2c:null 2d:["$","title",null,{"children":"Track 5: Rainfall Reliability Report | earth-observation | Observable"}] 32:T6c0,In this track, the aim is to build a decision-support tool designed to help users assess how the start and duration of the rainy season have changed across Kenyan counties, a critical factor for agricultural planning. By analysing historical and projected climate data, the tool offers insights into seasonal variability, highlighting counties facing shifts in rainfall onset or growing period length. These insights are especially valuable for smallholder farmers and stakeholders planning climate-smart agriculture, insurance schemes, and adaptation investments. Tailored for county planners, researchers, and proposal developers, the tool should allow users to filter by county, time period, and climate scenario, and to export maps, graphs, and key data summaries. Optional features include county comparisons and crop-specific overlays. With a user-friendly interface and decision-ready outputs, the notebook will support evidence-based planning to improve resilience in Kenya’s rain-fed agricultural systems. In this report, we explore: * When does the rainy season start, and what is the typical length and year-to-year variability of the length of growing period (LGP) across counties? * Which counties are experiencing significant shifts in onset or shortening of the rainy season? * What do these rainfall trends imply for key rain-fed crops (maize, sorghum, fodder) and for livestock populations (cattle, goats, sheep, poultry)? * How does rainfall variability correlate with crop yields, livestock vulnerability and rural population exposure to identify compounded-risk areas? * Which counties or sub-counties potentially require supplementary irrigation to mitigate shortened or unreliable rainy seasons? ---33:T114a,viewof kenya_geography = { // --- Wrapper setup --- const wrapper = html`

`; const container = html`

`; wrapper.append(container); // --- Create map --- const map = L.map(container).setView([0.1, 37.9], 6); L.tileLayer("https://{s}.basemaps.cartocdn.com/dark_all/{z}/{x}/{y}{r}.png", { attribution: "© OpenStreetMap contributors" }).addTo(map); // --- Define rainfall / geographic zones --- const regionClasses = { "High Rainfall Highlands": ["Kisii", "Kericho", "Nyeri", "Murang'A", "Embu", "Bomet", "Nyandarua"], "Semi-Arid Transition": ["Kitui", "Machakos", "Makueni", "Tharaka-Nithi"], "Arid & Semi-Arid Lands (ASAL)": ["Turkana", "Marsabit", "Wajir", "Mandera", "Garissa", "Isiolo", "Samburu", "Tana River"], "Lake Basin": ["Kisumu", "Homa Bay", "Migori", "Siaya", "Busia"], "Coastal Zone": ["Kwale", "Kilifi", "Lamu", "Tana River"] }; const regionData = kenya_boundaries.map(f => { const name = (f.properties.admin1_name || "").trim(); const region = Object.keys(regionClasses).find(k => regionClasses[k].includes(name)) || "Other"; return { admin1_name: name, value: region }; }); const kenyaGeo = buildKenyaGeojson(regionData); // --- Colors --- const colorMap = { "High Rainfall Highlands": "#2ecc71", "Semi-Arid Transition": "#f4d03f", "Arid & Semi-Arid Lands (ASAL)": "#e67e22", "Lake Basin": "#3498db", "Coastal Zone": "#1abc9c", "Other": "#bdc3c7" }; // --- Polygons --- L.geoJSON(kenyaGeo, { style: f => ({ fillColor: colorMap[f.properties.hazard_value], color: "#fff", weight: 1, fillOpacity: 0.8 }), onEachFeature: (f, layer) => layer.bindTooltip(`${f.properties.admin1_name}
${f.properties.hazard_value}`) }).addTo(map); // --- Region icons --- const regionIcons = { "High Rainfall Highlands": "🌳", "Semi-Arid Transition": "🌾", "Arid & Semi-Arid Lands (ASAL)": "☀️", "Lake Basin": "🌊", "Coastal Zone": "🏝️" }; for (const feature of kenyaGeo.features) { const region = feature.properties.hazard_value; const centroid = turf.centerOfMass(feature); const coords = [centroid.geometry.coordinates[1], centroid.geometry.coordinates[0]]; if (regionIcons[region]) { const icon = L.divIcon({ html: `

${regionIcons[region]}

`, iconSize: [24, 24], className: "" }); L.marker(coords, { icon }).addTo(map); } } // --- Floating animation CSS --- const style = document.createElement("style"); style.textContent = ` @keyframes float { 0% { transform: translateY(0px); } 50% { transform: translateY(-4px); } 100% { transform: translateY(0px); } } `; document.head.append(style); // --- Persistent HTML legend (not Leaflet-controlled) --- const legendHTML = html`

🌦️ Kenya Rainfall Zones
${Object.entries(colorMap) .map(([label, color]) => `

${regionIcons[label] || ""} ${label}

`) .join("")}

`; container.append(legendHTML); // --- Caption --- const caption = html`

Figure 1. Geographic classification of Kenya into major rainfall zones — represented with subtle icons for forested highlands (🌳), semi-arid transition belts (🌾), arid lands (☀️), lake basin (🌊), and coastal regions (🏝️). Each region illustrates Kenya’s diverse climatic gradients relevant to agricultural and adaptation planning.

`; wrapper.append(caption); // --- Ensure layout is ready --- requestAnimationFrame(() => map.invalidateSize()); return wrapper; } 34:T1a1e,viewof seasonalPlots = { const container = html`

`; const margin = { top: 30, right: 60, bottom: 40, left: 60 }; const width = 800, height = 260; const seasonOrder = ["DJF","JFM","FMA","MAM","AMJ","MJJ","JJA","JAS","ASO","SON","OND","NDJ"]; const seasons = seasonOrder.filter(s => kenyaPTOTFiltered.some(d => d.season === s) || kenyaNDWSFiltered.some(d => d.season === s) ); const seasonMonths = { DJF: "Dec–Jan–Feb", JFM: "Jan–Feb–Mar", FMA: "Feb–Mar–Apr", MAM: "Mar–Apr–May", AMJ: "Apr–May–Jun", MJJ: "May–Jun–Jul", JJA: "Jun–Jul–Aug", JAS: "Jul–Aug–Sep", ASO: "Aug–Sep–Oct", SON: "Sep–Oct–Nov", OND: "Oct–Nov–Dec", NDJ: "Nov–Dec–Jan" }; function meanSdBySeason(data, key) { const grouped = d3.rollups( data.filter(d => d.season !== "annual"), v => { const vals = v.map(d => d[key]); const m = d3.mean(vals); const s = Math.sqrt(d3.mean(vals.map(x => (x - m) ** 2))); return { mean: m, sd: s }; }, d => d.season ); const map = new Map(grouped); return seasons.map(s => ({ season: s, mean: map.get(s)?.mean ?? 0, sd: map.get(s)?.sd ?? 0 })); } const rainData = meanSdBySeason(kenyaPTOTFiltered, "seasonal_rain"); const ndwsData = meanSdBySeason(kenyaNDWSFiltered, "seasonal_ndws"); const allYears = [...new Set([ ...kenyaPTOTFiltered.map(d => d.year), ...kenyaNDWSFiltered.map(d => d.year) ].filter(Boolean))].sort((a,b)=>a-b); const minYear = allYears[0] ?? "start", maxYear = allYears.at(-1) ?? "end"; const x = d3.scaleBand().domain(seasons).range([margin.left, width - margin.right]).padding(0.2); const yRain = d3.scaleLinear().domain([0, d3.max(rainData, d => d.mean + d.sd)]).nice() .range([height - margin.bottom, margin.top]); const yNDWS = d3.scaleLinear().domain([0, d3.max(ndwsData, d => d.mean + d.sd)]).nice() .range([height - margin.bottom, margin.top]); // Reusable plot builder function drawPlot(data, yScale, color, ylabel) { const svg = d3.create("svg") .attr("width", width) .attr("height", height) .style("background", "linear-gradient(to top, #f5f5f5, #ffffff)"); const tooltip = d3.select("body") .append("div") .style("position", "absolute") .style("background", "rgba(0,0,0,0.75)") .style("color", "white") .style("padding", "4px 8px") .style("border-radius", "4px") .style("font-size", "12px") .style("pointer-events", "none") .style("opacity", 0) .style("transition", "opacity 0.2s ease") .style("box-shadow", "0 2px 6px rgba(0,0,0,0.3)"); svg.append("g") .attr("transform", `translate(0,${height - margin.bottom})`) .call(d3.axisBottom(x)) .selectAll("text").style("font-size", "11px"); svg.append("g") .attr("transform", `translate(${margin.left},0)`) .call(d3.axisLeft(yScale).ticks(5)) .selectAll("text").style("font-size", "11px"); svg.append("text") .attr("x", -height/2) .attr("y", 15) .attr("transform", "rotate(-90)") .attr("fill", color) .attr("text-anchor", "middle") .style("font-size","12px") .text(ylabel); // Bars svg.selectAll(".bar") .data(data) .join("rect") .attr("x", d => x(d.season)) .attr("y", d => yScale(d.mean)) .attr("width", x.bandwidth()) .attr("height", d => yScale(0) - yScale(d.mean)) .attr("fill", color) .attr("rx", 3) .style("opacity", 0.8) .on("mouseover", (event, d) => { tooltip.style("opacity", 1) .html(`${d.season}
Mean: ${d.mean.toFixed(2)}
SD: ${d.sd.toFixed(2)}`) .style("left", `${event.pageX + 10}px`) .style("top", `${event.pageY - 30}px`); d3.select(event.currentTarget) .transition().duration(100) .attr("fill", d3.color(color).darker(0.6)); }) .on("mousemove", (event) => { tooltip.style("left", `${event.pageX + 10}px`) .style("top", `${event.pageY - 30}px`); }) .on("mouseout", (event) => { tooltip.style("opacity", 0); d3.select(event.currentTarget) .transition().duration(300) .attr("fill", color); }); // Error bars const cap = x.bandwidth() * 0.3; svg.selectAll(".error") .data(data) .join("line") .attr("x1", d => x(d.season) + x.bandwidth()/2) .attr("x2", d => x(d.season) + x.bandwidth()/2) .attr("y1", d => yScale(d.mean + d.sd)) .attr("y2", d => yScale(Math.max(d.mean - d.sd, 0))) .attr("stroke", "#333"); svg.selectAll(".cap") .data(data) .join("line") .attr("x1", d => x(d.season)+x.bandwidth()/2 - cap/2) .attr("x2", d => x(d.season)+x.bandwidth()/2 + cap/2) .attr("y1", d => yScale(d.mean + d.sd)) .attr("y2", d => yScale(d.mean + d.sd)) .attr("stroke", "#333"); return svg.node(); } const title = html`

Seasonal Rainfall and NDWS (${minYear}–${maxYear})

`; container.append(title); const legend = html`

Rainfall (mm)

NDWS

${seasonOrder.map(code => seasons.includes(code) ? html`

${code}

${seasonMonths[code]}

` : null)}

`; container.append(legend); container.append(drawPlot(rainData, yRain, "steelblue", "Rainfall (mm)")); container.append(drawPlot(ndwsData, yNDWS, "orange", "NDWS")); const caption = html`

Figure 2. Mean seasonal rainfall (blue) and Number of Days of Water Stress (orange) across Kenya over ${minYear}–${maxYear}. Hover over a bar to view detailed values. Error bars show ±1 SD, highlighting variability between seasons.

`; container.append(caption); return container; } 35:T29ea,viewof phenologyMapS = { // --- Container setup --- const wrapper = html`

Season:

`; const container = wrapper.querySelector("div:nth-of-type(2)"); const map = L.map(container, { zoomControl: false }).setView([0.1, 37.9], 6); const legendDiv = container.querySelector("#phenologyLegend"); const seasonSelect = wrapper.querySelector("#seasonSelect"); // --- Dark basemap --- L.tileLayer("https://{s}.basemaps.cartocdn.com/dark_all/{z}/{x}/{y}{r}.png", { subdomains: "abcd", maxZoom: 19 }).addTo(map); const layerGroup = L.layerGroup().addTo(map); // --- Kenya borders --- const borderStyle = { color: "#ffffff", weight: 2.5, opacity: 0.9, fillOpacity: 0 }; L.geoJSON({ type: "FeatureCollection", features: kenya_boundaries }, { style: borderStyle }).addTo(map); // --- Year display --- const yearDisplay = L.control({ position: "topright" }); yearDisplay.onAdd = () => { const div = L.DomUtil.create("div", "year-display"); div.style.background = "rgba(0,0,0,0.5)"; div.style.color = "white"; div.style.padding = "5px 10px"; div.style.borderRadius = "6px"; div.style.fontSize = "18px"; div.style.fontWeight = "bold"; div.textContent = ""; return div; }; yearDisplay.addTo(map); // --- Custom styles --- const style = document.createElement("style"); style.textContent = ` .phenology-tooltip { background: rgba(0, 0, 0, 0.85); color: #fff; border-radius: 6px; padding: 6px 10px; font-size: 13px; line-height: 1.4; box-shadow: 0 0 5px rgba(255,255,255,0.2); pointer-events: none; } .year-slider-btn { background: #333; color: white; border: none; border-radius: 4px; width: 28px; height: 28px; font-size: 16px; cursor: pointer; } .year-slider-btn:hover { background: #555; } `; document.head.appendChild(style); (async () => { const raw = await phenology_db.query(` SELECT admin1_name, season_yr, season_num, origin, "DER.sos" AS SOS, "DER.eos" AS EOS FROM kenya_phenology WHERE season_num IN (1, 2) `); const data = raw.map(d => { const origin = new Date(d.origin); const sos = new Date(d.SOS); const eos = new Date(d.EOS); const SOS_day = (sos - origin) / (1000 * 60 * 60 * 24); let EOS_day = (eos - origin) / (1000 * 60 * 60 * 24); if (EOS_day < SOS_day) EOS_day += 365; return { admin1_name: d.admin1_name, season_yr: +d.season_yr, season_num: +d.season_num, SOS_day, EOS_day, LGP: EOS_day - SOS_day }; }); const grouped = d3.rollups( data, v => ({ SOS_day: d3.mean(v, d => d.SOS_day), EOS_day: d3.mean(v, d => d.EOS_day), LGP: d3.mean(v, d => d.LGP), LGP_sd: d3.deviation(v, d => d.LGP) }), d => `${d.admin1_name}_${d.season_num}`, d => d.season_yr ); const aggregated = []; for (const [key, byYear] of grouped) { const [county, season] = key.split("_"); for (const [year, stats] of byYear) { aggregated.push({ county, season_num: +season, season_yr: year, ...stats }); } } const lgpExtent = d3.extent(aggregated, d => d.LGP); const radiusScale = d3.scaleLinear().domain(lgpExtent).range([5, 25]); const haloScale = d3.scaleLinear() .domain([0, d3.max(aggregated, d => d.LGP_sd || 0)]) .range([0, 20]); const years = Array.from(new Set(aggregated.map(d => d.season_yr))).sort((a,b)=>a-b); const slider = wrapper.querySelector("#yearSlider"); const label = wrapper.querySelector("#yearLabel"); const btnUp = wrapper.querySelector("#yearUp"); const btnDown = wrapper.querySelector("#yearDown"); const defaultYear = 2015; let currentYearIndex = Math.max(0, years.indexOf(defaultYear)); let currentSeason = +seasonSelect.value; slider.max = years.length - 1; slider.value = currentYearIndex; label.textContent = years[currentYearIndex]; function update(targetYear, targetSeason) { layerGroup.clearLayers(); const filtered = aggregated.filter(d => d.season_yr === targetYear && d.season_num === targetSeason); yearDisplay.getContainer().textContent = `Year: ${targetYear}`; const sosExtent = d3.extent(filtered, d => d.SOS_day); const colorScale = d3.scaleSequential(d3.interpolateCool).domain(sosExtent.reverse()); // --- Persistent legend update --- legendDiv.innerHTML = ` Legend (${targetSeason === 1 ? "MAM" : "OND"})

Late SOSEarly SOS

◯ White halo = higher variability within the county `; filtered.forEach(d => { const feature = kenya_boundaries.find(f => f.properties.admin1_name === d.county); if (!feature) return; const centroid = turf.centroid(feature).geometry.coordinates; const marker = L.circleMarker([centroid[1], centroid[0]], { radius: radiusScale(d.LGP), fillColor: colorScale(d.SOS_day), color: "none", fillOpacity: 0.8 }).addTo(layerGroup); marker.bindTooltip( ``, { permanent: false, direction: "top", opacity: 0.9, className: "phenology-tooltip" } ); L.circleMarker([centroid[1], centroid[0]], { radius: radiusScale(d.LGP) + haloScale(d.LGP_sd), color: "white", weight: 2, fill: false, opacity: 0.4 }).addTo(layerGroup); }); // --- Update caption dynamically --- caption.innerHTML = ` Figure 3. Annual variation in the length of growing period (LGP) and start-of-season (SOS) day for Kenya’s ${targetSeason === 1 ? "long-rains (March–May)" : "short-rains (October–December)"} season over 2000–2025. Circle color indicates timing of SOS—green for early onset and blue for late start—while circle size represents the LGP. The white halo denotes the variability of LGP within the county. `; } // --- Slider and button interactions --- slider.addEventListener("input", e => { currentYearIndex = +e.target.value; label.textContent = years[currentYearIndex]; update(years[currentYearIndex], currentSeason); }); btnUp.onclick = () => { if (currentYearIndex < years.length - 1) { currentYearIndex++; slider.value = currentYearIndex; label.textContent = years[currentYearIndex]; update(years[currentYearIndex], currentSeason); } }; btnDown.onclick = () => { if (currentYearIndex > 0) { currentYearIndex--; slider.value = currentYearIndex; label.textContent = years[currentYearIndex]; update(years[currentYearIndex], currentSeason); } }; seasonSelect.onchange = e => { currentSeason = +e.target.value; update(years[currentYearIndex], currentSeason); }; // --- Animation controls --- const playPauseBtn = wrapper.querySelector("#playPauseBtn"); let isPlaying = false; let playInterval = null; function playAnimation() { if (isPlaying) return; isPlaying = true; playPauseBtn.textContent = "⏸"; playInterval = setInterval(() => { currentYearIndex = (currentYearIndex + 1) % years.length; slider.value = currentYearIndex; label.textContent = years[currentYearIndex]; update(years[currentYearIndex], currentSeason); }, 1200); } function pauseAnimation() { isPlaying = false; playPauseBtn.textContent = "▶"; clearInterval(playInterval); } playPauseBtn.onclick = () => isPlaying ? pauseAnimation() : playAnimation(); [slider, btnUp, btnDown, seasonSelect].forEach(el => el.addEventListener("input", pauseAnimation)); // --- Initialize view --- map.whenReady(() => update(years[currentYearIndex], currentSeason)); })(); invalidation.then(() => map.remove()); const caption = html`

`; wrapper.append(caption); return wrapper; } 36:T1ef5,viewof sosLgpTimeseriesHover = { const container = html`

`; // ---------- Header ---------- container.append(html`

County-Level Mean SOS and LGP with Intra-County Spatial Variability

`); // ---------- Controls ---------- const controls = html`

Season: Metric:

`; container.append(controls); // ---------- Floating Info Box ---------- const infoBox = html``; container.append(infoBox); // ---------- Plot Function ---------- function drawPlot(season_num, metric) { const filtered = statsByCountyYearFiltered.filter(d => d.season_num === +season_num); const meanKey = metric === "LGP" ? "meanLGP" : "meanSOS"; const sdKey = metric === "LGP" ? "sdLGP" : "sdSOS"; // --- Compute mean per county for sorting and color scale --- const countySeasonMeans = Array.from( d3.rollup(filtered, v => d3.mean(v, d => d[meanKey]), d => d.county), ([county, meanVal]) => ({ county, meanVal }) ).sort((a, b) => a.meanVal - b.meanVal); const sortedSeasonCounties = countySeasonMeans.map(d => d.county); // Color scale by sorted order const colorScale = d3.scaleSequential() .domain([0, countySeasonMeans.length - 1]) .interpolator(d3.interpolateViridis); const countyColor = new Map(countySeasonMeans.map((d, i) => [d.county, colorScale(i)])); let selectedCounty = null; const plotRow = html`

`; const plotContainer = html`

`; plotRow.append(plotContainer); // ---------- Main Plot ---------- const plot = Plot.plot({ width: 1100, height: 800, style: { background: "#fff", color: "#000" }, marginLeft: 60, x: { label: "Year", domain: [2000, 2025] }, y: { label: metric === "LGP" ? "Mean LGP (days)" : "Start of Season (day of year)", domain: metric === "LGP" ? [20, 400] : [-100, 400] }, color: { domain: sortedSeasonCounties, range: sortedSeasonCounties.map(c => countyColor.get(c)) }, marks: [ Plot.areaY(filtered, { x: "year", y1: d => d[meanKey] - (d[sdKey] || 0), y2: d => d[meanKey] + (d[sdKey] || 0), fill: "county", fillOpacity: 0.2, curve: "linear" }), Plot.line(filtered, { x: "year", y: meanKey, stroke: "county", strokeWidth: 2.2, curve: "linear" }), Plot.dot(filtered, { x: "year", y: meanKey, fill: "county", r: 2.8 }) ] }); plotContainer.append(plot); // ---------- Legend ---------- const legendBox = html`

`; countySeasonMeans.forEach(d => { const item = html`

${d.county}

`; item.addEventListener("mouseenter", () => (item.style.background = "rgba(0,0,0,0.08)")); item.addEventListener("mouseleave", () => (item.style.background = "transparent")); item.addEventListener("click", () => { selectedCounty = selectedCounty === d.county ? null : d.county; updateSelection(); }); legendBox.append(item); }); plotRow.append(legendBox); // ---------- Update Function ---------- function updateSelection() { const svg = plotContainer.querySelector("svg"); if (!svg) return; const lines = svg.querySelectorAll("path[stroke]"); const dots = svg.querySelectorAll("circle"); const areas = svg.querySelectorAll("path[fill]"); if (!selectedCounty) { infoBox.style.opacity = 0; setTimeout(() => (infoBox.style.display = "none"), 300); lines.forEach(l => (l.style.display = "", l.style.opacity = 1)); dots.forEach(d => (d.style.display = "", d.style.opacity = 1)); areas.forEach(a => (a.style.display = "")); return; } const color = countyColor.get(selectedCounty); lines.forEach(l => (l.style.display = l.getAttribute("stroke") === color ? "" : "none")); dots.forEach(d => (d.style.display = d.getAttribute("fill") === color ? "" : "none")); areas.forEach(a => (a.style.display = a.getAttribute("fill") === color ? "" : "none")); // --- Info Box (updated wording) --- const series = filtered.filter(d => d.county === selectedCounty); const mean = d3.mean(series, d => d[meanKey]).toFixed(1); const sd = d3.mean(series, d => d[sdKey] || 0).toFixed(1); const range = d3.extent(series, d => d.year).join("–"); infoBox.innerHTML = ` ${selectedCounty}
Long-term mean ${metric}: ${mean}
Intra-county spatial SD (avg): ${sd}
Years represented: ${range} `; infoBox.style.display = "block"; requestAnimationFrame(() => (infoBox.style.opacity = 1)); } requestAnimationFrame(() => updateSelection()); return plotRow; } // ---------- Initial Plot ---------- let plotSection; requestAnimationFrame(() => { plotSection = drawPlot(1, "LGP"); container.append(plotSection); }); // ---------- Caption ---------- container.append(html`

Figure 4. Lines show each county’s annual mean SOS or annual mean LGP for 2000–2024. Shaded envelopes represent the intra-county spatial variability (standard deviation across pixels within the county in that year). Click a county in the legend to isolate its trajectory and display a summary of long-term mean and spatial SD.

`); // ---------- Update on Control Change ---------- const update = () => { const season = +controls.querySelector("#seasonSel").value; const metric = controls.querySelector("#metricSel").value; plotSection.remove(); plotSection = drawPlot(season, metric); container.insertBefore(plotSection, container.lastElementChild); }; controls.querySelector("#seasonSel").addEventListener("change", update); controls.querySelector("#metricSel").addEventListener("change", update); return container; } 37:T2b93,viewof kenyaPhenologyDashboard = (async () => { const container = html`

Select Season:

`; const caption = container.querySelector("#caption"); const seasonSelect = container.querySelector("#seasonSelect"); let activeMaps = []; // ---------- Leaflet Map Builder ---------- function createMap(div, data, valueKey, label, colorScale, tooltipLabel, legendLabel) { if (div._leaflet_id) { div._leaflet_id = null; div.innerHTML = ""; } const map = L.map(div, { zoomControl: false }).setView([-0.5, 39.0], 6.2); L.tileLayer("https://{s}.basemaps.cartocdn.com/dark_all/{z}/{x}/{y}{r}.png", { subdomains: "abcd", maxZoom: 19 }).addTo(map); // --- Choropleth layer --- L.geoJSON(data, { style: f => ({ fillColor: isFinite(f.properties[valueKey]) ? colorScale(f.properties[valueKey]) : "#333", weight: 1, color: "white", fillOpacity: 0.7 }), onEachFeature: (f, layer) => { const val = f.properties[valueKey]; const tooltip = L.tooltip({ direction: "top", offset: [0, -5], className: "phenology-tooltip" }) .setContent(`${f.properties.admin1_name}
${tooltipLabel}: ${isFinite(val) ? val.toFixed(1) : "NA"}`); layer.on("mouseover", e => { tooltip.setLatLng(e.latlng); tooltip.addTo(map); }); layer.on("mousemove", e => tooltip.setLatLng(e.latlng)); layer.on("mouseout", () => map.removeLayer(tooltip)); } }).addTo(map); // --- Legend --- const vals = data.map(f => f.properties[valueKey]).filter(v => isFinite(v)); if (vals.length > 1) { const [min, max] = d3.extent(vals); const steps = 6; const legend = L.control({ position: "bottomright" }); legend.onAdd = () => { const div = L.DomUtil.create("div", "info legend"); Object.assign(div.style, { background: "rgba(30,30,30,0.8)", padding: "6px", borderRadius: "6px", color: "white", fontSize: "12px" }); for (let i = 0; i <= steps; i++) { const val = min + i * (max - min) / steps; div.innerHTML += `${val.toFixed(1)}
`; } div.innerHTML += `${legendLabel}`; return div; }; map.whenReady(() => { legend.addTo(map); function ensureLegendReady(callback, retries = 10) { const legendDiv = legend.getContainer?.(); if (legendDiv && map.getContainer().contains(legendDiv)) { callback?.(); } else if (retries > 0) { setTimeout(() => ensureLegendReady(callback, retries - 1), 100); } } ensureLegendReady(() => map.invalidateSize()); }); } map.whenReady(() => map.invalidateSize()); activeMaps.push(map); return map; } // ---------- Hover Tooltip Maker ---------- function makeTooltip() { return html``; } // ---------- Hover Logic for Bar Plots ---------- function attachHover(plotContainer, tooltip) { const svg = plotContainer.querySelector("svg"); if (!svg) return; const bars = svg.querySelectorAll("rect"); bars.forEach(bar => { const title = bar.querySelector("title")?.textContent; if (!title) return; bar.addEventListener("mouseenter", e => { tooltip.style.display = "block"; tooltip.innerHTML = title.replace(/\n/g, "
"); bar.style.stroke = "white"; bar.style.strokeWidth = "1.5px"; }); bar.addEventListener("mousemove", e => { const bbox = plotContainer.getBoundingClientRect(); tooltip.style.left = `${e.clientX - bbox.left + 15}px`; tooltip.style.top = `${e.clientY - bbox.top + 10}px`; }); bar.addEventListener("mouseleave", () => { tooltip.style.display = "none"; bar.style.stroke = "none"; }); }); } // ---------- Main Drawing ---------- async function drawMaps(season_num) { activeMaps.forEach(m => m.remove()); activeMaps = []; const filtered = soslgpTrends.filter(d => d.season_num === season_num); function enrich(key, valueKey) { return JSON.parse(JSON.stringify(kenya_boundaries)).map(f => { const match = filtered.find(d => d.county === f.properties.admin1_name); f.properties[valueKey] = match ? match[key] : NaN; return f; }); } const meanSOS = enrich("meanSOS_overYears", "meanSOS"); const meanLGP = enrich("meanLGP_overYears", "meanLGP"); const sdSOS = enrich("sdSOS_overYears", "sdSOS"); const sdLGP = enrich("sdLGP_overYears", "sdLGP"); function makeScale(features, key, interp) { const vals = features.map(f => f.properties[key]).filter(v => isFinite(v)); return vals.length > 1 ? d3.scaleSequential(d3.extent(vals), interp) : d3.scaleSequential([0, 1], interp); } const sosScale = makeScale(meanSOS, "meanSOS", d3.interpolateBlues); const lgpScale = makeScale(meanLGP, "meanLGP", d3.interpolateYlGn); const sdSOSScale = makeScale(sdSOS, "sdSOS", d3.interpolatePiYG); const sdLGPScale = makeScale(sdLGP, "sdLGP", d3.interpolatePiYG); ["meanSOSMap", "meanLGPMap", "sdSOSMap", "sdLGPMap"].forEach(id => { container.querySelector(`#${id}`).innerHTML = ""; }); createMap(container.querySelector("#meanSOSMap"), meanSOS, "meanSOS", "Mean SOS (DOY)", sosScale, "Mean SOS (DOY)", "Start of Season"); createMap(container.querySelector("#meanLGPMap"), meanLGP, "meanLGP", "Mean LGP (days)", lgpScale, "Mean LGP (days)", "Length of Growing Period"); createMap(container.querySelector("#sdSOSMap"), sdSOS, "sdSOS", "SD SOS (days)", sdSOSScale, "SD of SOS (days)", "SD Start of Season"); createMap(container.querySelector("#sdLGPMap"), sdLGP, "sdLGP", "SD LGP (days)", sdLGPScale, "SD of LGP (days)", "SD Length of Growing Period"); // --- Bar Plots with Custom Tooltip --- const lgpBarsContainer = container.querySelector("#lgpBarsContainer"); const sosBarsContainer = container.querySelector("#sosBarsContainer"); lgpBarsContainer.innerHTML = ""; sosBarsContainer.innerHTML = ""; const lgpTooltip = makeTooltip(); const sosTooltip = makeTooltip(); lgpBarsContainer.appendChild(lgpTooltip); sosBarsContainer.appendChild(sosTooltip); // --- LGP Bars --- const lgpPlot = Plot.plot({ width: 500, height: 700, marginLeft: 120, marginBottom: 50, style: { background: "black", color: "white" }, title: `Average Length of Growing Period (${season_num === 1 ? "MAM" : "OND"})`, x: { label: "LGP (days)", labelOffset: 35 }, y: { domain: filtered.sort((a, b) => d3.ascending(a.meanLGP_overYears, b.meanLGP_overYears)).map(d => d.county), label: null }, marks: [ Plot.barX(filtered, { x: "meanLGP_overYears", y: "county", fill: "steelblue", title: d => `${d.county}\nMean: ${d.meanLGP_overYears.toFixed(1)} days\n±${d.sdLGP_overYears.toFixed(1)}\nRange: ${d.minLGP.toFixed(1)} – ${d.maxLGP.toFixed(1)}` }), Plot.barX(filtered, { x: d => d.meanLGP_overYears + (d.sdLGP_overYears || 0), x2: "meanLGP_overYears", y: "county", fill: "purple", opacity: 0.7, title: d => `${d.county}\nSD: ${isFinite(d.sdLGP_overYears) ? d.sdLGP_overYears.toFixed(1) : 'NA'} days` }), Plot.ruleX(filtered, { x1: "minLGP", x2: "maxLGP", y: "county", stroke: "white" }) ] }); lgpBarsContainer.appendChild(lgpPlot); await new Promise(r => setTimeout(r, 50)); attachHover(lgpBarsContainer, lgpTooltip); lgpBarsContainer.querySelectorAll("title").forEach(t => t.remove()); // --- SOS Bars --- const sosPlot = Plot.plot({ width: 500, height: 700, marginLeft: 120, marginBottom: 50, style: { background: "black", color: "white" }, title: `Average Start of Season (${season_num === 1 ? "MAM" : "OND"})`, x: { label: "Day of Year (SOS)", labelOffset: 35 }, y: { domain: filtered.sort((a, b) => d3.ascending(a.meanSOS_overYears, b.meanSOS_overYears)).map(d => d.county), label: null }, marks: [ Plot.barX(filtered, { x: "meanSOS_overYears", y: "county", fill: "darkgreen", title: d => `${d.county}\nMean SOS: DOY ${d.meanSOS_overYears.toFixed(0)}\n±${d.sdSOS_overYears.toFixed(0)} days\nRange: ${d.minSOS.toFixed(0)} – ${d.maxSOS.toFixed(0)}` }), Plot.barX(filtered, { x: d => d.meanSOS_overYears + (d.sdSOS_overYears || 0), x2: "meanSOS_overYears", y: "county", fill: "gold", opacity: 0.7, title: d => `${d.county}\nSD: ${isFinite(d.sdSOS_overYears) ? d.sdSOS_overYears.toFixed(0) : 'NA'} days` }), Plot.ruleX(filtered, { x1: "minSOS", x2: "maxSOS", y: "county", stroke: "white" }) ] }); sosBarsContainer.appendChild(sosPlot); await new Promise(r => setTimeout(r, 50)); attachHover(sosBarsContainer, sosTooltip); sosBarsContainer.querySelectorAll("title").forEach(t => t.remove()); caption.innerHTML = `Figure 5. Mean and variability of phenological indicators across Kenyan counties for ${season_num === 1 ? "MAM (March–May)" : "OND (October–December)"} season evalauted over the period 2000-2024. Maps show spatial patterns; bar plots summarize county-level averages.`; } drawMaps(1); seasonSelect.addEventListener("change", e => drawMaps(+e.target.value)); return container; })(); 38:T5ef,Preliminary analysis indicates that Kenya’s western highlands and central regions exhibit the longest and most reliable *long-rains* (MAM) growing periods, typically lasting **200–240 days**, with *short-rains* (OND) periods of **100–150 days**. These extended and stable LGPs reflect the bimodal rainfall regime and sustained soil moisture retention in these high-altitude zones. Conversely, **arid and semi-arid counties**—including *Wajir, Mandera, Isiolo, Tana River, Garissa, Kitui, and Tharaka-Nithi*—experience **much shorter growing periods** (often <100 days), indicative of their reliance on short, erratic rainfall events. The **start of the growing season (SOS)** varies substantially across Kenya, shaped by strong climatic gradients and the country’s bimodal rainfall pattern. Counties in the southern and western highlands (e.g., *Narok, Kajiado*) typically show **early greening onset** around *Day of Year (DOY) 40* (early February), coinciding with the arrival of the long rains. In contrast, semi-arid counties in eastern and northern Kenya (e.g., *Tana River, Isiolo, Waji*r) exhibit **delayed SOS values** (*DOY 120–150*; May–June), consistent with later ITCZ passage and higher interannual rainfall variability. These spatial gradients align with bibliographic regional climatological findings, which report **earlier, longer growing periods in western Kenya** and **progressively shorter, more variable seasons toward the northeast** (Vrieling, 2013; Ndungu et al., 2019). ---39:T1d7a,viewof TimeseriesHovers = { const container = html`

`; // ---------- Figure title ---------- const header = html`

Figure 6. Long-Term Trends in County-Level LGP and SOS (2000–2024)

`; container.appendChild(header); // ---------- County selection ---------- const significantCountiesSeason1SOS = countySeasonTrends .filter(d => d.season_num === 1 && d.sosTrend > 1) .map(d => d.county); const significantCountiesSeason2SOS = countySeasonTrends .filter(d => d.season_num === 2 && d.sosTrend > 1) .map(d => d.county); const significantCountiesSeason1LGP = countySeasonTrends .filter(d => d.season_num === 1 && d.lgpTrend < -1) .map(d => d.county); const significantCountiesSeason2LGP = countySeasonTrends .filter(d => d.season_num === 2 && d.lgpTrend < -1) .map(d => d.county); // ---------- Reusable plot builder ---------- function makeTrendPlot(topCounties, yKey, yLabel, titleText, subtitleText) { const filteredData = statsByCountyYearFiltered .sort((a, b) => a.year - b.year) .filter(d => topCounties.includes(d.county)); //------------------------------------------------------ // Local color scale only based on the counties displayed //------------------------------------------------------ const localColorScale = d3.scaleSequential() .domain([0, topCounties.length - 1]) .interpolator(d3.interpolateMagma); // or Viridis // Turbo gives better separation when n is small (3–12) // Map county -> color const countyColor = new Map( topCounties.map((c, i) => [c, localColorScale(i)]) ); const plotContainer = html`

`; const tooltip = html``; plotContainer.appendChild(tooltip); const countyStats = new Map(); topCounties.forEach(c => { const data = filteredData.filter(d => d.county === c); const years = data.map(d => d.year); const meanVal = d3.mean(data, d => d[yKey]); const sdVal = d3.deviation(data, d => d[yKey]); countyStats.set(c, { minYear: d3.min(years), maxYear: d3.max(years), meanVal, sdVal }); }); const plot = Plot.plot({ width: 500, height: 360, style: { background: "#fff", color: "#000" }, marginLeft: 50, marginRight: 0, x: { domain: [2000, 2025], label: "Year" }, y: { label: yLabel }, color: { domain: topCounties, range: topCounties.map(c => countyColor.get(c)) }, marks: [ Plot.line(filteredData, { x: "year", y: yKey, stroke: "county", strokeWidth: 2.5, curve: "linear", title: d => { const s = countyStats.get(d.county); return `${d.county} Period: ${s.minYear}–${s.maxYear} Average ${yLabel}: ${s.meanVal.toFixed(1)} ± ${s.sdVal?.toFixed(1) ?? 0} (annual mean ± intra-county SD)`; } }), Plot.dot(filteredData, { x: "year", y: yKey, fill: "county", r: 3 }) ] }); plotContainer.appendChild(plot); // ---------- Hover interactivity ---------- const svg = plotContainer.querySelector("svg"); const lines = svg.querySelectorAll("path[stroke]"); lines.forEach(line => { line.style.transition = "stroke-width 0.25s, opacity 0.25s"; line.addEventListener("mouseenter", e => { const county = line.getAttribute("stroke"); const stats = countyStats.get(county); lines.forEach(l => (l.style.opacity = 0.15)); line.style.opacity = 1; line.style.strokeWidth = 4; tooltip.innerHTML = ` ${county}
Period: ${stats.minYear}–${stats.maxYear}
Avg ${yLabel}: ${stats.meanVal.toFixed(1)} ± ${stats.sdVal.toFixed(1)}
(mean annual value ± intra-county SD) `; tooltip.style.display = "block"; }); line.addEventListener("mousemove", e => { const bbox = plotContainer.getBoundingClientRect(); tooltip.style.left = e.clientX - bbox.left + 12 + "px"; tooltip.style.top = e.clientY - bbox.top + 12 + "px"; }); line.addEventListener("mouseleave", e => { lines.forEach(l => { l.style.opacity = 1; l.style.strokeWidth = 2.5; }); tooltip.style.display = "none"; }); }); // ---------- Legend ---------- const legendBox = html`

`; topCounties.forEach(c => { const item = html`

${c}

`; legendBox.appendChild(item); }); const wrapper = html`

${titleText}

${subtitleText}

${plotContainer}${legendBox}

`; return wrapper; } // ---------- Create all 4 panels ---------- const grid = html`

`; grid.appendChild(makeTrendPlot( significantCountiesSeason1LGP, "meanLGP", "Mean LGP (days)", "Season 1 (MAM) – LGP", "Counties where annual LGP is shortening by ≥ 1 day per year" )); grid.appendChild(makeTrendPlot( significantCountiesSeason1SOS, "meanSOS", "Mean SOS (DOY)", "Season 1 (MAM) – SOS", "Counties where SOS is delaying by ≥ 1 day per year" )); grid.appendChild(makeTrendPlot( significantCountiesSeason2LGP, "meanLGP", "Mean LGP (days)", "Season 2 (OND) – LGP", "Counties where annual LGP is shortening by ≥ 1 day per year" )); grid.appendChild(makeTrendPlot( significantCountiesSeason2SOS, "meanSOS", "Mean SOS (DOY)", "Season 2 (OND) – SOS", "Counties where SOS is delaying by ≥ 1 day per year" )); container.appendChild(grid); // ---------- Caption ---------- const caption = html`

Each panel displays annual mean values of Length of Growing Period (LGP) or Start of Season (SOS) for counties exhibiting statistically meaningful long-term change. Annual means are computed from all available pixel-level observations for that county, season, and year. The shaded range shown in the tooltip (mean ± SD) reflects intra-county spatial variability in that year’s SOS or LGP distribution. Counties are selected based on trend slopes derived from the 2000–2024 record.

`; container.appendChild(caption); return container; } 3a:T630, ### ** Counties Exhibiting Significant Shifts in Rainfall Onset or Growing Period Duration (2000–2024)** Counties along the **coastal and low-elevation belt**, notably *Kwale, Kilifi,* and *Lamu*, show a pronounced **shortening of the growing period** during MAM, with LGP decreasing by approximately **1 day per year or more**. Similar negative trends are observed in *Narok, Nyeri,* and *Murang’a* during the short rains (OND). These results echo *Vrieling (2013)*, who documented comparable declines (~–0.9 days/yr) in eastern Kenya. In contrast, several counties in the **northwestern highlands**, such as *Turkana* and *Elgeyo-Marakwet*, display **lengthening growing seasons** of up to **+2 days/yr**, suggesting localized increases in early-season rainfall and improved soil moisture retention. High-potential agricultural regions including *Homa Bay, Murang’a,* and *Migori* exhibit **delayed onsets** of the growing season, with SOS shifting later by roughly **1 days/yr** during MAM. Such delays, coupled with shortening LGPs elsewhere, imply a progressive compression of Kenya’s primary crop-growing window—consistent with recent climatological reports (*Kotikot et al., 2024*). These trends highlight **diverging trajectories of climate risk** across Kenya: while some western counties may benefit from marginally extended growing periods, much of the eastern and coastal belt faces increasing climatic stress from delayed and truncated rains. *Note: Trends were computed via least-squares linear regression of seasonal means for each county.* --- 3b:T164a,// --- 3️⃣ MAPS WITH SIGNIFICANCE FILTER --- viewof trendMaps = { const container = html`

Figure 7 — Spatial Trends in LGP and SOS (2000–2025). Spatial distribution of trend magnitude (days per year) in the Length of Growing Period (LGP) and Start of Season (SOS) across Kenya, shown separately for the MAM (Season 1) and OND (Season 2) rainfall seasons. Shading intensity represents the trend magnitude, while opacity encodes statistical significance (p < 0.05 → solid; p ≥ 0.2 → transparent). Blue areas (LGP) indicate declining growing-period lengths (-ve trend), red indicates increases (+ve trend). For SOS, greener shades imply earlier onset (-ve trend), while lighter tones denote delays (+ve trend). Western highlands exhibit increasing LGP and delayed SOS (MAM season), whereas arid northern and eastern counties show shortening and earlier growing periods, signalling greater climatic stress.

`; // --- Map creation helper --- function makeBaseMap(divId) { const div = container.querySelector(`#${divId}`); const map = L.map(div).setView([0.1, 37.9], 6); L.tileLayer("https://{s}.basemaps.cartocdn.com/dark_all/{z}/{x}/{y}{r}.png", { subdomains: 'abcd', maxZoom: 19 }).addTo(map); return { map, div }; } const maps = { lgp1: makeBaseMap("lgpMap1"), sos1: makeBaseMap("sosMap1"), lgp2: makeBaseMap("lgpMap2"), sos2: makeBaseMap("sosMap2") }; // --- Split trends by season --- const trendsBySeason = d3.group(countySeasonTrends, d => d.season_num); // --- Color scales --- const lgpColorScale = vals => d3.scaleDiverging(d3.interpolateRdBu) .domain([d3.max(vals), 0, d3.min(vals)]); const sosColorScale = vals => d3.scaleSequential() .domain([d3.max(vals), d3.min(vals)]) // reversed: greener = earlier SOS .interpolator(d3.interpolateGreens); // --- Create layer + persistent legend/title --- function createLayer(mapObj, mapType, season, values, colorScale, valueProp, pProp) { const { map, div } = mapObj; const geoColored = kenya_boundaries.map(f => { const match = trendsBySeason.get(season)?.find(d => d.county === f.properties.admin1_name); f.properties[valueProp] = match ? match[valueProp] : 0; f.properties[pProp] = match ? match[pProp] : 1; return f; }); L.geoJSON(geoColored, { style: f => ({ fillColor: colorScale(f.properties[valueProp]), weight: 1, color: "white", fillOpacity: (() => { const p = +f.properties[pProp]; if (p < 0.05) return 1; if (p < 0.1) return 0.8; if (p < 0.2) return 0.6; return 0.35; })() }), onEachFeature: (f, layer) => { layer.bindTooltip(` ${f.properties.admin1_name}
${mapType} Trend (Season ${season}): ${f.properties[valueProp].toFixed(2)} days/yr
p = ${f.properties[pProp].toFixed(3)} `, { sticky: true, direction: "top", opacity: 0.9 }); layer.on({ mouseover: e => { e.target.setStyle({ weight: 3, color: "yellow" }); e.target.bringToFront(); }, mouseout: e => e.target.setStyle({ weight: 1, color: "white" }) }); } }).addTo(map); // --- ✅ Persistent HTML legend (instead of L.control) --- const range = d3.extent(values); const step = (range[1] - range[0]) / 8; const legendDiv = html`

`; let htmlLegend = `${mapType} Trend (days/yr)
Season ${season}
`; for (let i = 0; i <= 8; i++) { const val = range[0] + i * step; htmlLegend += `${val.toFixed(1)}
`; } legendDiv.innerHTML = htmlLegend; div.appendChild(legendDiv); // --- ✅ Persistent title overlay --- const titleDiv = html`

${mapType} Trend (Season ${season})

`; div.appendChild(titleDiv); map.whenReady(() => map.invalidateSize()); } // --- Create all maps --- for (const [season, arr] of trendsBySeason) { const lgpVals = arr.map(d => d.lgpTrend); const sosVals = arr.map(d => d.sosTrend); createLayer(maps[`lgp${season}`], "LGP", season, lgpVals, lgpColorScale(lgpVals), "lgpTrend", "lgpP"); createLayer(maps[`sos${season}`], "SOS", season, sosVals, sosColorScale(sosVals), "sosTrend", "sosP"); } return container; } 3c:Tfa4,viewof kenyaCropsOverview = { const container = html`

`; // --- Paragraph --- const para = html`

Kenya’s agricultural diversity reflects its climatic range. Cash crops such as tea, coffee, and horticultural produce thrive in the cool, wet highlands, while staples like maize, beans, and potatoes dominate nationwide. Drought-tolerant crops such as sorghum and millet sustain the semi-arid regions. The two rainy seasons—the March–May (MAM) long rains and October–December (OND) short rains—govern the agricultural calendar. Fodder crops like Napier grass and pasture species rely on consistent rainfall to maintain livestock feed. Erratic rainfall shortens forage availability, reducing milk yields, body weight, and herd resilience, while prolonged dry spells exacerbate feed shortages. Poultry, though less water-dependent, are indirectly affected by reduced grain supply. Favorable rainfall, in contrast, boosts pasture biomass, improves animal health, and stabilizes household income and food security.

`; container.append(para); // --- Table data --- const data = [ { icon: "🌾", category: "Rain-fed Food Crops", examples: "Maize, Sorghum, Millet, Beans", rainfall: "High", regions: "Highlands and Semi-arid lowlands", impact: "Reduced yields, delayed planting, crop failure during droughts" }, { icon: "🌿", category: "Fodder & Pasture", examples: "Napier grass, Rhodes grass, Pasture species", rainfall: "Very High", regions: "Rift Valley, Central, Eastern", impact: "Feed shortages, overgrazing, livestock migration" }, { icon: "🐄", category: "Livestock", examples: "Cattle, Goats, Sheep, Poultry", rainfall: "Indirect (through pasture and water availability)", regions: "Nationwide, especially arid & semi-arid lands", impact: "Reduced milk/meat yield, higher mortality, loss of income" }, { icon: "🌴", category: "Resilient Crops", examples: "Sorghum, Millet, Cassava", rainfall: "Moderate (drought-tolerant)", regions: "Eastern, Coastal, Semi-arid regions", impact: "Better survival under dry conditions; alternative food source" } ]; // --- Fancy interactive cards --- const fancyGrid = html`

${data.map(row => html`

${row["icon"]} ${row["category"]}

Major Crops: ${row["examples"]}

Main Regions: ${row["regions"]}

Rainfall Dependence: ${row["rainfall"]}

Impact: ${row["impact"]}

`)}

`; // --- Add interactivity --- fancyGrid.querySelectorAll(".cropCard").forEach(card => { card.addEventListener("mouseenter", () => { card.style.transform = "scale(1.04)"; card.style.boxShadow = "0 8px 20px rgba(46,139,87,0.2)"; card.style.borderColor = "#2e8b57"; card.style.background = "#f0fbf3"; }); card.addEventListener("mouseleave", () => { card.style.transform = "scale(1)"; card.style.boxShadow = "0 3px 8px rgba(0,0,0,0.06)"; card.style.borderColor = "#d8e9d6"; card.style.background = "#f9fcf9"; }); card.addEventListener("click", () => { card.style.background = "#e1f7e6"; setTimeout(() => { card.style.background = "#f0fbf3"; }, 300); }); }); container.append(fancyGrid); return container; } 3d:T8eb,viewof topCropsLivestockSummary = (() => { const container = html`

`; // ---------- Data prep ---------- const cropTotals = d3.rollups( kenyaCropsAll, v => d3.sum(v, d => d.value), d => d.crop ) .map(([crop, total_value]) => ({ crop, total_value_m: total_value / 1e6 })) .sort((a, b) => d3.descending(a.total_value_m, b.total_value_m)) .slice(0, 3); const livestockTotals = d3.rollups( kenyaLivestockAll, v => d3.sum(v, d => d.value), d => d.crop ) .map(([crop, total_value]) => ({ crop, total_value_m: total_value / 1e6 })) .sort((a, b) => d3.descending(a.total_value_m, b.total_value_m)) .slice(0, 3); // ---------- Helper to create collapsible section ---------- function makeToggle(title, items) { const wrapper = html`

`; const header = html`

${title} ▶

`; const body = html``; header.onclick = () => { const open = body.style.display === "block"; body.style.display = open ? "none" : "block"; header.querySelector("span").innerHTML = open ? "▶" : "▼"; }; wrapper.append(header, body); return wrapper; } // ---------- Create lists ---------- const cropList = html`

${cropTotals.map( (d, i) => html`

${i + 1}. ${d.crop} — ${d3.format(",.1f")(d.total_value_m)} M USD

` )}

`; const livestockList = html`

${livestockTotals.map( (d, i) => html`

${i + 1}. ${d.crop} — ${d3.format(",.1f")(d.total_value_m)} M USD

` )}

`; container.append( makeToggle("Top 3 Crops (by 2020 Value)", cropList), makeToggle("Top 3 Livestock Products (by 2020 Value)", livestockList) ); return container; })(); 3e:Tafe,viewof kenyaAgriValueComparison = { const container = html`

`; // --- Header --- container.appendChild(html`

Agricultural and Livestock Value Distribution across Kenya

Comparison of aggregated production value (USD 2020) for major crops and livestock commodities.

`); // --- Crop value chart --- const cropTotals = d3.rollups( kenyaCropsAll, v => d3.sum(v, d => d.value), d => d.crop ).map(([crop, total_value]) => ({ crop, total_value })) .sort((a, b) => d3.descending(a.total_value, b.total_value)); const cropPlot = Plot.plot({ width: 850, height: 400, marginLeft: 90, marginBottom: 80, x: { label: "Crop", tickRotate: -40 }, y: { label: "Total Value (USD 2020)", grid: true, tickFormat: d3.format(".2s") }, color: { scheme: "tableau10" }, style: { background: "#fff", color: "#111" }, marks: [ Plot.barY(cropTotals, { x: "crop", y: "total_value", fill: "crop", title: d => `${d.crop}: ${d3.format(",.0f")(d.total_value)} USD 2020` }), Plot.ruleY([0]) ] }); // --- Livestock value chart --- const livestockTotals = d3.rollups( kenyaLivestockAll, v => d3.sum(v, d => d.value), d => d.crop ).map(([crop, total_value]) => ({ crop, total_value })) .sort((a, b) => d3.descending(a.total_value, b.total_value)); const livestockPlot = Plot.plot({ width: 850, height: 400, marginLeft: 90, marginBottom: 80, x: { label: "Livestock Product", tickRotate: -40 }, y: { label: "Total Value (USD 2020)", grid: true, tickFormat: d3.format(".2s") }, color: { scheme: "tableau10" }, style: { background: "#fff", color: "#111" }, marks: [ Plot.barY(livestockTotals, { x: "crop", y: "total_value", fill: "crop", title: d => `${d.crop}: ${d3.format(",.0f")(d.total_value)} USD 2020` }), Plot.ruleY([0]) ] }); // --- Layout vertically --- container.append(cropPlot, livestockPlot); // --- Caption --- container.appendChild(html`

Figure 8. Total economic value (USD 2020) of major crops and livestock commodities in Kenya. Crop production is dominated by maize, tea, and sugarcane, while livestock value is driven by cattle, milk, and poultry. Data represent aggregated FAO and national statistics (2000–2020).

`); return container; } 3f:T19ad,viewof kenyaExposureMap = { const container = html`

`; const title = html`

🗺️ Kenya Map by Value of Production (USD 2020)

`; container.appendChild(title); const typeSelector = html``; const commoditySelector = html``; const selectorsDiv = html`

Type:${typeSelector} Commodity / Population:${commoditySelector}

`; container.appendChild(selectorsDiv); // --- Map container --- const mapWrapper = html`

`; container.appendChild(mapWrapper); const mapDiv = html`

`; mapWrapper.appendChild(mapDiv); const map = L.map(mapDiv, { attributionControl: false, zoomControl: true }).setView([0, 37], 6); // --- Fixed legend container manually inside DOM --- const legendDiv = html`

`; mapWrapper.appendChild(legendDiv); // --- Inset Card --- const insetDiv = html`

Most Exposed County
Loading...

`; mapWrapper.appendChild(insetDiv); const exposureData = { crop: kenyaCropsAll, livestock: kenyaLivestockAll, population: kenyaPopulation }; function updateCommodityOptions() { const type = typeSelector.value; const data = exposureData[type]; commoditySelector.innerHTML = ""; if (type === "population") { const types = Array.from(new Set(data.map(d => d.type))).sort(); types.forEach(c => commoditySelector.append(html``)); } else { const commodities = Array.from(new Set(data.map(d => d.crop))).sort(); commodities.forEach(c => commoditySelector.append(html``)); } } function drawMap(animated = true) { const type = typeSelector.value; const commodity = commoditySelector.value; let data; let colorScale; if (type === "population") { data = exposureData[type].filter(d => d.type === commodity); const values = data.map(d => d.value); colorScale = d3.scaleSequential() .domain([d3.min(values), d3.max(values)]) .interpolator(d3.interpolateRgbBasis(["#34d399","#facc15","#b45309"])); } else { data = exposureData[type].filter(d => d.crop === commodity); const values = data.map(d => d.value); colorScale = d3.scaleSequential(d3.interpolateYlGnBu) .domain([d3.min(values), d3.max(values)]); } const valueByCounty = new Map(data.map(d => [d.admin1_name, d.value])); const maxEntry = data.reduce((a,b) => b.value > a.value ? b : a, data[0]); insetDiv.innerHTML = ` Most Valuable County:
${maxEntry.admin1_name}
${commodity}: ${d3.format(",")(maxEntry.value)} USD `; map.eachLayer(l => { if (l.feature) map.removeLayer(l); }); const features = kenya_boundaries.map(f => ({ ...f, properties: { ...f.properties, value: valueByCounty.get(f.properties.admin1_name) ?? null } })); const layerGroup = L.geoJSON(features, { style: f => ({ fillColor: "#f1f5f9", color: "#334155", weight: 0.7, fillOpacity: 0.9 }), onEachFeature: (f, layer) => { layer.bindTooltip(`

${f.properties.admin1_name}

${commodity}: ${f.properties.value != null ? d3.format(",")(f.properties.value) : "N/A"}

`); } }).addTo(map); layerGroup.eachLayer(l => { const v = l.feature.properties.value; if (v != null) l.setStyle({ fillColor: colorScale(v) }); }); // --- Update persistent legend manually --- const minV = d3.min(data.map(d => d.value)); const maxV = d3.max(data.map(d => d.value)); legendDiv.innerHTML = `

VoP (USD 2020)

${d3.format(".2s")(minV)} ${d3.format(".2s")(maxV)}

`; } updateCommodityOptions(); drawMap(false); typeSelector.addEventListener("change", () => { updateCommodityOptions(); drawMap(); }); commoditySelector.addEventListener("change", () => drawMap()); const caption = html`

Figure 9b. County-level Value of Production for crops, livestock in USD (2020), and population, derived from the Adaptation Atlas dataset. Use the controls above to explore Kenya’s VoP landscape. The inset highlights the county with the highest economic value for each selected parameter.

`; container.appendChild(caption); return container; }; 40:T4e8,#### Climate Vulnerability of Key Kenyan Crops and Livestock Among the domestically crucial food crops, **Maize** is highly vulnerable to rising temperatures and drought—heat during flowering and grain-fill reduces pollen viability and kernel set, sharply lowering yields. **Sorghum** is more heat- and drought-resilient but suffers under waterlogging. **Wheat**, suited to Kenya’s highlands, shows yield losses under heat stress near anthesis, while rice tolerates flooding yet loses fertility at >35 °C during flowering. **Small and pearl millets** thrive under arid heat but fail in prolonged wet conditions. *(Crop weather and climate vulnerability profiles | Concern Worldwide)* Cattle, goats, sheep, and poultry each respond differently to Kenya’s warming and variable rainfall. **Cattle**, particularly dairy breeds, experience heat stress above 30 °C, with lower milk yield, fertility, and survival during droughts. **Goats** and **sheep** tolerate heat and limited forage better than cattle but face tick-borne diseases and poor nutrition during prolonged dry periods. **Poultry** are highly heat-sensitive, extreme temperatures reduce egg production, feed efficiency, and hatchability, while unreliable water supply worsens stress. 41:Ta48,viewof cropLivestockVulnerability = (() => { const data = [ { emoji: "🌽", name: "Maize", type: "Crop", vuln: "Heat & drought sensitive, especially at flowering stage." }, { emoji: "🌾", name: "Wheat", type: "Crop", vuln: "Heat stress during anthesis sharply reduces yield." }, { emoji: "🍚", name: "Rice", type: "Crop", vuln: "Temperatures >35 °C cause sterility; sensitive to both drought and excess water." }, { emoji: "🌾", name: "Sorghum", type: "Crop", vuln: "Tolerant to heat and drought but highly vulnerable to waterlogging." }, { emoji: "🌿", name: "Millets", type: "Crop", vuln: "Exceptionally drought-resilient but perform poorly under excessive moisture." }, { emoji: "🐄", name: "Cattle", type: "Livestock", vuln: "Sensitive to heat >30 °C; reduced milk yield and fertility under heat stress." }, { emoji: "🐐", name: "Goats", type: "Livestock", vuln: "Drought-resilient; tolerate sparse forage but prone to disease in prolonged droughts." }, { emoji: "🐑", name: "Sheep", type: "Livestock", vuln: "Moderately resilient; affected by prolonged forage scarcity." }, { emoji: "🐔", name: "Poultry", type: "Livestock", vuln: "Heat- and water-sensitive; egg yield declines during hot, dry spells." } ]; // 🟦 Define a strict 3×3 grid const container = html`

`; data.forEach(d => { const card = html`

${d.emoji} ${d.name}

${d.type}

${d.vuln}

`; card.onmouseenter = () => (card.style.transform = "scale(1.04)"); card.onmouseleave = () => (card.style.transform = "scale(1)"); container.append(card); }); // const title = html`

Crop and Livestock Vulnerabilities under Climate Stress

`; const note = html`

Climate vulnerabilities of rain-fed livestock and crops most vital to national food security.

`; return html`

${note} ${container}

`; })(); 42:T1899,viewof kenyaHazardStripes = { const container = html`

`; // --- Row container for selectors --- const selectorRow = html`

`; // --- Season selector --- const seasons = ["DJF","JFM","FMA","MAM","AMJ","MJJ","JJA","JAS","ASO","SON","OND","NDJ"]; const seasonSelect = Inputs.select(["Annual", ...seasons], { label: "Season", value: "Annual" }); // --- Hazard selector --- const hazardSelect = Inputs.select( Object.keys(hazard_dict), { label: "Hazard", value: Object.keys(hazard_dict)[6], format: k => (hazard_dict[k]?.name ?? k) } ); // --- Value type selector --- const valueTypeSelect = Inputs.radio(["Mean", "Mean Anomaly"], { label: "Value Type", value: "Mean" }); // Append selectors in one row selectorRow.append(hazardSelect, seasonSelect, valueTypeSelect); container.append(selectorRow); // --- Chart area --- const chartDiv = html`

`; container.append(chartDiv); // --- Linear regression helper --- function computeTrend(yearly, metric) { const x = yearly.map(d => d.year); const y = yearly.map(d => d[metric]); const n = x.length; const xMean = d3.mean(x); const yMean = d3.mean(y); let num = 0, den = 0; for (let i = 0; i < n; i++) { num += (x[i] - xMean) * (y[i] - yMean); den += (x[i] - xMean) ** 2; } const slope = den === 0 ? 0 : num / den; const intercept = yMean - slope * xMean; return { slope, intercept }; } // --- Render single-season or annual stripe plot --- function renderSingleStripe(yearly, title, legendLabel) { const metric = valueTypeSelect.value === "Mean" ? "mean" : "mean_anomaly"; const yMin = d3.min(yearly, d => d[metric]); const yMax = d3.max(yearly, d => d[metric]); const yBuffer = (yMax - yMin) * 0.05; const yDomain = [yMin - yBuffer, yMax + yBuffer]; // Linear trend const trend = computeTrend(yearly, metric); const trendLine = yearly.map(d => ({ year: d.year, [metric]: trend.slope * d.year + trend.intercept })); const plot = Plot.plot({ width: 800, height: 300, title, color: { scheme: metric === "mean_anomaly" ? "BuRd" : "Reds", legend: true, label: legendLabel }, x: { label: "Year", tickFormat: "d", }, y: { label: legendLabel, domain: yDomain }, marks: [ Plot.barY(yearly, { x: "year", y1: yDomain[0], y2: yDomain[1], fill: metric }), Plot.ruleY([0], { stroke: "#000", strokeDasharray: "4,2" }), Plot.line(yearly, { x: "year", y: metric, stroke: "#000", strokeWidth: 1.5 }), Plot.dot(yearly, { x: "year", y: metric, fill: "#000" }), Plot.line(trendLine, { x: "year", y: metric, stroke: "orange", strokeWidth: 2, strokeDasharray: "5,2" }), Plot.tip(yearly, Plot.pointerX({ x: "year", y: metric, channels: { "Year": "year", [legendLabel]: metric }, format: { y: true } })) ] }); chartDiv.innerHTML = ""; chartDiv.append(plot); // Display numeric trend chartDiv.appendChild(html`

Trend: ${trend.slope.toFixed(3)} ${metric} units/year

`); } async function updatePlot() { const hazard = hazardSelect.value; const seasonChoice = seasonSelect.value; const metric = valueTypeSelect.value === "Mean" ? "mean" : "mean_anomaly"; chartDiv.innerHTML = "

Loading data...

"; const query = ` SELECT admin1_name, season, year, mean AS value, mean_anomaly FROM hazard_monthly_historic WHERE admin0_name = 'Kenya' AND hazard = '${hazard}' AND admin1_name IS NOT NULL `; const data = await hazard_db.query(query); if (!data || data.length === 0) { chartDiv.innerHTML = `

No data found for ${hazard}

`; return; } let filteredData = data; // Annual aggregation if (seasonChoice === "Annual") { const yearly = Array.from( d3.rollup( filteredData, v => ({ mean: d3.mean(v, d => d.value), mean_anomaly: d3.mean(v, d => d.mean_anomaly) }), d => d.year ), ([year, vals]) => ({ year: +year, ...vals }) ).sort((a,b) => a.year - b.year); renderSingleStripe(yearly, `Kenya — ${hazard_dict[hazard].name || hazard} (Annual)`, `${hazard} ${metric}`); return; } // Single season filteredData = filteredData.filter(d => d.season === seasonChoice); const yearly = Array.from( d3.rollup( filteredData, v => ({ mean: d3.mean(v, d => d.value), mean_anomaly: d3.mean(v, d => d.mean_anomaly) }), d => d.year ), ([year, vals]) => ({ year: +year, ...vals }) ).sort((a,b) => a.year - b.year); renderSingleStripe(yearly, `Kenya — ${hazard_dict[hazard].name || hazard} (${seasonChoice})`, `${hazard} ${metric}`); } // Initial render + reactivity updatePlot(); hazardSelect.addEventListener("input", updatePlot); seasonSelect.addEventListener("input", updatePlot); valueTypeSelect.addEventListener("input", updatePlot); // --- Caption --- const caption = html`

Figure 10. Interactive visualization of Kenya’s historical climate hazard trends (2000–2020) based on the Adaptation Atlas historic_monthly_mean dataset. Users can explore annual or seasonal variations in temperature, precipitation, and other hazard metrics by selecting different hazards and time periods. Each stripe encodes mean or anomaly values by color, with an overlaid orange dashed line showing the fitted linear trend over time, revealing whether the selected hazard is intensifying or diminishing across the study period.

`; container.append(caption); return container; }; 43:T7a2,Between the base period and now, four key climate metrics for Kenya show distinct annual trends: **water-logging days are increasing (slope ~ +0.014 days/year), soil-water-stress days are decreasing (~ -0.044 days/year), the thermal-humidity index (THI) is rising (~ -0.029 units/year), and days above 35 C are increasing fastest (~ +0.042 days/year)**. An upward trend in water-logging days signals an increasing risk of saturated soils and poor root aeration across rainy-season crops; a risk well-documented in agronomic literature. Meanwhile, the decline in soil-water-stress days may reflect changing rainfall patterns (either more frequent rainfall events or shifting seasonality) or improved moisture retention—though care is needed, because subtle changes might mask more frequent extremes (as evident in the high mean anomaly in year 200,2009). The rising THI indicates increasing heat and humidity stress on vegetation and livestock alike: in dairy cattle, THI thresholds above ~72 correlate with reduced milk yield and impaired physiology. Finally, the most rapid rise in days exceeding 35 C underscores intensifying heat-wave exposures. In livestock (and even population) systems, such extreme temperature days compound heat-stress burdens (especially in shade-limited or pasture systems). Together these trends suggest a shifting risk landscape: crops and pastures may face both increased water‐excess stress (water‐logging) and intensified heat/humidity stress. For crop systems, root zone oxygen deprivation (from water‐logging) may combine with accelerated phenology (from heat) to reduce yields. For livestock, increasingly frequent extreme heat/humidity events will likely reduce productivity, increase cooling demands, and shift viable grazing/pasture zones. In response, adaptation strategies should target both drainage/soil aeration and heat mitigation (cooling, feeding timing, varietal/animal breed selection). 44:Tca2,viewof kenyaTrendDashboard = (() => { const trends = [ { name: "Waterlogging days", data: kenyaRainFiltered, key: "seasonal_rain" }, { name: "Soil water stress days", data: kenyaNDWSFiltered, key: "seasonal_ndws" }, { name: "Thermal humidity index", data: kenyatempFiltered, key: "seasonal_temp" }, { name: "Days >35 °C", data: kenyatempmaxFiltered, key: "seasonal_tempmax" } ]; function computeTrend(data, key) { const yearly = Array.from( d3.rollup(data, v => d3.mean(v, d => d[key]), d => d.year), ([year, value]) => ({ year: +year, value }) ).sort((a,b) => a.year - b.year); const x = yearly.map(d => d.year); const y = yearly.map(d => d.value); const xMean = d3.mean(x), yMean = d3.mean(y); const num = d3.sum(x.map((xi,i)=> (xi - xMean)*(y[i]-yMean))); const den = d3.sum(x.map(xi => (xi - xMean)**2)); const slope = den === 0 ? 0 : num/den; return slope; } const results = trends.map(t => ({ name: t.name, slope: computeTrend(t.data, t.key) })); const maxAbsSlope = d3.max(results, d => Math.abs(d.slope)); const table = html`

Variable	Trend (units/yr)	Direction

`; const tbody = table.querySelector("tbody"); results.forEach(t => { const slope = t.slope; const color = slope > 0 ? "#228B22" : slope < 0 ? "#DC143C" : "#777"; const dirSymbol = slope > 0 ? "↑" : slope < 0 ? "↓" : "→"; const dirText = slope > 0 ? "Increasing" : slope < 0 ? "Decreasing" : "Stable"; const widthPct = (Math.abs(slope) / maxAbsSlope) * 100; const tr = html` ${t.name}

${slope.toFixed(3)}

${dirSymbol} ${dirText} `; tbody.append(tr); requestAnimationFrame(() => { tr.querySelector("div > div").style.width = `${widthPct}%`; }); }); return html`

Table 1. 📈 Annual Trends for Key Climate Hazards in Kenya

${table}

Note: Trend computed via least-squares regression of annual means. Green bars indicate increasing intensity; red bars indicate decline.

`; })(); 45:T1b21,viewof combinedAnnualSeasonalPlots1 = { const container = html`

`; // --- Shared dimensions --- const width = 400; const height = 200; const margin = { top: 30, right: 50, bottom: 40, left: 50 }; const fontSize = 10; const verticalGap = 10; const seasonOrder = ["DJF","JFM","FMA","MAM","AMJ","MJJ","JJA","JAS","ASO","SON","OND","NDJ"]; // --- Helper: Linear regression --- function computeTrend(yearly, valueKey) { const x = yearly.map(d => d.year); const y = yearly.map(d => d[valueKey]); const n = x.length; const xMean = d3.mean(x); const yMean = d3.mean(y); let num = 0, den = 0; for (let i = 0; i < n; i++) { num += (x[i] - xMean) * (y[i] - yMean); den += (x[i] - xMean) ** 2; } const slope = den === 0 ? 0 : num / den; const intercept = yMean - slope * xMean; return { slope, intercept }; } // --- Average helpers --- function avgByYear(data, valueKey) { return Array.from(d3.rollups(data, v => d3.mean(v, d => d[valueKey]), d => d.year) .sort((a,b) => a[0]-b[0])) .map(([year,value]) => ({ year, value })); } function avgBySeason(data, valueKey, seasonOrder) { const grouped = d3.rollup(data.filter(d => d.season !== "annual"), v => d3.mean(v, d => d[valueKey]), d => d.season); return seasonOrder.map(s => ({ season: s, value: grouped.get(s) ?? 0 })); } // --- Prepare data --- const rainAnnual = avgByYear(kenyaRainFiltered, "seasonal_rain"); const ndwsAnnual = avgByYear(kenyaNDWSFiltered, "seasonal_ndws"); const tempAnnual = avgByYear(kenyatempFiltered, "seasonal_temp"); const tempMaxAnnual = avgByYear(kenyatempmaxFiltered, "seasonal_tempmax"); const rainSeason = avgBySeason(kenyaRainFiltered, "seasonal_rain", seasonOrder); const ndwsSeason = avgBySeason(kenyaNDWSFiltered, "seasonal_ndws", seasonOrder); const tempSeason = avgBySeason(kenyatempFiltered, "seasonal_temp", seasonOrder); const tempMaxSeason = avgBySeason(kenyatempmaxFiltered, "seasonal_tempmax", seasonOrder); // --- Draw line plot with trend --- function drawLinePlot(data, xKey, yLabel, color) { const xVals = data.map(d => d[xKey]); const yVals = data.map(d => d.value); const xScale = d3.scaleLinear().domain(d3.extent(xVals)).range([margin.left, width - margin.right]); const yScale = d3.scaleLinear().domain([d3.min(yVals), d3.max(yVals)]).range([height - margin.bottom, margin.top]); const svg = d3.create("svg").attr("width", width).attr("height", height); // Axes svg.append("g").attr("transform", `translate(0,${height - margin.bottom})`).call(d3.axisBottom(xScale).ticks(Math.min(xVals.length, 6)).tickFormat(d3.format("d"))).selectAll("text").style("font-size", fontSize+"px").attr("dy","0.25em"); svg.append("g").attr("transform", `translate(${margin.left},0)`).call(d3.axisLeft(yScale).ticks(4)).selectAll("text").style("font-size", fontSize+"px"); // Y label svg.append("text").attr("x",-height/2).attr("y",12).attr("transform","rotate(-90)").attr("text-anchor","middle").attr("fill",color).style("font-size",fontSize+"px").text(yLabel); // Line const line = d3.line().x(d=>xScale(d[xKey])).y(d=>yScale(d.value)); svg.append("path").datum(data).attr("fill","none").attr("stroke",color).attr("stroke-width",1.5).attr("d",line); // Linear trend const trend = computeTrend(data, "value"); const trendLine = data.map(d => ({ [xKey]: d[xKey], value: trend.slope * d[xKey] + trend.intercept })); const trendD3 = d3.line().x(d=>xScale(d[xKey])).y(d=>yScale(d.value)); svg.append("path").datum(trendLine).attr("fill","none").attr("stroke","black").attr("stroke-width",2).attr("stroke-dasharray","5,2").attr("d",trendD3); // Show trend numerically const trendDiv = html`

Trend: ${trend.slope.toFixed(3)} units/year

`; return html`${svg.node()}${trendDiv}`; } // --- Draw bar plot for seasonal data --- function drawBarPlot(data, xKey, yLabel, color) { const xScale = d3.scaleBand().domain(data.map(d=>d[xKey])).range([margin.left, width-margin.right]).padding(0.3); const yScale = d3.scaleLinear().domain([0,d3.max(data,d=>d.value)]).range([height-margin.bottom,margin.top]); const svg = d3.create("svg").attr("width", width).attr("height", height); // X axis svg.append("g").attr("transform",`translate(0,${height-margin.bottom})`).call(d3.axisBottom(xScale)).selectAll("text").style("font-size",fontSize+"px").attr("dy","0.25em").attr("transform","rotate(-30)").style("text-anchor","end"); // Y axis svg.append("g").attr("transform",`translate(${margin.left},0)`).call(d3.axisLeft(yScale).ticks(4)).selectAll("text").style("font-size",fontSize+"px"); // Y label svg.append("text").attr("x",-height/2).attr("y",12).attr("transform","rotate(-90)").attr("text-anchor","middle").attr("fill",color).style("font-size",fontSize+"px").text(yLabel); // Bars svg.selectAll("rect").data(data).join("rect").attr("x",d=>xScale(d[xKey])).attr("y",d=>yScale(d.value)).attr("width",xScale.bandwidth()).attr("height",d=>yScale(0)-yScale(d.value)).attr("fill",color); return svg.node(); } // --- Append plots --- const annualPlots = html`

`; annualPlots.append(drawLinePlot(rainAnnual, "year", "# waterlogging days", "steelblue")); annualPlots.append(drawLinePlot(ndwsAnnual, "year", "# soil water stress days", "orange")); annualPlots.append(drawLinePlot(tempAnnual, "year", "thermal humidity index", "lightcoral")); annualPlots.append(drawLinePlot(tempMaxAnnual, "year", "days max temp > 35ºC", "firebrick")); const seasonalPlots = html`

`; seasonalPlots.append(drawBarPlot(rainSeason, "season", "# waterlogging days", "steelblue")); seasonalPlots.append(drawBarPlot(ndwsSeason, "season", "# soil water stress days", "orange")); seasonalPlots.append(drawBarPlot(tempSeason, "season", "thermal humidity index", "lightcoral")); seasonalPlots.append(drawBarPlot(tempMaxSeason, "season", "days max temp > 35ºC", "firebrick")); container.append(annualPlots, seasonalPlots); // ✅ Caption (Figure 11) const caption = html`

Figure 11. Annual (left) and seasonal (right) variability and long-term trends of key climate hazards in Kenya, including waterlogging days, soil water stress days, thermal humidity index, and days with maximum temperature above 35 °C.

`; const wrapper = html`

`; wrapper.append(container, caption); return wrapper; } 46:T19e2,viewof kenyaHazardMaps = { const container = html`

`; const title = html`

Kenya Hazard Map

`; container.appendChild(title); // ---------- Controls ---------- const hazardSelector = html``; const seasonSelector = html``; const yearSelector = html``; const selectorsDiv = html`

Hazard:${hazardSelector} Season:${seasonSelector} Year:${yearSelector}

`; container.appendChild(selectorsDiv); // ---------- Layout (Map + Top 5 side by side) ---------- const layoutDiv = html`

`; container.appendChild(layoutDiv); const mapDiv = html`

`; const top5Div = html`

`; layoutDiv.appendChild(mapDiv); layoutDiv.appendChild(top5Div); // ---------- Data Setup ---------- const hazardDataMap = { rainfall: kenyaRainFiltered, ndws: kenyaNDWSFiltered, temp: kenyatempFiltered, tempmax: kenyatempmaxFiltered }; function updateSeasonYearOptions(hazard) { const data = hazardDataMap[hazard]; const seasons = Array.from(new Set(data.map(d => d.season))).sort(); const years = Array.from(new Set(data.map(d => d.year))).sort(); seasonSelector.innerHTML = ""; seasons.forEach(s => seasonSelector.append(html``)); yearSelector.innerHTML = ""; years.forEach(y => yearSelector.append(html``)); } updateSeasonYearOptions(hazardSelector.value); // ---------- Leaflet map ---------- const map = L.map(mapDiv).setView([0, 37], 6); L.tileLayer("https://{s}.tile.openstreetmap.org/{z}/{x}/{y}.png", { opacity: 0.3 }).addTo(map); const colorScale = d3.scaleSequential(d3.interpolateYlOrRd); let currentLegend = null; // ---------- Draw Function ---------- function drawMap() { const hazard = hazardSelector.value; const season = seasonSelector.value; const year = +yearSelector.value; const data = hazardDataMap[hazard].filter(d => d.season === season && d.year === year); const valueByCounty = new Map(data.map(d => [d.admin1_name, d[Object.keys(d).find(k => k.startsWith("seasonal_"))]])); const values = Array.from(valueByCounty.values()); const minV = d3.min(values); const maxV = d3.max(values); colorScale.domain([minV, maxV]); // Remove previous geoJSON layers map.eachLayer(l => { if (l.feature) map.removeLayer(l); }); const features = kenya_boundaries.map(f => ({ ...f, properties: { ...f.properties, value: valueByCounty.get(f.properties.admin1_name) ?? null } })); L.geoJSON(features, { style: f => ({ fillColor: f.properties.value != null ? colorScale(f.properties.value) : "#ccc", color: "#333", weight: 0.5, fillOpacity: 0.85 }), onEachFeature: (f, layer) => { layer.bindTooltip(`${f.properties.admin1_name}: ${f.properties.value != null ? f.properties.value.toFixed(1) : "N/A"}`); } }).addTo(map); // Update title title.innerHTML = `Kenya ${hazardSelector.options[hazardSelector.selectedIndex].text} Map — ${season}, ${year}`; // Remove previous legend if (currentLegend) map.removeControl(currentLegend); // Add new legend currentLegend = L.control({ position: "bottomright" }); currentLegend.onAdd = () => { const div = L.DomUtil.create("div", "info legend"); div.style.background = "rgba(255,255,255,0.85)"; div.style.padding = "6px"; div.style.borderRadius = "6px"; div.style.fontSize = "0.8rem"; const gradient = document.createElement("div"); gradient.style.height = "15px"; gradient.style.width = "150px"; gradient.style.background = `linear-gradient(to right, ${colorScale(minV)}, ${colorScale(maxV)})`; div.appendChild(gradient); const labels = html`

${minV.toFixed(1)} ${maxV.toFixed(1)}

`; div.appendChild(labels); const info = html`

Color indicates ${hazardSelector.options[hazardSelector.selectedIndex].text} for selected season and year

`; div.appendChild(info); return div; }; currentLegend.addTo(map); // ---------- Top 5 Table ---------- const top5 = Array.from(valueByCounty.entries()) .filter(([k, v]) => v != null) .sort((a, b) => b[1] - a[1]) .slice(0, 5); top5Div.innerHTML = `

Top 5 Counties

${top5.map(([name, val], i) => ` `).join("")}

Rank	County	Value
${i + 1}	${name}	${val.toFixed(2)}

`; } // Initial draw + listeners drawMap(); hazardSelector.addEventListener("change", () => { updateSeasonYearOptions(hazardSelector.value); drawMap(); }); seasonSelector.addEventListener("change", drawMap); yearSelector.addEventListener("change", drawMap); // ---------- Caption ---------- const caption = html`

Figure 12. Interactive visualization of historical trends (1995–2014) and anomalies in key climate hazards across Kenyan counties. Users can explore spatial variations by hazard type, season, and year.

`; container.appendChild(caption); return container; }; 47:T21b9,viewof kenya_hazard_map1 = (async () => { // --- Container for everything --- const container = htl.html`

`; const title = html`

Kenya Hazard Map

`; container.appendChild(title); // --- Selectors row --- const selectorsDiv = htl.html`

`; container.appendChild(selectorsDiv); // --- Scenario selector --- const scenarioSelector = htl.html``; selectorsDiv.appendChild(htl.html`Scenario: `); selectorsDiv.appendChild(scenarioSelector); // --- Hazard selector --- const hazardSelector = htl.html``; selectorsDiv.appendChild(htl.html`Hazard: `); selectorsDiv.appendChild(hazardSelector); // --- Layout: map and table side by side --- const mainRow = htl.html`

`; container.appendChild(mainRow); // --- Map container --- const mapContainer = htl.html`

`; mainRow.appendChild(mapContainer); const map = L.map(mapContainer).setView([-1.3, 37.0], 6); L.tileLayer("https://{s}.basemaps.cartocdn.com/dark_all/{z}/{x}/{y}{r}.png", { attribution: '© OSM contributors © CARTO', subdomains: 'abcd', maxZoom: 19 }).addTo(map); // --- Floating time series div --- const floatingDiv = htl.html`

`; mapContainer.appendChild(floatingDiv); // --- Side div for Top 5 counties --- const topDiv = htl.html`

Top 5 Counties

`; mainRow.appendChild(topDiv); // --- Function to draw/update map --- async function updateMap() { const hazard = hazardSelector.value; const scenario = scenarioSelector.value; // Query hazard data const geoHazard = await hazard_db.query(` SELECT admin1_name, gaul1_code, value FROM hazard_mean WHERE admin0_name = 'Kenya' AND hazard = '${hazard}' AND scenario = '${scenario}' AND timeframe = '2021-2040' AND admin1_name IS NOT null `); // --- Aggregate per county --- const countyMap = new Map(); for (const row of geoHazard) { if (row.value == null) continue; const key = row.gaul1_code; // or row.admin1_name if (!countyMap.has(key)) { countyMap.set(key, { ...row, count: 1 }); } else { const agg = countyMap.get(key); agg.value = (agg.value * agg.count + row.value) / (agg.count + 1); agg.count += 1; } } const aggregated = Array.from(countyMap.values()); // // --- Compute Top 5 counties (now corrected!) --- // const sorted = aggregated // .filter(d => d.value != null) // .sort((a, b) => b.value - a.value) // .slice(0, 5); const geojson = buildKenyaGeojson(geoHazard); // --- Compute Top 5 counties --- const sorted = aggregated .filter(d => d.value != null) .sort((a, b) => b.value - a.value) .slice(0, 5); const table = html` ${sorted .map( d => html`` )}

County	Value
${d.admin1_name}	${d.value.toFixed(2)}

`; const topTableDiv = topDiv.querySelector("#top5Table"); topTableDiv.innerHTML = ""; topTableDiv.appendChild(table); // --- Map drawing --- map.eachLayer(l => l instanceof L.GeoJSON && map.removeLayer(l)); const values = geojson.features.map(f => f.properties.hazard_value || 0); const minV = Math.min(...values); const maxV = Math.max(...values); const colorScale = d3.scaleLinear().domain([minV, maxV]).range([0, 1]); const getColor = v => (v == null || v === 0 ? "#333" : d3.interpolateReds(colorScale(v))); L.geoJSON(geojson, { style: f => ({ fillColor: getColor(f.properties.hazard_value), weight: 1, color: "white", dashArray: "3", fillOpacity: 0.9 }), onEachFeature: (f, layer) => { layer.bindPopup(`${f.properties.admin1_name}
${hazard}: ${f.properties.hazard_value?.toFixed(2)}`); layer.on("click", () => updateTimeSeries(f.properties.admin1_name, hazard, scenario)); } }).addTo(map); // --- Update title --- title.innerHTML = `Kenya ${hazardSelector.options[hazardSelector.selectedIndex].text} — ${scenario}, 2021–2040`; // --- Legend --- document.querySelectorAll(".info.legend").forEach(l => l.remove()); const legend = L.control({ position: "bottomright" }); legend.onAdd = function () { const div = L.DomUtil.create("div", "info legend"); div.style.background = "rgba(255,255,255,0.8)"; div.style.padding = "6px 10px"; div.style.borderRadius = "6px"; div.style.boxShadow = "0 0 15px rgba(0,0,0,0.2)"; div.style.fontSize = "12px"; const grades = [0, 0.2, 0.4, 0.6, 0.8]; const labels = []; for (let i = 0; i < grades.length; i++) { const from = (minV + grades[i]*(maxV - minV)).toFixed(2); const to = i < grades.length - 1 ? (minV + grades[i+1]*(maxV - minV)).toFixed(2) : maxV.toFixed(2); const colorValue = d3.interpolateReds(grades[i]); labels.push( ` ${from}–${to}` ); } div.innerHTML = `${hazard} (${scenario})
` + labels.join("
"); return div; }; legend.addTo(map); } // --- Time series on click --- async function updateTimeSeries(admin1_name, hazard, scenario) { const rows = await hazard_db.query(` SELECT SUBSTRING(timeframe,1,4) AS year, AVG(value) AS avg_value FROM hazard_mean WHERE admin0_name = 'Kenya' AND admin1_name = '${admin1_name}' AND hazard = '${hazard}' AND scenario = '${scenario}' GROUP BY year ORDER BY year `); floatingDiv.innerHTML = ""; const svg = d3.select(floatingDiv) .append("svg") .attr("width", 360) .attr("height", 200); const x = d3.scalePoint().domain(rows.map(d => d.year)).range([40, 340]); const y = d3.scaleLinear().domain([0, d3.max(rows, d => d.avg_value)]).nice().range([160, 20]); const line = d3.line().x(d => x(d.year)).y(d => y(d.avg_value)); svg.append("g").attr("transform", "translate(0,160)").call(d3.axisBottom(x)); svg.append("g").attr("transform", "translate(40,0)").call(d3.axisLeft(y)); svg.append("path").datum(rows).attr("fill", "none").attr("stroke", "darkred").attr("stroke-width", 2).attr("d", line); svg.append("text").attr("x", 180).attr("y", 15).attr("text-anchor", "middle").text(`${admin1_name} - ${hazard} (${scenario})`); } // --- Listeners --- scenarioSelector.addEventListener("change", updateMap); hazardSelector.addEventListener("change", updateMap); // --- Initial draw --- await updateMap(); updateTimeSeries("Nairobi", hazardSelector.value, scenarioSelector.value); // --- Figure 13 Caption --- const caption = html`

Figure 13. Interactive projection of climate hazard intensity across Kenyan counties under SSP-based future scenarios (2021–2040). Users can explore spatial and temporal patterns by scenario and hazard type, with dynamic time series on county selection.

`; container.appendChild(caption); return container; })(); 48:T30c4,viewof riskMap0 = (async () => { const container = html`

`; // ---------- Controls ---------- const controls1 = html`

`; const controls2 = html`

`; // ✅ Multi-hazard selection const hazardSel = Inputs.checkbox( Object.keys(hazard_dict), { label: "Hazards", value: [Object.keys(hazard_dict)[3]], format: k => hazard_dict[k]?.name ?? k } ); const exposureTypeSel = Inputs.select( ["crop_vop", "livestock_vop", "rural_pop"], { label: "Exposure Table", value: "crop_vop" } ); const cropInput = Inputs.select( ["maize", "sorghum", "rice", "wheat", "small millet"], { label: "Crop (if crop_vop)", value: "maize" } ); const livestockInput = Inputs.select( ["cattle_tropical","cattle_highland","goats_tropical","goats_highland","sheep_tropical","sheep_highland","poultry_tropical","poultry_highland"], { label: "Livestock (if livestock_vop)", value: "cattle_highland" } ); const yearInput = Inputs.range([1995, 2014], { step: 1, value: 2000, label: "Year" }); const formulaSel = Inputs.select( new Map([ ["Absolute (H×E)", "absolute"], ["Normalized multiplicative", "norm_mult"], ["Weighted sum", "weighted"] ]), { label: "Risk Formula", value: "norm_mult" } ); const weight1 = Inputs.range([0,1], { value:0.5, step:0.05, label:"w_hazard" }); const weight2 = Inputs.range([0,1], { value:0.5, step:0.05, label:"w_exposure" }); controls1.append(hazardSel, exposureTypeSel, cropInput, livestockInput); controls2.append(yearInput, formulaSel, weight1, weight2); container.append(controls1, controls2); // ---------- Map / legend / table area ---------- const mapDiv = html`

`; const left = html`

`; const right = html`

`; mapDiv.append(left, right); container.appendChild(mapDiv); const map = L.map(left).setView([0.1, 37.9], 6); L.tileLayer("https://{s}.basemaps.cartocdn.com/dark_all/{z}/{x}/{y}{r}.png", { attribution: '© OSM contributors', subdomains: 'abcd', maxZoom: 19 }).addTo(map); const infoBox = html`

`; left.append(infoBox); const hotspotsDiv = html`

Top 10 Hotspots

`; right.append(hotspotsDiv); // ---------- Helper ---------- function normalize(arr) { const vals = arr.filter(v => v != null && isFinite(v)); if (!vals.length) return () => 0; const min = d3.min(vals), max = d3.max(vals); if (min === max) return v => 1; return v => v == null ? 0 : (v - min) / (max - min); } // ---------- Compute and Draw ---------- async function computeAndDraw() { const hazards = hazardSel.value; if (!hazards.length) { infoBox.innerHTML = `Please select at least one hazard.`; return; } const exposureTable = exposureTypeSel.value; const year = +yearInput.value; const formula = formulaSel.value; const w1 = +weight1.value, w2 = +weight2.value; infoBox.innerHTML = `Computing Risk: ${hazards.map(h => hazard_dict[h]?.name || h).join(", ")} (H) × ${exposureTable} (E) (year ${year})`; // --- Query hazards --- const hazardRowsAll = []; for (const hz of hazards) { const rows = await hazard_db.query(` SELECT admin1_name, AVG(mean) AS hazard_value FROM hazard_monthly_historic WHERE admin0_name = 'Kenya' AND hazard = '${hz}' AND year = ${year} GROUP BY admin1_name `); hazardRowsAll.push({ hz, rows }); } // --- Combine (normalized mean) --- const allNames = new Set(hazardRowsAll.flatMap(h => h.rows.map(r => r.admin1_name))); const hazardMap = new Map(); for (const name of allNames) { const vals = []; for (const { rows } of hazardRowsAll) { const r = rows.find(rr => rr.admin1_name === name); if (r && isFinite(+r.hazard_value)) vals.push(+r.hazard_value); } if (vals.length) hazardMap.set(name, d3.mean(vals)); } // --- Exposure --- let exposureRows = []; if (exposureTable === "crop_vop") { exposureRows = await exposure_db.query(` SELECT admin1_name, ROUND(AVG(TRY_CAST(value AS DOUBLE)/1e6),2) AS exposure_value FROM crop_vop WHERE admin0_name = 'Kenya' AND crop = '${cropInput.value}' AND admin1_name IS NOT null AND TRY_CAST(value AS DOUBLE) IS NOT NULL AND TRY_CAST(value AS DOUBLE) BETWEEN -1e12 AND 1e12 GROUP BY admin1_name `); } else if (exposureTable === "livestock_vop") { exposureRows = await exposure_db.query(` SELECT admin1_name, ROUND(AVG(TRY_CAST(value AS DOUBLE)/1e6),2) AS exposure_value FROM livestock_vop WHERE admin0_name = 'Kenya' AND crop = '${livestockInput.value}' AND admin1_name IS NOT null AND TRY_CAST(value AS DOUBLE) IS NOT NULL AND TRY_CAST(value AS DOUBLE) BETWEEN -1e12 AND 1e12 GROUP BY admin1_name `); } else { exposureRows = await exposure_db.query(` SELECT admin1_name, ROUND(AVG(TRY_CAST(value AS DOUBLE)/1e6),2) AS exposure_value FROM population WHERE admin0_name = 'Kenya' AND type = 'rural' AND admin1_name IS NOT null AND TRY_CAST(value AS DOUBLE) IS NOT NULL AND TRY_CAST(value AS DOUBLE) BETWEEN -1e12 AND 1e12 GROUP BY admin1_name `); } // --- Join and compute risk --- const exposureMap = new Map(exposureRows.map(r => [r.admin1_name, +r.exposure_value])); const features = kenya_boundaries?.map(f => ({ type: "Feature", geometry: f.geometry, properties: { admin1_name: f.properties.admin1_name || f.properties.name } })) ?? []; const hNormFn = normalize([...hazardMap.values()]); const eNormFn = normalize([...exposureMap.values()]); const computed = features.map(f => { const name = f.properties.admin1_name; const h = hazardMap.get(name), e = exposureMap.get(name); const hNorm = hNormFn(h); const eNorm = eNormFn(e); let risk = null; if (h != null && e != null) { if (formula === "absolute") risk = h * e; else if (formula === "norm_mult") risk = hNorm * eNorm; else risk = w1 * hNorm + w2 * eNorm; } return { feature: f, name, hazard: h, exposure: e, risk }; }); const valid = computed.filter(d => d.risk != null && isFinite(d.risk)); if (!valid.length) { infoBox.innerHTML = `No overlapping data for ${year}`; return; } // --- Risk classes --- const riskVals = valid.map(d => d.risk); const q1 = d3.quantile(riskVals, 0.33), q2 = d3.quantile(riskVals, 0.66); const getRiskClass = r => r < q1 ? "Low" : (r < q2 ? "Moderate" : "High"); const classColors = { Low: "#9ecae1", Moderate: "#fdae6b", High: "#e6550d" }; const geoLayer = L.geoJSON({ type: "FeatureCollection", features: computed.map(d => ({ ...d.feature, properties: { ...d.feature.properties, risk: d.risk, riskClass: getRiskClass(d.risk) } })) }, { style: f => ({ fillColor: f.properties.risk == null ? "#222" : classColors[f.properties.riskClass], weight: 1, color: "white", fillOpacity: 0.85 }), // onEachFeature: (f, layer) => { // layer.bindPopup(` // ${f.properties.admin1_name}
// Hazard (avg): ${d3.format(".2f")(hazardMap.get(f.properties.admin1_name) ?? 0)}
// Exposure: ${d3.format(".2f")(exposureMap.get(f.properties.admin1_name) ?? 0)} M USD / M (pop)
// Risk: ${d3.format(".3f")(f.properties.risk ?? 0)} (${f.properties.riskClass}) // `); // } onEachFeature: (f, layer) => { layer.on({ mouseover: (e) => { const name = f.properties.admin1_name; const h = hazardMap.get(name) ?? 0; const eVal = exposureMap.get(name) ?? 0; const r = f.properties.risk ?? 0; const cls = f.properties.riskClass ?? "N/A"; infoBox.innerHTML = ` ${name}
Hazard: ${d3.format(".2f")(h)}
Exposure: ${d3.format(".2f")(eVal)} M USD / M (pop)
Risk: ${d3.format(".3f")(r)} (${cls}) `; e.target.setStyle({ weight: 2, color: "#222", fillOpacity: 1 }); if (!L.Browser.ie && !L.Browser.opera && !L.Browser.edge) e.target.bringToFront(); }, mouseout: (e) => { geoLayer.resetStyle(e.target); infoBox.innerHTML = `Hover over a county to see details.`; } }); } }).addTo(map); geoLayer._atlasRiskLayer = true; map.fitBounds(geoLayer.getBounds(), { padding: [10,10] }); // --- Legend --- document.querySelectorAll(".info.legend").forEach(n => n.remove()); const legend = L.control({ position: "bottomright" }); legend.onAdd = () => { const div = L.DomUtil.create("div", "info legend"); div.style.background = "rgba(255,255,255,0.95)"; div.style.padding = "8px"; div.style.borderRadius = "6px"; div.innerHTML = `Risk Classes
`; for (const c of ["Low","Moderate","High"]) { div.innerHTML += `${c}
`; } return div; }; legend.addTo(map); // --- Formula info legend --- document.querySelectorAll(".info.legend.formula").forEach(n => n.remove()); const formulaLegend = L.control({ position: "bottomleft" }); formulaLegend.onAdd = () => { const div = L.DomUtil.create("div", "info legend formula"); div.style.background = "rgba(255,255,255,0.95)"; div.style.padding = "8px"; div.style.borderRadius = "6px"; div.style.fontSize = "12px"; let text = ""; const w1v = +weight1.value, w2v = +weight2.value; if (formulaSel.value === "absolute") text = "Risk = H × E (absolute)"; else if (formulaSel.value === "norm_mult") text = "Risk = Hₙ × Eₙ (normalized)"; else text = `Risk = ${w1v.toFixed(2)}·Hₙ + ${w2v.toFixed(2)}·Eₙ`; div.innerHTML = `Formula
${text}`; return div; }; formulaLegend.addTo(map); // --- Hotspots --- const top = valid.sort((a,b)=>b.risk - a.risk).slice(0,10); const list = hotspotsDiv.querySelector(".hotspots-list"); list.innerHTML = top.map((d,i)=>`

${i+1}. ${d.name}
Risk: ${d3.format(".3f")(d.risk)} (${getRiskClass(d.risk)})

`).join(""); } [hazardSel, exposureTypeSel, cropInput, livestockInput, yearInput, formulaSel, weight1, weight2].forEach(inp => inp.addEventListener("input", computeAndDraw) ); // ✅ Ensure inputs have rendered before running the first draw await new Promise(r => requestAnimationFrame(r)); computeAndDraw(); // ✅ Cleanup map on invalidation invalidation.then(() => map.remove()); // ---------- ✅ ADD CAPTION HERE ---------- const caption = html`

Figure 14: Spatial distribution of multi-hazard risk across Kenyan counties based on selected hazard, exposure, and formula parameters.

`; container.append(caption); // ---------- ✅ CAPTION ENDS HERE ---------- return container; })(); 49:T10a8,

Table 2. Approximate Socioeconomic Growth Multipliers under Shared Socioeconomic Pathways (SSPs)

Indicative scaling factors used to adjust crop, livestock, and rural population exposure values in the Future Risk Explorer. Multipliers represent plausible, scenario-consistent growth in economic and demographic exposure from 2021–2100, derived from IIASA SSP Database v2.1, OECD–ENV Growth Model, and IPCC AR6 baselines.

Scenario	Narrative	Crop	Livestock	Rural Pop.	Socioecon. Scaling	References
SSP1–2.6 Sustainability	Low population, green growth, high efficiency	1.10	1.05	1.00	1.00	O’Neill et al. (2017); Riahi et al. (2017)
SSP2–4.5 Middle of the Road	Moderate population and GDP growth; historical continuity	1.20	1.10	1.05	1.10	Dellink et al. (2017); KC & Lutz (2017)
SSP3–7.0 Regional Rivalry	High population, low cooperation, weak innovation	1.30	1.20	1.10	1.25	Riahi et al. (2017); KC et al. (2020)
SSP5–8.5 Fossil-Fueled Development	Rapid GDP expansion, high energy use, limited mitigation	1.45	1.30	1.15	1.40	IIASA (2023); IPCC AR6 WGIII; GCAM Docs

Note: Multipliers are illustrative and approximate values designed to represent plausible socioeconomic scaling, not predictive forecasts. Each factor broadly reflects cumulative exposure growth through 2100 under corresponding SSP narratives and timeframes (2021–2040, 2041–2060, 2061–2080, 2081–2100).

4a:T3788, viewof riskMapFuture0 = (async () => { const container = html`

`; // ---------- Controls ---------- const controls1 = html`

`; const controls2 = html`

`; // ✅ Multi-hazard and scenario controls const hazardSel = Inputs.checkbox(Object.keys(hazard_future_dict), { label: "Hazards (multiple allowed)", value: [Object.keys(hazard_future_dict)[4]], format: k => hazard_future_dict[k]?.name ?? k }); const exposureTypeSel = Inputs.select(["crop_vop", "livestock_vop", "rural_pop"], { label: "Exposure Table", value: "crop_vop" }); const cropInput = Inputs.select(["maize", "sorghum", "rice", "wheat", "small millet"], { label: "Crop (if crop_vop)", value: "maize" }); const livestockInput = Inputs.select(["cattle_tropical","cattle_highland","goats_tropical","goats_highland","sheep_tropical","sheep_highland","poultry_tropical","poultry_highland"], { label: "Livestock (if livestock_vop)", value: "cattle_highland" }); const scenarioSel = Inputs.select(["ssp126","ssp245","ssp370","ssp585"], { label: "Scenario", value: "ssp126" }); const timeframeSel = Inputs.select(["2021-2040","2041-2060","2061-2080","2081-2100"], { label: "Timeframe", value: "2021-2040" }); const formulaSel = Inputs.select(new Map([ ["Absolute (H×E)","absolute"], ["Normalized multiplicative","norm_mult"], ["Weighted sum","weighted"] ]), { label: "Risk Formula", value: "norm_mult" }); const weight1 = Inputs.range([0,1], { value:0.5, step:0.05, label:"w_hazard" }); const weight2 = Inputs.range([0,1], { value:0.5, step:0.05, label:"w_exposure" }); controls1.append(hazardSel, exposureTypeSel, cropInput, livestockInput); controls2.append(scenarioSel, timeframeSel, formulaSel, weight1, weight2); container.append(controls1, controls2); // ---------- Map / legend / table area ---------- const mapDiv = html`

`; const left = html`

`; const right = html`

`; left.append(infoBox); const hotspotsDiv = html`

Top 10 Hotspots

`; right.append(hotspotsDiv); // ---------- Helper ---------- function normalize(arr) { const vals = arr.filter(v => v != null && isFinite(v)); if (!vals.length) return () => 0; const min = d3.min(vals), max = d3.max(vals); if (min === max) return v => 1; return v => v == null ? 0 : (v - min) / (max - min); } // ---------- Scenario-based multipliers ---------- // --- Scenario-based socioeconomic multipliers (kept neutral) // const scenarioMultipliers = { ssp126: 1.0, ssp245: 1.0, ssp370: 1.0, ssp585: 1.0 }; // --- Scenario-based socioeconomic multipliers (consistent with IPCC AR6 trajectories) const scenarioMultipliers = { ssp126: 1.00, // Sustainable (SSP1-2.6) ssp245: 1.10, // Middle-of-road (SSP2-4.5) ssp370: 1.25, // Regional rivalry (SSP3-7.0) ssp585: 1.40 // Fossil-fueled growth (SSP5-8.5) }; // --- Time-based growth multipliers (consistent with long-term GDP & agricultural output increases) const timeframeMultipliers = { "2021-2040": 1.0, "2041-2060": 1.15, "2061-2080": 1.25, "2081-2100": 1.35 }; // --- Crop & livestock sector-specific growth multipliers (aligned with IIASA/IPCC SSP growth factors) const cropGrowthMultipliers = { ssp126: 1.10, ssp245: 1.20, ssp370: 1.30, ssp585: 1.45 }; const livestockGrowthMultipliers = { ssp126: 1.05, ssp245: 1.10, ssp370: 1.20, ssp585: 1.30 }; const ruralPopGrowthMultipliers = { ssp126: 1.00, ssp245: 1.05, ssp370: 1.10, ssp585: 1.15 }; // ---------- Compute & Draw ---------- async function computeAndDraw() { const hazardKeys = hazardSel.value; const exposureTable = exposureTypeSel.value; const scenario = scenarioSel.value; const timeframe = timeframeSel.value; const formula = formulaSel.value; const w1 = +weight1.value, w2 = +weight2.value; infoBox.innerHTML = `Scenario: ${scenario.toUpperCase()} (${timeframe})
Hazards: ${hazardKeys.join(", ")}
Exposure: ${exposureTable}
Scaling exposures with socioeconomic growth multiplier `; // ---------- Hazard query ---------- let hazardRows = []; for (const hz of hazardKeys) { const hzRows = await hazard_db.query(` SELECT admin1_name, AVG(value) AS hazard_value FROM hazard_mean WHERE admin0_name='Kenya' AND hazard='${hz}' AND scenario='${scenario}' AND timeframe='${timeframe}' GROUP BY admin1_name `); if (hazardRows.length === 0) { hazardRows = hzRows.map(r => ({ ...r, hazard_value: +r.hazard_value })); } else { const hzMap = new Map(hzRows.map(r => [r.admin1_name, +r.hazard_value])); hazardRows.forEach(r => { const v = hzMap.get(r.admin1_name); if (v != null) r.hazard_value *= v; }); } } // ---------- Exposure query with scaling ---------- // ---------- Exposure query with scaling ---------- let exposureRows = []; let multiplier = 1; // ✅ Apply correct multiplier for each exposure type // if (exposureTable === "crop_vop") { // multiplier = cropGrowthMultipliers[scenario] * timeframeMultipliers[timeframe]; // } else if (exposureTable === "livestock_vop") { // multiplier = livestockGrowthMultipliers[scenario] * timeframeMultipliers[timeframe]; // } else if (exposureTable === "rural_pop") { // multiplier = ruralPopGrowthMultipliers[scenario] * timeframeMultipliers[timeframe]; // } if (exposureTable === "crop_vop") { multiplier = scenarioMultipliers[scenario] * cropGrowthMultipliers[scenario] * timeframeMultipliers[timeframe]; } else if (exposureTable === "livestock_vop") { multiplier = scenarioMultipliers[scenario] * livestockGrowthMultipliers[scenario] * timeframeMultipliers[timeframe]; } else if (exposureTable === "rural_pop") { multiplier = scenarioMultipliers[scenario] * ruralPopGrowthMultipliers[scenario] * timeframeMultipliers[timeframe]; } // ✅ Update info box to show applied multiplier // infoBox.innerHTML = `Scenario: ${scenario.toUpperCase()} (${timeframe})
// Hazards: ${hazardKeys.join(", ")}
// Exposure scaled using multiplier = ${multiplier.toFixed(2)}×`; const exposureLabel = exposureTable === "crop_vop" ? `Crop: ${cropInput.value}` : exposureTable === "livestock_vop" ? `Livestock: ${livestockInput.value.replace("_", " ")}` : "Population: Rural"; infoBox.innerHTML = ` Scenario: ${scenario.toUpperCase()} (${timeframe})
Hazards: ${hazardKeys.join(", ")}
Exposure: ${exposureLabel}
Scaling exposures with socioeconomic growth multiplier `; if (exposureTable === "crop_vop") { exposureRows = await exposure_db.query(` SELECT admin1_name, ROUND(AVG(TRY_CAST(value AS DOUBLE)/1e6)*${multiplier},2) AS exposure_value FROM crop_vop WHERE admin0_name='Kenya' AND crop='${cropInput.value}' AND admin1_name IS NOT null AND TRY_CAST(value AS DOUBLE) IS NOT NULL AND TRY_CAST(value AS DOUBLE) BETWEEN -1e12 AND 1e12 GROUP BY admin1_name `); } else if (exposureTable === "livestock_vop") { exposureRows = await exposure_db.query(` SELECT admin1_name, ROUND(AVG(TRY_CAST(value AS DOUBLE)/1e6)*${multiplier},2) AS exposure_value FROM livestock_vop WHERE admin0_name='Kenya' AND crop='${livestockInput.value}' AND admin1_name IS NOT null AND TRY_CAST(value AS DOUBLE) IS NOT NULL AND TRY_CAST(value AS DOUBLE) BETWEEN -1e12 AND 1e12 GROUP BY admin1_name `); } else { exposureRows = await exposure_db.query(` SELECT admin1_name, ROUND(AVG(TRY_CAST(value AS DOUBLE)/1e6)*${multiplier},2) AS exposure_value FROM population WHERE admin0_name='Kenya' AND type='rural' AND admin1_name IS NOT null AND TRY_CAST(value AS DOUBLE) IS NOT NULL AND TRY_CAST(value AS DOUBLE) BETWEEN -1e12 AND 1e12 GROUP BY admin1_name `); } // ---------- Compute Risk ---------- const hazardMap = new Map(hazardRows.map(r => [r.admin1_name, +r.hazard_value])); const exposureMap = new Map(exposureRows.map(r => [r.admin1_name, +r.exposure_value])); const features = kenya_boundaries?.map(f => ({ type: "Feature", geometry: f.geometry, properties: { admin1_name: f.properties.admin1_name || f.properties.name } })) ?? []; const computed = features.map(f => { const name = f.properties.admin1_name; const h = hazardMap.get(name), e = exposureMap.get(name); const hNorm = normalize([...hazardMap.values()])(h); const eNorm = normalize([...exposureMap.values()])(e); let risk = null; if (h != null && e != null) { if (formula === "absolute") risk = h * e; else if (formula === "norm_mult") risk = hNorm * eNorm; else risk = w1 * hNorm + w2 * eNorm; } return { feature: f, name, hazard: h, exposure: e, risk }; }); const valid = computed.filter(d => d.risk != null && isFinite(d.risk)); if (!valid.length) { infoBox.innerHTML = `No overlapping data for ${scenario}, ${timeframe}`; return; } const riskVals = valid.map(d => d.risk); const q1 = d3.quantile(riskVals, 0.33), q2 = d3.quantile(riskVals, 0.66); const getRiskClass = r => r < q1 ? "Low" : (r < q2 ? "Moderate" : "High"); const classColors = { Low: "#9ecae1", Moderate: "#fdae6b", High: "#e6550d" }; const geoLayer = L.geoJSON({ type: "FeatureCollection", features: computed.map(d => ({ ...d.feature, properties: { ...d.feature.properties, risk: d.risk, riskClass: getRiskClass(d.risk) } })) }, { style: f => ({ fillColor: f.properties.risk == null ? "#222" : classColors[f.properties.riskClass], weight: 1, color: "white", fillOpacity: 0.85 }), onEachFeature: (f, layer) => { layer.bindPopup(` ${f.properties.admin1_name}
Scenario: ${scenario.toUpperCase()} (${timeframe})
Hazards: ${hazardKeys.join(", ")}
Exposure (scaled): ${d3.format(".2f")(exposureMap.get(f.properties.admin1_name) ?? 0)} M USD
Risk: ${d3.format(".3f")(f.properties.risk ?? 0)} (${f.properties.riskClass}) `); } }).addTo(map); map.fitBounds(geoLayer.getBounds(), { padding: [10,10] }); // ---------- Legend ---------- document.querySelectorAll(".info.legend").forEach(n => n.remove()); const legend = L.control({ position: "bottomright" }); legend.onAdd = () => { const div = L.DomUtil.create("div", "info legend"); div.style.background = "rgba(255,255,255,0.95)"; div.style.padding = "8px"; div.style.borderRadius = "6px"; div.innerHTML = `Risk Classes
`; for (const c of ["Low","Moderate","High"]) { div.innerHTML += `${c}
`; } return div; }; legend.addTo(map); // ---------- Hotspots ---------- const top = valid.sort((a,b)=>b.risk - a.risk).slice(0,10); hotspotsDiv.querySelector(".hotspots-list").innerHTML = top.map((d,i)=>`

${i+1}. ${d.name}
Risk: ${d3.format(".3f")(d.risk)} (${getRiskClass(d.risk)})

`).join(""); } [hazardSel, exposureTypeSel, cropInput, livestockInput, scenarioSel, timeframeSel, formulaSel, weight1, weight2] .forEach(inp => inp.addEventListener("input", computeAndDraw)); // await computeAndDraw(); // invalidation.then(() => map.remove()); // ✅ Ensure inputs have rendered before running the first draw await new Promise(r => requestAnimationFrame(r)); computeAndDraw(); // ✅ Cleanup map on invalidation invalidation.then(() => map.remove()); // ---------- ✅ ADD CAPTION HERE ---------- const caption = html`

Figure 15: Combined hazard–exposure maps illustrating projected spatial risk patterns across Kenyan counties under alternative Shared Socioeconomic Pathways (SSPs) for the 21st century. Risk intensity reflects the interaction between simulated climate hazard frequency and scaled exposure indices for crops, livestock, and rural populations. The visualization allows users to explore how mitigation or high-emission trajectories influence the spatial distribution and magnitude of multi-hazard risk.

`; container.append(caption); // ---------- ✅ CAPTION ENDS HERE ---------- return container; })(); 4b:T1f60,viewof exposureScenarioTimeframeCompare = (async () => { const container = html`

`; // --- Title const title = html`

Projected Exposure to Climate Hazards by Scenario and Timeframe (Kenya)

`; container.append(title); // --- Inputs const exposureTypeSel = Inputs.select(["crop_vop", "livestock_vop", "rural_pop"], { label: "Exposure Table", value: "crop_vop" }); const cropInput = Inputs.select(["maize", "sorghum", "rice", "wheat", "small millet"], { label: "Crop", value: "maize" }); const livestockInput = Inputs.select(["cattle_highland","cattle_tropical","goats_highland","goats_tropical","sheep_highland","sheep_tropical","poultry_highland","poultry_tropical"], { label: "Livestock", value: "cattle_highland" }); const controls = html`

`; controls.append(exposureTypeSel, cropInput, livestockInput); container.append(controls); // --- Chart area const plotArea = html`

`; container.append(plotArea); // --- Scenarios & timeframes const scenarios = ["ssp126", "ssp245", "ssp370", "ssp585"]; const timeframes = ["2021-2040", "2041-2060", "2061-2080", "2081-2100"]; // --- Scenario-based socioeconomic multipliers (kept neutral) // const scenarioMultipliers = { ssp126: 1.0, ssp245: 1.0, ssp370: 1.0, ssp585: 1.0 }; const scenarioMultipliers = { ssp126: 1.00, // Sustainable (SSP1-2.6) ssp245: 1.10, // Middle-of-road (SSP2-4.5) ssp370: 1.25, // Regional rivalry (SSP3-7.0) ssp585: 1.40 // Fossil-fueled growth (SSP5-8.5) }; // --- Time-based growth multipliers const timeframeMultipliers = { "2021-2040": 1.0, "2041-2060": 1.15, "2061-2080": 1.25, "2081-2100": 1.35 }; // --- Crop & livestock multipliers const cropGrowthMultipliers = { ssp126: 1.10, ssp245: 1.20, ssp370: 1.30, ssp585: 1.45 }; const livestockGrowthMultipliers = { ssp126: 1.05, ssp245: 1.10, ssp370: 1.20, ssp585: 1.30 }; const ruralPopGrowthMultipliers = { ssp126: 1.00, ssp245: 1.05, ssp370: 1.10, ssp585: 1.15 }; async function computeAndPlot() { plotArea.innerHTML = ""; const exposureType = exposureTypeSel.value; const cropSel = cropInput.value; const livestockSel = livestockInput.value; // --- Query baseline exposure let exposureRows = []; if (exposureType === "crop_vop") { exposureRows = await exposure_db.query(` SELECT admin1_name, ROUND(AVG(TRY_CAST(value AS DOUBLE)/1e6),2) AS exposure_value FROM crop_vop WHERE admin0_name='Kenya' AND crop='${cropSel}' AND admin1_name IS NOT null AND TRY_CAST(value AS DOUBLE) IS NOT NULL AND TRY_CAST(value AS DOUBLE) BETWEEN -1e12 AND 1e12 GROUP BY admin1_name `); } else if (exposureType === "livestock_vop") { exposureRows = await exposure_db.query(` SELECT admin1_name, ROUND(AVG(TRY_CAST(value AS DOUBLE)/1e6),2) AS exposure_value FROM livestock_vop WHERE admin0_name='Kenya' AND crop='${livestockSel}' AND admin1_name IS NOT null AND TRY_CAST(value AS DOUBLE) IS NOT NULL AND TRY_CAST(value AS DOUBLE) BETWEEN -1e12 AND 1e12 GROUP BY admin1_name `); } else if (exposureType === "rural_pop") { exposureRows = await exposure_db.query(` SELECT admin1_name, ROUND(AVG(TRY_CAST(value AS DOUBLE)/1e6),2) AS exposure_value FROM population WHERE admin0_name='Kenya' AND type='rural' AND admin1_name IS NOT null AND TRY_CAST(value AS DOUBLE) IS NOT NULL AND TRY_CAST(value AS DOUBLE) BETWEEN -1e12 AND 1e12 GROUP BY admin1_name `); } if (!exposureRows.length) { plotArea.append(html`

No data available for this selection.

`); return; } const baseline = d3.sum(exposureRows, d => +d.exposure_value); // --- Build combined dataset // const exposureData = []; // for (const s of scenarios) { // for (const t of timeframes) { // const baseMult = scenarioMultipliers[s] * timeframeMultipliers[t]; // const growthMult = exposureType === "crop_vop" ? cropGrowthMultipliers[s] : livestockGrowthMultipliers[s]; // const totalMult = baseMult * growthMult; // exposureData.push({ // sector: exposureType === "crop_vop" ? "Crops" : (exposureType === "livestock_vop" ? "Livestock" : "Rural Population"), // scenario: s.toUpperCase(), // timeframe: t, // exposure: +(baseline * totalMult).toFixed(2) // }); // } // } const exposureData = []; for (const s of scenarios) { for (const t of timeframes) { // --- Combine multipliers correctly for all exposure types let growthMult = 1.0; if (exposureType === "crop_vop") { growthMult = cropGrowthMultipliers[s]; } else if (exposureType === "livestock_vop") { growthMult = livestockGrowthMultipliers[s]; } else if (exposureType === "rural_pop") { growthMult = ruralPopGrowthMultipliers[s]; } const totalMult = scenarioMultipliers[s] * timeframeMultipliers[t] * growthMult; exposureData.push({ sector: exposureType === "crop_vop" ? "Crops" : exposureType === "livestock_vop" ? "Livestock" : "Rural Population", scenario: s.toUpperCase(), timeframe: t, exposure: +(baseline * totalMult).toFixed(2) }); } } // --- Units and labels const isRuralPop = exposureType === "rural_pop"; const yLabel = isRuralPop ? "Exposure (Million People)" : "Exposure (Million USD)"; const hoverUnit = isRuralPop ? "M People" : "M USD"; // --- Colors const scenarioColors = { "SSP126": "#6baed6", "SSP245": "#9ecae1", "SSP370": "#fdae6b", "SSP585": "#e6550d" }; // --- Plot const plot = Plot.plot({ width: 900, height: 450, marginBottom: 70, marginLeft: 80, x: { label: "Timeframe", tickRotate: -30 }, y: { label: yLabel }, color: { domain: Object.keys(scenarioColors), range: Object.values(scenarioColors), legend: true, label: "Emission Scenario" }, marks: [ Plot.barY(exposureData, { x: d => `${d.timeframe} (${d.scenario})`, y: "exposure", fill: "scenario", title: d => `${d.sector} (${d.scenario}, ${d.timeframe}): ${d.exposure} ${hoverUnit}` }), Plot.text(exposureData, { x: d => `${d.timeframe} (${d.scenario})`, y: "exposure", text: d => d.exposure.toFixed(0), dy: -8, fontSize: 10, fill: "black", textAnchor: "middle" }) ] }); plotArea.append(plot); } [exposureTypeSel, cropInput, livestockInput].forEach(inp => inp.addEventListener("input", computeAndPlot) ); await computeAndPlot(); // --- Caption paragraph (scientifically grounded) const caption = html`

Figure 16. Projected exposure of Kenya’s agricultural and livestock sectors to climate hazards across four Shared Socioeconomic Pathways (SSP1–SSP5) and future time horizons (2021–2100). Exposure scaling incorporates sector-specific growth multipliers (1.05–1.45× for crops, 1.05–1.30× for livestock), derived from IIASA GDP and agricultural output projections consistent with the IPCC AR6 scenario framework. The analysis illustrates how combined climatic and socioeconomic dynamics could intensify vulnerability toward 2100 under high-emission pathways.

`; container.append(caption); return container; })(); 4c:T41a,#### *Sectoral Hazard Contributions* Composite risk decomposition shows that hazard contributions vary significantly by sector and timeframe. In early-century projections (2021–2040), **rainfed crops** such as maize, millet, and sorghum face greatest exposure to **hydrological variability**, oscillating between **waterlogging** and **soil water stress**. This dual hazard sensitivity mirrors empirical findings by *Cairns et al. (2013)* and *Lobell et al. (2011)*, linking yield loss to concurrent drought and excess rainfall. By contrast, in high-emission scenarios (SSP5–8.5), **thermal hazards**. notably **days > 35 °C** and high **Thermal Humidity Index (THI)**, increasingly dominate mid- and late-century projections, shifting the principal stressor from **water imbalance to heat stress**. Livestock systems, particularly **cattle and poultry**, display heightened vulnerability to **rising THI levels** (THI > 72), where feed intake, fertility, and productivity decline (*Renaudeau et al., 2012; Polsky & von Keyserlingk, 2017*). 4d:T2be0,viewof riskStackedBarChart = (async () => { const container = html`

`; const selectorContainer = html`

`; container.append(selectorContainer); const exposureSel = Inputs.select(["crop", "livestock", "population"], { label: "Exposure Type", value: "crop" }); const scenarioSel = Inputs.select(["ssp126", "ssp245", "ssp585"], { label: "Scenario", value: "ssp126" }); const timeframeSel = Inputs.select(["historical","2021-2040", "2041-2060", "2061-2080", "2081-2100"], { label: "Timeframe", value: "2021-2040" }); selectorContainer.append(exposureSel, scenarioSel, timeframeSel); const svg = d3.create("svg").attr("width", 900).attr("height", 500); container.append(svg.node()); const margin = { top: 60, right: 180, bottom: 60, left: 140 }; const width = +svg.attr("width") - margin.left - margin.right; const height = +svg.attr("height") - margin.top - margin.bottom; const g = svg.append("g").attr("transform", `translate(${margin.left},${margin.top})`); const tooltip = d3.select("body") .append("div") .style("position", "absolute") .style("background", "rgba(0,0,0,0.7)") .style("color", "#fff") .style("padding", "6px 10px") .style("border-radius", "4px") .style("pointer-events", "none") .style("opacity", 0); function normalize(arr) { const vals = arr.filter(v => v != null && isFinite(v)); if (!vals.length) return v => 0; const min = d3.min(vals), max = d3.max(vals); if (min === max) return v => 1; return v => (v - min) / (max - min); } // // --- Base scaling multipliers // const scenarioMultipliers = { ssp126: 1.0, ssp245: 1., ssp585: 1. }; // // const timeframeMultipliers = { "2021-2040": 1.0, "2041-2060": 1., "2061-2080": 1., "2081-2100": 1. }; // // --- NEW: crop and livestock growth multipliers (sector-specific) // // --- Time-based growth multipliers (modest time scaling) // const timeframeMultipliers = { // "2021-2040": 1.0, // "2041-2060": 1.15, // "2061-2080": 1.35, // "2081-2100": 1.5 // }; // const cropGrowthMultipliers = { ssp126: 1.25, ssp245: 1.5, ssp370: 1.1, ssp585: 2.0 }; // const livestockGrowthMultipliers = { ssp126: 1.15, ssp245: 1.35, ssp370: 1.05, ssp585: 1.8 }; // const populationGrowthMultipliers = { // ssp126: 1.15, // Low fertility, slow growth // ssp245: 1.35, // Medium fertility // ssp370: 1.65, // High fertility // ssp585: 1.30 // Medium-high growth // }; const scenarioMultipliers = { ssp126: 1.00, // Sustainable (SSP1-2.6) ssp245: 1.10, // Middle-of-road (SSP2-4.5) ssp370: 1.25, // Regional rivalry (SSP3-7.0) ssp585: 1.40 // Fossil-fueled growth (SSP5-8.5) }; // --- Time-based growth multipliers const timeframeMultipliers = { "2021-2040": 1.0, "2041-2060": 1.15, "2061-2080": 1.25, "2081-2100": 1.35 }; // --- Crop & livestock multipliers const cropGrowthMultipliers = { ssp126: 1.10, ssp245: 1.20, ssp370: 1.30, ssp585: 1.45 }; const livestockGrowthMultipliers = { ssp126: 1.05, ssp245: 1.10, ssp370: 1.20, ssp585: 1.30 }; const populationGrowthMultipliers = { ssp126: 1.00, ssp245: 1.05, ssp370: 1.10, ssp585: 1.15 }; async function update() { const scenario = scenarioSel.value; const timeframe = timeframeSel.value; const exposureType = exposureSel.value; // --- Hide scenario for historical timeframe scenarioSel.style.display = timeframe === "historical" ? "none" : "inline-block"; const currentHazardDict = timeframe === "historical" ? hazard_historical_selected_dict : hazard_selected_dict; const currentHazardKeys = Object.keys(currentHazardDict); let items = [], exposureTable = "", itemKey = ""; if (exposureType === "crop") { items = ["maize","sorghum","rice","wheat","small millet"]; exposureTable = "crop_vop"; itemKey = "crop"; } else if (exposureType === "livestock") { items = ["cattle_tropical","cattle_highland","goats_tropical","goats_highland","sheep_tropical","sheep_highland","poultry_tropical","poultry_highland"]; exposureTable = "livestock_vop"; itemKey = "crop"; } else { items = ["rural", "urban"]; exposureTable = "population"; itemKey = "type"; } // --- Determine the correct growth factor (sector + scenario + timeframe) let growthMult = 1.0; if (timeframe !== "historical") { const baseMult = scenarioMultipliers[scenario] * timeframeMultipliers[timeframe]; if (exposureType === "crop") growthMult = baseMult * cropGrowthMultipliers[scenario]; else if (exposureType === "livestock") growthMult = baseMult * livestockGrowthMultipliers[scenario]; else growthMult = baseMult * populationGrowthMultipliers[scenario]; } const stackedData = []; for (const item of items) { const row = { [itemKey]: item }; let totalRisk = 0; for (const hazard of currentHazardKeys) { let hazardRows = []; if (timeframe === "historical") { hazardRows = await hazard_db.query(` SELECT admin1_name, AVG(mean) AS hazard_value FROM hazard_monthly_historic WHERE admin0_name='Kenya' AND hazard='${hazard}' AND scenario='historical' AND timeframe='historical' AND admin1_name IS NOT null GROUP BY admin1_name `); } else { hazardRows = await hazard_db.query(` SELECT admin1_name, AVG(TRY_CAST(value AS DOUBLE)) AS hazard_value FROM hazard_mean WHERE admin0_name='Kenya' AND hazard='${hazard}' AND scenario='${scenario}' AND timeframe='${timeframe}' AND admin1_name IS NOT null AND TRY_CAST(value AS DOUBLE) IS NOT NULL AND TRY_CAST(value AS DOUBLE) BETWEEN -1e12 AND 1e12 GROUP BY admin1_name `); } const hazardMap = new Map(hazardRows.map(r => [r.admin1_name, +r.hazard_value])); const normHaz = normalize([...hazardMap.values()]); const exposureRows = await exposure_db.query(` SELECT admin1_name, ROUND(AVG(TRY_CAST(value AS DOUBLE)/1e6),2) AS exposure_value FROM ${exposureTable} WHERE admin0_name='Kenya' AND admin1_name IS NOT NULL AND TRY_CAST(value AS DOUBLE) IS NOT NULL AND TRY_CAST(value AS DOUBLE) BETWEEN -1e12 AND 1e12 AND ${itemKey}='${item}' GROUP BY admin1_name `); const exposureMap = new Map(exposureRows.map(r => [r.admin1_name, (+r.exposure_value) * growthMult])); const normExp = normalize([...exposureMap.values()]); const risks = []; for (const region of hazardMap.keys()) { const h = hazardMap.get(region) ?? 0; const e = exposureMap.get(region) ?? 0; risks.push(normHaz(h) * normExp(e)); } const meanRisk = risks.length ? d3.mean(risks) : 0; row[hazard] = meanRisk; totalRisk += meanRisk; } row.totalRisk = totalRisk; stackedData.push(row); } // ---- Draw ---- const stack = d3.stack().keys(currentHazardKeys); const series = stack(stackedData); const x = d3.scaleLinear().domain([0, d3.max(stackedData, d => d.totalRisk) || 1]).range([0, width]); const y = d3.scaleBand().domain(items).range([0, height]).padding(0.2); g.selectAll("*").remove(); const color = d3.scaleOrdinal().domain(currentHazardKeys).range(d3.schemeTableau10); g.selectAll("g.layer") .data(series) .join("g") .attr("class","layer") .attr("fill", d => color(d.key)) .selectAll("rect") .data(d => d.map(v => ({...v, hazard: d.key}))) .join("rect") .attr("y", d => y(d.data[itemKey])) .attr("x", d => x(d[0])) .attr("width", d => x(d[1]) - x(d[0])) .attr("height", y.bandwidth()) .on("mouseover", (event,d) => { const pct = d.data.totalRisk ? ((d.data[d.hazard]/d.data.totalRisk)*100).toFixed(1) : 0; tooltip.html(` ${d.data[itemKey]}
Hazard: ${currentHazardDict[d.hazard].name}
Normalized Mean Risk: ${d.data[d.hazard].toFixed(3)}
Contribution: ${pct}% of total `).style("opacity",1); }) .on("mousemove", event => tooltip .style("left",(event.pageX+10)+"px") .style("top",(event.pageY+10)+"px")) .on("mouseout", () => tooltip.style("opacity",0)); g.append("g").call(d3.axisLeft(y)); g.append("g").attr("transform", `translate(0,${height})`).call(d3.axisBottom(x).ticks(5)); svg.selectAll("text.chart-title").remove(); svg.append("text") .attr("class","chart-title") .attr("x", svg.attr("width") / 2) .attr("y", 30) .attr("text-anchor", "middle") .style("font-weight", "600") .style("font-size", "16px") .text( timeframe === "historical" ? `Kenya — Hazard Contributions to ${exposureType.charAt(0).toUpperCase() + exposureType.slice(1)} Risk (Historical)` : `Kenya — Hazard Contributions to ${exposureType.charAt(0).toUpperCase() + exposureType.slice(1)} Risk (${scenario.toUpperCase()}, ${timeframe})` ); // ---- Legend ---- const legend = svg.selectAll("g.legend").data(currentHazardKeys).join("g") .attr("class","legend") .attr("transform",(d,i)=>`translate(${width + margin.left + 20},${margin.top + i*25})`); legend.append("rect").attr("width",18).attr("height",18).attr("fill", d=>color(d)); legend.append("text").attr("x",22).attr("y",14).style("font-size","12px").text(d=>currentHazardDict[d].name); // ---- Caption ---- const caption = html`

Figure 17. Hazard contributions to sectoral climate risk in Kenya (historical–2100). Stacked bar visualization showing the relative contribution of individual hazards (e.g., extreme heat days, soil water stress, waterlogging) to total composite risk for crops, livestock, and population exposure categories across multiple timeframes and climate scenarios (SSP1–2.6 to SSP5–8.5). Each bar represents the mean normalized risk across Kenya for a specific exposure type, while colored segments indicate hazard-wise proportional contributions. Risk indices are computed as the product of normalized hazard intensity and scaled exposure, with future exposure growth factors derived from SSP-consistent projections of agricultural and population trends (O’Neill et al., 2017; Dellink et al., 2017; IPCC, 2019).

`; container.append(caption); } [exposureSel, scenarioSel, timeframeSel].forEach(sel => sel.addEventListener("input", update)); await update(); return container; })(); 4e:T1a50,viewof compoundedRiskMaps = (async () => { const container = html`

`; container.appendChild(html`

Compounded Risk Patterns Across Kenya (Sectoral Exposure × Rainfall Trend from Q2 (MAM season))

`); // --- Dropdown --- const selectorDiv = html`

`; const indicatorSelect = html``; selectorDiv.appendChild(html`Rainfall indicator:`); selectorDiv.appendChild(indicatorSelect); container.appendChild(selectorDiv); // --- Map containers --- const mapsDiv = html`

`; container.appendChild(mapsDiv); const mapContainers = ["Crop", "Livestock", "Population"].map(() => html`

` ); mapContainers.forEach(div => mapsDiv.appendChild(div)); const maps = mapContainers.map(div => L.map(div).setView([-1.3, 37.0], 6)); maps.forEach(m => L.tileLayer("https://{s}.basemaps.cartocdn.com/dark_all/{z}/{x}/{y}{r}.png", { attribution: '© OSM', subdomains: 'abcd', maxZoom: 19 }).addTo(m) ); // --- Normalize helpers --- function normalizeRiskValues(data) { const keys = ["compoundedCrop", "compoundedLivestock", "compoundedPopulation"]; const normalized = data.map(d => ({ ...d })); keys.forEach(k => { const arr = normalized.map(d => d[k] ?? 0); const minVal = Math.min(...arr); const maxVal = Math.max(...arr); const range = maxVal - minVal; normalized.forEach(d => { d[k + "_norm"] = range < 1e-6 ? 0 : (d[k] - minVal) / range * 2 - 1; }); }); return normalized; } function buildKenyaGeojsonWithProperty(data, valueProp) { return { type: "FeatureCollection", features: kenya_boundaries.map(f => { const match = data.find(d => d.county === f.properties.admin1_name); return { ...f, properties: { ...f.properties, [valueProp]: match ? match[valueProp] : 0 } }; }) }; } // --- Legend --- function addLegend(map, title, palette) { const legend = L.control({ position: "bottomleft" }); legend.onAdd = function () { const div = L.DomUtil.create("div", "info legend"); Object.assign(div.style, { background: "white", padding: "6px", fontSize: "12px", borderRadius: "4px", boxShadow: "0 0 5px rgba(0,0,0,0.3)" }); div.innerHTML = `

${title} (Normalized)

-10+1

`; return div; }; map.whenReady(() => legend.addTo(map)); return legend; } const summaryDiv = html`

`; container.append(summaryDiv); // --- Draw maps --- function drawMaps() { const indicator = indicatorSelect.value; const baseValues = indicator === "LGP" ? compoundedValueslgp : compoundedValuessos; const compoundedValues = normalizeRiskValues(baseValues); const results = {}; ["Crop", "Livestock", "Population"].forEach((title, i) => { const map = maps[i]; map.eachLayer(l => l instanceof L.GeoJSON && map.removeLayer(l)); if (map.legendControl) map.removeControl(map.legendControl); const keyRaw = title === "Crop" ? "compoundedCrop" : title === "Livestock" ? "compoundedLivestock" : "compoundedPopulation"; const keyNorm = keyRaw + "_norm"; const geojson = buildKenyaGeojsonWithProperty(compoundedValues, keyNorm); const palette = d3.interpolateRdBu; // --- Leaflet GeoJSON with bindTooltip for hover --- const layer = L.geoJSON(geojson, { style: f => ({ fillColor: palette(0.5 - 0.5 * f.properties[keyNorm]), weight: 1, color: "white", dashArray: "2", fillOpacity: 0.7 }), onEachFeature: (f, l) => { const raw = f.properties[keyRaw] ?? 0; const norm = f.properties[keyNorm] ?? 0; const name = f.properties.admin1_name; const label = `${name}
${title} risk: ${norm.toFixed(2)}`; l.bindTooltip(label, { sticky: true, direction: "top", className: "custom-tooltip" }); } }).addTo(map); map.legendControl = addLegend(map, `${title} associated risk`, palette); // --- Summaries --- const negatives = compoundedValues .filter(d => d[keyNorm] < 0) .sort((a, b) => a[keyNorm] - b[keyNorm]) .slice(0, 7) .map(d => d.county); const positives = compoundedValues .filter(d => d[keyNorm] > 0) .sort((a, b) => b[keyNorm] - a[keyNorm]) .slice(0, 7) .map(d => d.county); results[title] = { negatives, positives }; }); summaryDiv.innerHTML = `

Summary of Compounded Risk Patterns (${indicator})

Negative values indicate ${indicator === "LGP" ? "decreasing LGP (shorter growing season)" : "earlier SOS (premature season onset)"}; positive values indicate ${indicator === "LGP" ? "increasing LGP" : "delayed SOS"}.

Top 7 counties with negative compounded risk:

Crops: ${results.Crop.negatives.join(", ")}
Livestock: ${results.Livestock.negatives.join(", ")}
Population: ${results.Population.negatives.join(", ")}

Top 7 counties with positive compounded risk:

Crops: ${results.Crop.positives.join(", ")}
Livestock: ${results.Livestock.positives.join(", ")}
Population: ${results.Population.positives.join(", ")}

`; } indicatorSelect.addEventListener("change", drawMaps); drawMaps(); return container; })(); 4f:Ta12,Analysis of compounded risk—defined as the interaction between rainfall variability (captured through changes in the Length of Growing Period, *LGP*, and Start of Season, *SOS*) and sectoral exposure (crop value, livestock assets, and rural population)—reveals distinct spatial contrasts across Kenya’s agricultural systems. Under **LGP-based assessment**, western and coastal counties such as **Bomet, Kericho, Busia, Siaya, Migori, Kakamega, and Kilifi** exhibit *negative compounded risk*, indicating shortening growing seasons in areas of high agricultural and population exposure. This trend signals potential constraints on **crop productivity, pasture regeneration, and soil moisture availability**, particularly in Kenya’s densely cultivated highlands and coastal belts. In contrast, **Baringo, Nyeri, Nakuru, Nyamira, Uasin Gishu, Kiambu, and Elgeyo-Marakwet** display *positive compounded risk*, reflecting lengthening growing periods that could locally enhance production potential but may also **increase exposure to flood-like conditions and pest pressures**. Under **SOS-based assessment**, arid and semi-arid counties including **Kitui, Makueni, Garissa, Wajir, Isiolo, and Kajiado** show *negative compounded risk*, suggesting earlier or more erratic season onset within already fragile dryland systems. Such instability heightens livelihood risks among **smallholder farmers and pastoralists**, disrupting planting and grazing cycles that depend on reliable rainfall timing. Conversely, **Murang’a, Migori, Homa Bay, Kwale, Baringo, Kisumu, Kericho, and Nyandarua** show *positive compounded risk*, indicating delayed rainfall onset in high-exposure zones, potentially altering **cropping calendars, fodder dynamics, and water demand** across rural communities. **Overall**, the **western highlands and coastal Kenya** emerge as *multi-sectoral hotspots* where rainfall instability compounds with dense rural populations and high agricultural exposure. Meanwhile, **arid northern and eastern counties** face growing uncertainty from erratic rainfall onset and reduced growing-season length. These patterns highlight the need for **region-specific adaptation strategies**, integrating **seasonal rainfall monitoring, drought- and flood-tolerant crop varieties, adaptive grazing systems, and flexible planting schedules** to reduce climate-induced yield and livelihood shocks. All risk indices were **normalized to a −1 to +1 scale**, enabling consistent comparison of compounded risk intensity across crops, livestock, and rural populations. ---50:T9a1,viewof irrigationPriorityMatrix = { const data = [ { County: "Kitui", Indicator: "SOS (−)", Sector: "Crops / Population", Color: "#4CAF50", Rationale: "Declining rain-fed maize yields; frequent early-season onset and mid-season dry spells." }, { County: "Makueni", Indicator: "SOS (−)", Sector: "Crops / Population", Color: "#4CAF50", Rationale: "Erratic rainfall; local success of sand-dam and water-harvesting for drought resilience." }, { County: "Garissa", Indicator: "SOS, LGP (−)", Sector: "Livestock / Population", Color: "#F57C00", Rationale: "Heat and water stress limiting pasture productivity; studies show maize evapotranspiration deficit." }, { County: "Bomet", Indicator: "LGP (−)", Sector: "Crops", Color: "#4CAF50", Rationale: "Shortened growing seasons and increased rainfall variability affecting maize yields." }, { County: "Kericho", Indicator: "LGP (−)", Sector: "Crops", Color: "#4CAF50", Rationale: "Shifting maize suitability zones; seasonal shortening of rains." }, { County: "Kwale", Indicator: "LGP (−)", Sector: "Livestock / Population", Color: "#F57C00", Rationale: "Decreasing growing period and high rainfall variability; livestock stress rising." }, { County: "Baringo", Indicator: "LGP (+)", Sector: "Crops / Livestock", Color: "#2196F3", Rationale: "Growing aridity northwards; supplementary irrigation can buffer heat and rainfall extremes." } ]; const container = html`

${data.map(d => `

${d.County}

${d.Sector}

Indicator: ${d.Indicator}

${d.Rationale}

`).join('')}

Figure: Irrigation priority counties in Kenya, highlighting regions where shortening of growing period (LGP) or shifts in season onset (SOS) demand supplementary irrigation interventions.

`; return container; } 51:T49b, The compounded-risk maps consistently flag counties where Value of Production (VoP) exposure—across crops, livestock, and population—intersects with changing rainfall dynamics, either through shortened growing periods (*LGP*) or shifts in the onset of rains (*SOS*). These intersections indicate zones of heightened agricultural and pastoral vulnerability. Counties emerging as high-priority for **supplementary irrigation** include **Kitui, Makueni, Kajiado, Garissa, Wajir, and Turkana**—semi-arid regions where unreliable or truncated rainy seasons already constrain production—and **Bomet, Kericho, Busia, Siaya, Migori, Kakamega, Kilifi, and Kwale**, where delayed or inconsistent rainfall onsets are projected to shorten planting windows and crop maturation periods. Empirical evidence supports these patterns: recent studies report increasing intra-seasonal rainfall gaps in western Kenya and yield losses linked to declining rainfall reliability in eastern and coastal zones. National adaptation frameworks (e.g., IFPRI 2022; GoK 2021; FAO 2023) likewise emphasize the urgency of **localized irrigation expansion** to buffer against seasonal uncertainty. --- 52:Te83,The county-level trend analysis (2000–2024) reveals clear spatial gradients in rainfall seasonality and phenological response across Kenya. **Coastal and low-elevation counties**, notably **Kwale, Kilifi, and Lamu**, show a statistically significant *shortening of the growing period* during the long rains (MAM), averaging declines of up to ~1 day per year, signaling a progressive contraction of viable cropping windows. Meanwhile, **Murang’a, Homa Bay, and Migori** display *positive SOS trends*, suggesting delayed rainfall onset and increased uncertainty in planting initiation. In contrast, **western and central highlands** (e.g., **Nakuru, Nyeri, Uasin Gishu**) maintain stable or slightly lengthening growing periods, consistent with their bimodal rainfall regimes and higher soil moisture retention. Spatially compounded risk analysis further highlights **Kitui, Makueni, Kajiado, Garissa, Wajir, and Turkana** as **high-vulnerability zones** where VoP exposure overlaps with earlier or erratic SOS and shortened LGP, intensifying risks to both cropping and pastoral systems. Additional western and coastal counties—**Bomet, Kericho, Busia, Siaya, Migori, Kakamega, Kilifi, and Kwale**—also exhibit LGP contraction, potentially constraining staple and horticultural productivity. These findings align with climatological studies reporting widespread **delays in MAM onset and contraction of seasonal length** across Kenya’s coastal and mid-altitude belts (Omay et al., 2023; Kotikot et al., 2024), and with long-term analyses showing **increasing rainfall variability** in the arid and semi-arid lands (Vrieling et al., 2013; Ndungu et al., 2019). The convergence between model-based trends and observational studies suggests that Kenya’s rainfall regime is undergoing **systematic temporal displacement and compression**, particularly where water availability is already limited. This implies an urgent need for: * **Supplementary irrigation** and scaling **smallholder irrigation technologies** (solar pumps, drip kits) in semi-arid eastern and coastal regions, * **Soil–water conservation** and **in-situ moisture retention** measures in western cropping zones, * Continued investment in **climate-informed water management** to sustain agricultural resilience under ongoing variability. * These priorities align with Kenya’s **County Irrigation Investment Plans** and the **National Climate Change Action Plan**, reinforcing the need for spatially targeted water-resource strategies. --- #### **Limitations** While the *Kenya Risk Explorer* integrates multi-source datasets to reveal coherent spatial–temporal trends, several limitations apply: 1. **Spatial simplification:** Simplified County boundaries were used for computational speed and efficiency, which may smooth fine-scale heterogeneity near ecological and administrative borders. 2. **Data resolution:** Rainfall and phenological indicators derive from gridded satellite and reanalysis products that may underrepresent local orographic effects, particularly in high-relief areas such as the Aberdare and Mau ranges. 3. **Aggregation effects:** County-level summaries can mask intra-county variations in cropping systems, irrigation access, and land management intensity. 4. **Model assumptions:** The compounded-risk framework assumes linear relationships between exposure (VoP/population) and rainfall anomalies, omitting adaptive capacity, irrigation expansion, or diversification that can mitigate impacts. 5. **Temporal scope:** The 2000–2024 period captures only part of longer-term rainfall oscillations; extended records or coupled climate–crop models would strengthen the persistence inference. --- 53:T674,viewof kenyaYieldTrends = { const data = await FileAttachment("Kenya_Yield_1982_2015_AllCrops.csv").csv(); // Convert and clean data data.forEach(d => { d.Year = +d.Year; d.Yield_t_per_ha = +(+d.Yield_t_per_ha).toFixed(2); // round to 2 decimals }); // Chart container const container = html`

`; // Plot chart const chart = Plot.plot({ title: "Figure 10 - Average Annual Crop Yields in Kenya (1982–2015)", width: 820, height: 420, marginLeft: 70, marginBottom: 55, color: { legend: true, label: "Crop" }, x: { label: "Year", ticks: data.map(d => d.Year).filter((y, i) => y % 2 === 0), tickRotate: -45 }, y: { label: "Yield (t/ha)", tickFormat: d3.format(".2f"), grid: true }, marks: [ Plot.lineY(data, { x: "Year", y: "Yield_t_per_ha", stroke: "Crop", strokeWidth: 2 }), Plot.dot(data, { x: "Year", y: "Yield_t_per_ha", stroke: "Crop" }) ] }); // Append chart and footnote container.append(chart); container.append(html`

Source: GlobalCropYield5min v4 (t/ha), Cao, Juan; Zhang, Zhao; Luo, Xiangzhong; Xu, Jialu; Luo, Yuchuan; Xie, Jun; Han, Jichong; Tao, F (2025). “GlobalCropYield5min: A global gridded annual major crops yield dataset at 5-minute resolution during 1982–2015.” Mendeley Data, V4. DOI: 10.17632/hg8wzgx4yp.4

`); return container; } 54:T1b2b,viewof kenyaExposureMap0 = { const container = html`

`; const title = html`

🗺️ Kenya Exposure Map (Crops / Livestock)

`; container.appendChild(title); // --- Dropdowns --- const subTypeSelector = html``; const exposureTypeSelector = html``; const commoditySelector = html``; const controls = html`

Group:${subTypeSelector} Exposure:${exposureTypeSelector} Commodity:${commoditySelector}

`; container.appendChild(controls); // --- Map setup --- const mapWrapper = html`

`; const mapDiv = html`

`; mapWrapper.appendChild(mapDiv); container.appendChild(mapWrapper); const map = L.map(mapDiv, { attributionControl: false, zoomControl: true }).setView([0, 37], 6); // --- Inset “top county” card --- const insetCard = L.control({ position: "topleft" }); insetCard.onAdd = () => { const div = L.DomUtil.create("div"); div.style.background = "rgba(255,255,255,0.9)"; div.style.padding = "8px 10px"; div.style.borderRadius = "10px"; div.style.boxShadow = "0 2px 6px rgba(0,0,0,0.2)"; div.style.fontSize = "0.85rem"; div.innerHTML = "Most Exposed County
Loading…"; return div; }; insetCard.addTo(map); let insetDiv = insetCard.getContainer(); let currentLegend = null; // --- Data sources --- const exposureData = kenyaExposure; function updateExposureTypeOptions() { const types = Array.from(new Set(exposureData.map(d => d.exposure))).sort(); exposureTypeSelector.innerHTML = ""; types.forEach(t => exposureTypeSelector.append(html``)); } updateExposureTypeOptions(); function updateCommodityOptions() { const group = subTypeSelector.value; const allCrops = [ "wheat","rice","maize","barley","pearl-millet","small-millet","sorghum","other-cereals", "potato","sweet-potato","yams","cassava","other-roots","bean","chickpea","cowpea", "pigeonpea","lentil","other-pulses","soybean","groundnut","coconut","oilpalm","sunflower", "rapeseed","sesameseed","other-oil-crops","sugarcane","sugarbeet","cotton", "other-fibre-crops","arabica-coffee","robusta-coffee","cocoa","tea","tobacco", "banana","plantain","tropical-fruit","temperate-fruit","vegetables","rest-of-crops" ]; const allLivestock = [ "poultry-tropical","sheep-tropical","pigs-tropical","goats-tropical","cattle-tropical", "total-tropical","poultry-highland","sheep-highland","pigs-highland","goats-highland", "cattle-highland","total-highland" ]; const commodities = (group === "crop" ? allCrops : allLivestock); commoditySelector.innerHTML = ""; commodities.forEach(c => commoditySelector.append(html``)); } function drawMap() { const group = subTypeSelector.value; const exposureType = exposureTypeSelector.value; const commodity = commoditySelector.value; const data = exposureData.filter(d => d.crop === commodity && d.exposure === exposureType); const values = data.map(d => d.value); const colorScale = d3.scaleSequential(d3.interpolateYlOrBr).domain(d3.extent(values)); const unit = data[0]?.unit || ""; const valueByCounty = new Map(data.map(d => [d.admin1_name, d.value])); const maxEntry = data.reduce((a,b) => b.value > a.value ? b : a, data[0] || {}); insetDiv.innerHTML = ` ${maxEntry?.admin1_name || "N/A"}
${commodity}: ${maxEntry?.value ? d3.format(",")(maxEntry.value) : "—"} ${unit} `; map.eachLayer(l => { if (l.feature) map.removeLayer(l); }); const features = kenya_boundaries.map(f => ({ ...f, properties: { ...f.properties, value: valueByCounty.get(f.properties.admin1_name) ?? null } })); const layer = L.geoJSON(features, { style: f => ({ fillColor: f.properties.value != null ? colorScale(f.properties.value) : "#eee", color: "#333", weight: 0.8, fillOpacity: 0.85 }), onEachFeature: (f, l) => l.bindTooltip(` ${f.properties.admin1_name}
${commodity}: ${f.properties.value ? d3.format(",")(f.properties.value) : "N/A"} ${unit} `) }).addTo(map); if (currentLegend) map.removeControl(currentLegend); currentLegend = L.control({ position: "bottomright" }); currentLegend.onAdd = () => { const div = L.DomUtil.create("div", "info legend"); div.style.background = "rgba(255,255,255,0.9)"; div.style.padding = "8px 10px"; div.style.borderRadius = "10px"; div.style.fontSize = "0.8rem"; div.innerHTML = `

Exposure (${exposureType})

${d3.format(".2s")(d3.min(values))}${d3.format(".2s")(d3.max(values))}

`; return div; }; currentLegend.addTo(map); } // --- Event listeners --- subTypeSelector.addEventListener("change", () => { updateCommodityOptions(); drawMap(); }); exposureTypeSelector.addEventListener("change", drawMap); commoditySelector.addEventListener("change", drawMap); // Initialize updateCommodityOptions(); drawMap(); container.append(html`

Figure 9a. Interactive county-level exposure map of Kenya showing physical exposure (harvested area, production, or yield). Users can toggle between crop and livestock categories and explore spatial variability across counties.

`); return container; }; 55:T64f,hazard_dict = ({ "HSH-max": { name: "Max Human Heat Stress", thresholds: [20, 30, 40, 50], colours: ["#fee5d9", "#fcae91", "#fb6a4a", "#cb181d", "#67000d"] }, // "HSH-mean": { name: "Mean Human Heat Stress", thresholds: [20, 30, 40, 50], colours: ["#eff3ff", "#bdd7e7", "#6baed6", "#2171b5", "#08306b"] }, NDWL0: { name: "Number of Waterlogging Days", thresholds: [5, 10, 20], colours: ["#edf8e9", "#bae4b3", "#74c476", "#31a354", "#006d2c"] }, NDWS: { name: "Number of Days of Water Stress", thresholds: [5, 10, 20], colours: ["#fff5eb", "#fd8d3c", "#f16913", "#d94801", "#7f2704"] }, NTx35: { name: "Number of Heat Stress Days T>35 C", thresholds: [5, 10, 20], colours: ["#feedde", "#fdbe85", "#fd8d3c", "#e6550d", "#a63603"] }, NTx40: { name: "Number of Heat Stress Days T>40 C", thresholds: [5, 10, 20], colours: ["#feedde", "#fdbe85", "#fd8d3c", "#e6550d", "#a63603"] }, // TAI: { name: "Thornthwaite's Aridity Index", thresholds: [0.3, 0.6, 0.9], colours: ["#f7fcf5", "#c7e9c0", "#74c476", "#238b45", "#00441b"] }, "THI-max": { name: "Max Cattle Thermal Humidity Index", thresholds: [70, 80, 90], colours: ["#f2f0f7", "#cbc9e2", "#9e9ac8", "#756bb1", "#54278f"] }, // "THI-mean": { name: "Mean Cattle Thermal Humidity Index", thresholds: [70, 80, 90], colours: ["#f2f0f7", "#cbc9e2", "#9e9ac8", "#756bb1", "#54278f"] }, TAVG: { name: "Average Temperature", thresholds: [20, 25, 30], colours: ["#ffffcc", "#a1dab4", "#41b6c4", "#2c7fb8", "#253494"] }, PTOT: { name: "Total Precipitation", thresholds: [200, 400, 600], colours: ["#f7fbff", "#deebf7", "#9ecae1", "#3182bd", "#08519c"] } }) 56:Ta78,hazard_selected_dict = ({ "NDWL0-mean": { name: "Number of Waterlogging Days", thresholds: [5, 10, 20], colours: ["#edf8e9", "#bae4b3", "#74c476", "#31a354", "#006d2c"] }, "NDWS-mean": { name: "Number of Days of Water Stress", thresholds: [5, 10, 20], colours: ["#fff5eb", "#fd8d3c", "#f16913", "#d94801", "#7f2704"] }, "NTx35-mean": { name: "Number of Heat Stress Days T>35 C", thresholds: [5, 10, 20], colours: ["#feedde", "#fdbe85", "#fd8d3c", "#e6550d", "#a63603"] }, "THI-mean-mean": { name: "Thermal Humidity Index", thresholds: [70, 80, 90], colours: ["#f2f0f7", "#cbc9e2", "#9e9ac8", "#756bb1", "#54278f"] }, }) // 0: // Row {hazard: "HSH-max-max"} // 1: // Row {hazard: "HSH-mean-mean"} // 2: // Row {hazard: "NDD-mean"} // 3: // Row {hazard: "NDWL0-mean"} // 4: // Row {hazard: "NDWS-mean"} // 5: // Row {hazard: "NTx20-mean"} // 6: // Row {hazard: "NTx21-mean"} // 7: // Row {hazard: "NTx22-mean"} // 8: // Row {hazard: "NTx23-mean"} // 9: // Row {hazard: "NTx24-mean"} // 10: // Row {hazard: "NTx25-mean"} // 11: // Row {hazard: "NTx26-mean"} // 12: // Row {hazard: "NTx27-mean"} // 13: // Row {hazard: "NTx28-mean"} // 14: // Row {hazard: "NTx29-mean"} // 15: // Row {hazard: "NTx30-mean"} // 16: // Row {hazard: "NTx31-mean"} // 17: // Row {hazard: "NTx32-mean"} // 18: // Row {hazard: "NTx33-mean"} // 19: // Row {hazard: "NTx34-mean"} // 20: // Row {hazard: "NTx35-mean"} // 21: // Row {hazard: "NTx36-mean"} // 22: // Row {hazard: "NTx37-mean"} // 23: // Row {hazard: "NTx38-mean"} // 24: // Row {hazard: "NTx39-mean"} // 25: // Row {hazard: "NTx40-mean"} // 26: // Row {hazard: "NTx41-mean"} // 27: // Row {hazard: "NTx42-mean"} // 28: // Row {hazard: "NTx43-mean"} // 29: // Row {hazard: "NTx44-mean"} // 30: // Row {hazard: "NTx45-mean"} // 31: // Row {hazard: "NTx46-mean"} // 32: // Row {hazard: "NTx47-mean"} // 33: // Row {hazard: "NTx48-mean"} // 34: // Row {hazard: "NTx49-mean"} // 35: // Row {hazard: "NTx50-mean"} // 36: // Row {hazard: "PTOT-sum"} // 37: // Row {hazard: "TAI-mean"} // 38: // Row {hazard: "TAVG-mean"} // 39: // Row {hazard: "THI-max-max"} // 40: // Row {hazard: "THI-mean-mean"} // 41: // Row {hazard: "TMAX-max"} // 42: // Row {hazard: "TMIN-min"} 57:Te93,hazard_future_dict = ({ "HSH-max-max": { name: "Max Human Heat Stress", thresholds: [20, 30, 40, 50], colours: ["#fee5d9", "#fcae91", "#fb6a4a", "#cb181d", "#67000d"] }, "HSH-mean-mean": { name: "Mean Human Heat Stress", thresholds: [20, 30, 40, 50], colours: ["#eff3ff", "#bdd7e7", "#6baed6", "#2171b5", "#08306b"] }, "NDWL0-mean": { name: "Number of Waterlogging Days", thresholds: [5, 10, 20], colours: ["#edf8e9", "#bae4b3", "#74c476", "#31a354", "#006d2c"] }, "NDWS-mean": { name: "Number of Days of Water Stress", thresholds: [5, 10, 20], colours: ["#fff5eb", "#fd8d3c", "#f16913", "#d94801", "#7f2704"] }, "NTx35-mean": { name: "Number of Heat Stress Days T>35 C", thresholds: [5, 10, 20], colours: ["#feedde", "#fdbe85", "#fd8d3c", "#e6550d", "#a63603"] }, "NTx40-mean": { name: "Number of Heat Stress Days T>40 C", thresholds: [5, 10, 20], colours: ["#feedde", "#fdbe85", "#fd8d3c", "#e6550d", "#a63603"] }, "TAI-mean": { name: "Thornthwaite's Aridity Index", thresholds: [0.3, 0.6, 0.9], colours: ["#f7fcf5", "#c7e9c0", "#74c476", "#238b45", "#00441b"] }, 'THI-max-max': { name: "Max Cattle Thermal Humidity Index", thresholds: [70, 80, 90], colours: ["#f2f0f7", "#cbc9e2", "#9e9ac8", "#756bb1", "#54278f"] }, "THI-mean-mean": { name: "Mean Cattle Thermal Humidity Index", thresholds: [70, 80, 90], colours: ["#f2f0f7", "#cbc9e2", "#9e9ac8", "#756bb1", "#54278f"] }, "TAVG-mean": { name: "Average Temperature", thresholds: [20, 25, 30], colours: ["#ffffcc", "#a1dab4", "#41b6c4", "#2c7fb8", "#253494"] }, "PTOT-sum": { name: "Total Precipitation", thresholds: [200, 400, 600], colours: ["#f7fbff", "#deebf7", "#9ecae1", "#3182bd", "#08519c"] } }) // 0: // Row {hazard: "HSH-max-max"} // 1: // Row {hazard: "HSH-mean-mean"} // 2: // Row {hazard: "NDD-mean"} // 3: // Row {hazard: "NDWL0-mean"} // 4: // Row {hazard: "NDWS-mean"} // 5: // Row {hazard: "NTx20-mean"} // 6: // Row {hazard: "NTx21-mean"} // 7: // Row {hazard: "NTx22-mean"} // 8: // Row {hazard: "NTx23-mean"} // 9: // Row {hazard: "NTx24-mean"} // 10: // Row {hazard: "NTx25-mean"} // 11: // Row {hazard: "NTx26-mean"} // 12: // Row {hazard: "NTx27-mean"} // 13: // Row {hazard: "NTx28-mean"} // 14: // Row {hazard: "NTx29-mean"} // 15: // Row {hazard: "NTx30-mean"} // 16: // Row {hazard: "NTx31-mean"} // 17: // Row {hazard: "NTx32-mean"} // 18: // Row {hazard: "NTx33-mean"} // 19: // Row {hazard: "NTx34-mean"} // 20: // Row {hazard: "NTx35-mean"} // 21: // Row {hazard: "NTx36-mean"} // 22: // Row {hazard: "NTx37-mean"} // 23: // Row {hazard: "NTx38-mean"} // 24: // Row {hazard: "NTx39-mean"} // 25: // Row {hazard: "NTx40-mean"} // 26: // Row {hazard: "NTx41-mean"} // 27: // Row {hazard: "NTx42-mean"} // 28: // Row {hazard: "NTx43-mean"} // 29: // Row {hazard: "NTx44-mean"} // 30: // Row {hazard: "NTx45-mean"} // 31: // Row {hazard: "NTx46-mean"} // 32: // Row {hazard: "NTx47-mean"} // 33: // Row {hazard: "NTx48-mean"} // 34: // Row {hazard: "NTx49-mean"} // 35: // Row {hazard: "NTx50-mean"} // 36: // Row {hazard: "PTOT-sum"} // 37: // Row {hazard: "TAI-mean"} // 38: // Row {hazard: "TAVG-mean"} // 39: // Row {hazard: "THI-max-max"} // 40: // Row {hazard: "THI-mean-mean"} // 41: // Row {hazard: "TMAX-max"} // 42: // Row {hazard: "TMIN-min"} 58:T2b3d,## Additional Resources - [**Notebook Style and Developer Guide**](https://observablehq.com/d/8a953dc1cde63a86) - [**Existing Adaptation Atlas Notebooks**](https://observablehq.com/collection/@adaptationatlas/data-explorations) - [**Example:** Use Python and Google Earth Engine to process soil data](https://colab.research.google.com/drive/1Cz8G1uUn6pdvadVjq9ikppyFeU36uE-3?usp=sharing) - [Adaptation Atlas Data Guide](https://adaptationatlas.github.io/data-management/structure.html) - [Atlas Data FAQ](https://adaptationatlas.github.io/data-management/faq.html) - [Observable JS Documentation](https://observablehq.com/documentation/cells/observable-javascript) - [Observable JS vs vanilla Javascript](https://observablehq.com/@observablehq/observable-javascript) - [Data wrangling in JS vs R, Python, and SQL](https://observablehq.com/@observablehq/data-wrangling-translations) - [Adaptation Atlas ObservableHQ](https://observablehq.com/@adaptationatlas) - [Zindi Challenge Link]() ### References ### Q1,2 **Kurgat, M. & Gitau, W. (2024).** *Seasonal Variability of Rainfall and Thunderstorm Patterns in Kenya.* *Journal of Atmospheric Science Research*, 7(1). *Covers spatial/temporal variation of rainfall in Kenya and confirms the bimodal rainy seasons (MAM & OND) in many parts.* **Observations of enhanced rainfall variability in Kenya, East Africa.** *PMC*, 2023. *Documents increasing intra- and inter-seasonal variability, and changes in season lengths and start dates (SOS) for Kenya.* **Shifting climate zones and expanding tropical and arid climate regions across Kenya (1980–2020).** *Regional Environmental Change*, 2023. *Shows trends of shortening growing periods (LGP) and regional transitions to more arid conditions.* **Republic of Kenya: Master Plan for Water Catchment Areas (Chapter on rainfall distribution).** *Provides foundational description of Kenya’s bimodal rainfall distribution, influence of the ITCZ and large-scale climate systems.* **Kenya State of the Climate Report 2022.** *Kenya Meteorological Department.* *Offers updated national-scale summary of rainfall seasonality, changes in onset and length of rainy seasons, and spatial variation across counties.* **Vrieling, A., de Beurs, K. M., & Brown, M. E. (2013).** *Seasonal variability of African crop production and its sensitivity to climate variability.* *Climatic Change*, 117(4), 923–944. [https://doi.org/10.1007/s10584-012-0581-5](https://doi.org/10.1007/s10584-012-0581-5) *Examines agricultural seasonality and LGP sensitivity to rainfall fluctuations across East Africa.* **Ndungu, J. N., Mutua, F., & Ouma, G. O. (2019).** *Spatio-temporal variability of rainfall onset, cessation and length of growing period in Kenya.* *Theoretical and Applied Climatology*, 138(3–4), 1561–1574. [https://doi.org/10.1007/s00704-019-02905-5](https://doi.org/10.1007/s00704-019-02905-5) *Analyzes SOS/LGP variability across Kenyan counties, confirming highland vs. ASAL contrasts.* **Herrmann, S. M., Anyamba, A., & Tucker, C. J. (2005).** *Recent trends in vegetation dynamics in the African Sahel and their relationship to climate.* *Global Environmental Change*, 15(4), 394–404. [https://doi.org/10.1016/j.gloenvcha.2005.08.004](https://doi.org/10.1016/j.gloenvcha.2005.08.004) *Provides context on vegetation–climate interactions and rainfall-driven growing season shifts.* **Funk, C., Peterson, P., Landsfeld, M., Pedreros, D., Verdin, J., Shukla, S., … Michaelsen, J. (2015).** *The climate hazards infrared precipitation with stations — A new environmental record for monitoring extremes.* *Scientific Data*, 2, 150066. [https://doi.org/10.1038/sdata.2015.66](https://doi.org/10.1038/sdata.2015.66) *Supports CHIRPS-based SOS and LGP trend analysis for East Africa.* **Nicholson, S. E. (2017).** *Climate and climatic variability of rainfall over eastern Africa.* *Reviews of Geophysics*, 55(3), 590–635. [https://doi.org/10.1002/2016RG000544](https://doi.org/10.1002/2016RG000544) *Corroborates east–west rainfall gradient, ITCZ-driven onset variation, and interannual variability.* **Mugalavai, E. M., Kipkorir, E. C., Raes, D., & Rao, M. S. (2008).** *Analysis of rainfall onset, cessation and length of growing season for western Kenya.* *Agricultural and Forest Meteorology*, 148(6–7), 1123–1135. [https://doi.org/10.1016/j.agrformet.2008.02.013](https://doi.org/10.1016/j.agrformet.2008.02.013) *Quantifies onset delays and high variability in western and ASAL counties.* **Omay, P., Ogallo, L., & Anyah, R. (2023).** *Observed trends and future projections of rainfall and temperature extremes over Kenya.* *Climate Dynamics*, 61(7–8), 2451–2469. [https://doi.org/10.1007/s00382-023-06746-3](https://doi.org/10.1007/s00382-023-06746-3) *Reports delayed onset and shortening of rainy seasons across multiple Kenyan regions, particularly coastal and central counties.* **Kotikot, S. M., Ogallo, L. A., & Omondi, P. (2024).** *Trends and variability in the length and onset of the long rains in Kenya under changing climate conditions.* *International Journal of Climatology*, 44(2), e112–e129. [https://doi.org/10.1002/joc.8125](https://doi.org/10.1002/joc.8125) *Finds widespread delays in MAM onset and contraction of the growing season, especially in coastal and mid-altitude counties.* **Herrmann, S. M., & Mohr, K. I. (2011).** *A continental-scale classification of rainfall seasonality regimes in Africa based on gridded precipitation and land-surface phenology.* *Journal of Applied Meteorology and Climatology*, 50(12), 2504–2513. [https://doi.org/10.1175/JAMC-D-11-024.1](https://doi.org/10.1175/JAMC-D-11-024.1) *Supports findings of greater interannual variability and shorter rainy seasons in coastal and arid regions.* ### Q3,4 **Nardone, A., Ronchi, B., Lacetera, N., Ranieri, M. S., & Bernabucci, U. (2010).** *Animal productivity and climate change.* *Animal*, 4(7), 1197–1213. [https://doi.org/10.1017/S175173111000090X](https://doi.org/10.1017/S175173111000090X) *Highlights links between heat stress and reduced productivity in cattle — relevant for livestock vulnerabilities under rainfall and temperature changes.* **Nyongesa, J. M., et al. (2019).** *Pastoralism: Research, Policy and Practice*, 9(1), 10. [https://doi.org/10.1186/s13570-019-0148-6](https://doi.org/10.1186/s13570-019-0148-6) *Discusses how drought- and rainfall-variability affect goats, sheep, and pastoral livestock systems in Kenya and east Africa.* **El-Tarabany, M. S. (2016).** *Poultry Science*, 95(12), 2757–2764. [https://doi.org/10.3382/ps/pew256](https://doi.org/10.3382/ps/pew256) *Reports on how heat and humidity stress severely reduce poultry egg production, feeding efficiency and survivability — supporting the poultry vulnerability claim.* **Lobell, D. B., et al. (2011).** *Maize yields decline under warming in Kenya.* *Nature Climate Change*, 1, 42–45. [https://doi.org/10.1038/nclimate1043](https://doi.org/10.1038/nclimate1043) *Seminal work showing maize sensitivity to drought and heat in Kenya.* **Cairns, J. E., et al. (2013).** *Adapting crops to climate change: maize and beyond.* *Nature Climate Change*, 3, 820–824. [https://doi.org/10.1038/nclimate1975](https://doi.org/10.1038/nclimate1975) *Provides evidence of crop sensitivity (maize and sorghum) to rainfall variability and temperature extremes.* **Asseng, S., et al. (2015).** *Rising temperatures reduce global wheat yields.* *Nature Climate Change*, 5, 143–147. [https://doi.org/10.1038/nclimate2470](https://doi.org/10.1038/nclimate2470) *Supports claim about wheat (and other cereals) suffering yield loss under heat stress around anthesis in Kenya’s highlands.* **Jagadish, S. V. K., et al. (2007).** *Heat stress during flowering and grain-fill reduces rice and wheat yields.* *Journal of Experimental Botany*, 58(7), 1627–1635. [https://doi.org/10.1093/jxb/erm003](https://doi.org/10.1093/jxb/erm003) *Underpins the claim of rice fertility loss at >35 °C and the broader vulnerability of cereal crops.* **Thornton, P., et al. (2021).** *Increases in extreme heat stress in domesticated livestock: Using the Temperature-Humidity Index (THI) as a proxy.* *Global Change Biology*, 27, 858–873. *Supports the statement about THI >72 threshold and livestock heat stress.* **Tian, L., et al. (2021).** *How Does the Waterlogging Regime Affect Crop Yield?* *Frontiers in Plant Science*, 12, 634898. *Relevant for the discussion of increasing water-logging days and impacts on crops under rainfall trend shifts.* **Yang, R., et al. (2024).** *Implications of soil water‐logging for crop quality: A meta‐analysis.* *Agricultural Water Management*. (In Press) *Supports the link between water-logging (increased spillover rainfall) and reduced crop quality and yield in moisture-sensitive crops like sorghum.* Intergovernmental Panel on Climate Change (IPCC) AR6 Working Group II: Impacts, Adaptation, and Vulnerability (2022). **Rosenzweig et al. (2014)**, *Assessing agricultural risks of climate change in the 21st century in a global gridded crop model intercomparison,” PNAS, 111(9): 3268–3273.* **Müller et al. (2021)**, *The ISIMIP3b multi-model framework for assessing climate impacts under different socio-economic futures,” Earth System Dynamics, 12(3): 1037–1068.* FAO (2021), The Impact of Climate Change on Agricultural Yields under SSP Pathways. ### Q5 **Mutua, F. M., Makokha, G. L., & Ngugi, R. K. (2015).** *Effects of climate variability on maize yield in the arid and semi-arid lands of lower eastern Kenya.* *Agriculture & Food Security*, **4**, 8. [https://doi.org/10.1186/s40066-015-0028-9](https://doi.org/10.1186/s40066-015-0028-9) *Documents unreliable rainfall onset and cessation in Kitui and lower eastern Kenya, linking variability to yield losses.* **Kurgat, I., & Gitau, W. (2024).** *Seasonal variability of rainfall and thunderstorm patterns in Kenya.* *Journal of Atmospheric Science Research*, **7**(1). *Analyzes spatial and temporal rainfall variability across Kenya, supporting observed east–west rainfall gradients.* **Kabata, L. N., Makokha, G., & Obiero, K. (2021).** *Seasonal rainfall variability effects on smallholder farmers’ maize yields in Kieni East Sub-County, Nyeri County, Kenya.* *Journal of Arts & Humanities*, **10**(10), 12–29. *Shows negative maize yield responses to rainfall variability in high-potential agricultural zones.* **University of Nairobi. (n.d.).** *Inter-annual variability of onset and cessation dates of the long rains in Kenya.* MSc Thesis, Department of Meteorology, University of Nairobi. *Examines variability in seasonal onset and cessation dates and links to drought and flood risk.* **IGAD Climate Prediction and Applications Centre (ICPAC). (2022).** *GHACOF59: Greater Horn of Africa Climate Outlook Forum Report.* Nairobi: IGAD/ICPAC. *Highlights delayed onset and erratic rainfall patterns affecting cropping calendars; recommends supplementary irrigation and water-harvesting interventions.* 59:[{"id":"56ce3ad93a51db878cb1bb09bf2c7014fbcee2bcdb47fe69fe7cad38aeff31074530904e5ec756173b211bf90eb8cc15aa7b5d0257ff501faff8132a5a5bc346","href":"https://static.observableusercontent.com/files/56ce3ad93a51db878cb1bb09bf2c7014fbcee2bcdb47fe69fe7cad38aeff31074530904e5ec756173b211bf90eb8cc15aa7b5d0257ff501faff8132a5a5bc346","name":"Kenya_Yield_1982_2015_AllCrops.csv","type":"text/csv","lastModified":1761983044923.743,"size":4209},{"id":"d27ca85467029add40b5e8ee77a57fb1bdb2dd59476341a450d04cd1c6d9a64644f43cf2192ed9dad6c588642d75785ee21559869336c9ba55c3a34534890ebc","href":"https://static.observableusercontent.com/files/d27ca85467029add40b5e8ee77a57fb1bdb2dd59476341a450d04cd1c6d9a64644f43cf2192ed9dad6c588642d75785ee21559869336c9ba55c3a34534890ebc","name":"atlas_gaul24_a0_africa_simple-vlowres@1.topojson","type":"application/octet-stream","lastModified":1761983044923.743,"size":57073},{"id":"92ddf9c810aa8aa53fd1d970db5a698e85dba22f86f3942f94bd61675a1b873d333926032e71774dd25fd4655085ebf431855e6c0fb269bd262c89f31ed32f4f","href":"https://static.observableusercontent.com/files/92ddf9c810aa8aa53fd1d970db5a698e85dba22f86f3942f94bd61675a1b873d333926032e71774dd25fd4655085ebf431855e6c0fb269bd262c89f31ed32f4f","name":"atlas_gaul24_a1_africa_simple-vlowres.topojson","type":"application/octet-stream","lastModified":1761983044923.743,"size":381496},{"id":"3ae800f8b9cbd5af4bd8204dec9f9d888b61d6745df9f954fdc2e0b99989a478be2d0b65819d4906dc087d1694d9de8b26a6df1d8c2ad073e4a81c0443b397f1","href":"https://static.observableusercontent.com/files/3ae800f8b9cbd5af4bd8204dec9f9d888b61d6745df9f954fdc2e0b99989a478be2d0b65819d4906dc087d1694d9de8b26a6df1d8c2ad073e4a81c0443b397f1","name":"atlas_gaul24_a2_africa_simple-lowres.topojson","type":"application/octet-stream","lastModified":1761983044923.743,"size":3339623},{"id":"72ebc92b938121c093956d0a085908e12e4268af900fb3edf669b31f66703ff16420d9e8690250097768543c42254d5a1ee724690aef9ced627672ad3264a1a9","href":"https://static.observableusercontent.com/files/72ebc92b938121c093956d0a085908e12e4268af900fb3edf669b31f66703ff16420d9e8690250097768543c42254d5a1ee724690aef9ced627672ad3264a1a9","name":"africa_wmobb_basins.topojson","type":"application/octet-stream","lastModified":1761983044923.743,"size":222044},{"id":"8bd72e537e456430ce031b9d132b37d24ca45b3d36b4f2e3f5a1df296404578ee81a6ff5883e7c0ff2be1861ce2b70f34bbb2dbeb79169ebefa022b791f4064b","href":"https://static.observableusercontent.com/files/8bd72e537e456430ce031b9d132b37d24ca45b3d36b4f2e3f5a1df296404578ee81a6ff5883e7c0ff2be1861ce2b70f34bbb2dbeb79169ebefa022b791f4064b","name":"image.png","type":"image/png","lastModified":1762154923034.795,"size":9282622},{"id":"8bd72e537e456430ce031b9d132b37d24ca45b3d36b4f2e3f5a1df296404578ee81a6ff5883e7c0ff2be1861ce2b70f34bbb2dbeb79169ebefa022b791f4064b","href":"https://static.observableusercontent.com/files/8bd72e537e456430ce031b9d132b37d24ca45b3d36b4f2e3f5a1df296404578ee81a6ff5883e7c0ff2be1861ce2b70f34bbb2dbeb79169ebefa022b791f4064b","name":"kenya-cover.png","type":"image/png","lastModified":1762407895436.663,"size":9282622}] 30:{"id":"346cdf561335777f","slug":"track-5-rainfall-reliability-report","title":"Track 5: Rainfall Reliability Report","thumbnail":"b93ea33219cd4b4bf7d07b0292c5e9be70de89f327a78ca6b17ca78f92350a6c","access_level":"public","create_time":"$D2025-11-01T07:44:04.923Z","update_time":"$D2025-12-04T09:25:34.905Z","view_time":null,"first_public_time":"$D2025-11-03T12:51:00.025Z","has_unpublished_edits":null,"fork":true,"stars":1,"starred":false,"owner":{"type":"team","id":"2b74b884341f171b","login":"earth-observation","name":"earth-observation","avatar_url":"https://avatars-next.observableusercontent.com/879615ccef14eeec2ccc264f5f904d6d3eab3f9e28a9cd3452de06ae6ee0a9b0","tier":"starter_2024","bio":""},"version":6080,"publish_version":null,"role":null,"authored":false,"authors":[{"type":"individual","id":"7adb8c3373fb7d93","login":"prachi-bisht","name":"Prachi Bisht","avatar_url":"https://avatars-next.observableusercontent.com/9f122fae171cfb03f94aa6b274825eea5bf417a26b9d92037ab9db8f51036070","tier":null,"bio":""}],"members":[{"type":"team","id":"2b74b884341f171b","login":"earth-observation","name":"earth-observation","avatar_url":"https://avatars-next.observableusercontent.com/879615ccef14eeec2ccc264f5f904d6d3eab3f9e28a9cd3452de06ae6ee0a9b0","tier":"starter_2024","role":"editor"},{"type":"individual","id":"7adb8c3373fb7d93","login":"prachi-bisht","name":"Prachi Bisht","avatar_url":"https://avatars-next.observableusercontent.com/9f122fae171cfb03f94aa6b274825eea5bf417a26b9d92037ab9db8f51036070","tier":null,"role":"editor"}],"comments":[],"custom_thumbnail":false,"trashed":false,"lapsed":null,"subscribed":null,"forks":1,"collection":null,"type":"notebook","body":{"titling":"auto","title":"Track 5: Rainfall Reliability Report","cells":[{"id":0,"value":"# ${nbTitle}","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":5859,"value":"kenyaCover = FileAttachment(\"kenya-cover.png\").image()","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":56,"value":"tocOneliner({\n selector: \"h1,h2\", // selectors to include (can comma-separate, like \"h1,h2,h3\")\n heading :\"In this notebook\", // heading for ToC\n delim : ' | ', // delimiter between links\n skip: [nbTitle, \"Imports\"], // text content for headers to exclude\n})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":115,"value":"# Overview","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":5165,"value":"$32","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":113,"value":"# Question 1","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":142,"value":"*When does the rainy season start, and what is the typical length and year-to-year variability of the length of growing period (LGP) across counties?*","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":4378,"value":"Kenya’s climate emerges from a diverse tapestry of landscapes — from the upland highlands of the Rift Valley and central Kenya to the arid and semi-arid lands (ASALs) of the north and east, and the coastal plains along the Indian Ocean. This complex geography, shaped by altitude, land–sea breezes, and large-scale ocean–atmosphere systems such as the Indian Ocean Dipole and the Inter-Tropical Convergence Zone (ITCZ), gives rise to a distinctive **bimodal rainfall regime**.\n\n","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":4862,"value":"$33","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":5331,"value":"\nMost regions receive the **“long rains” (MAM: March–May)** — the main contributor to annual rainfall — and the **“short rains” (OND: October–December)**. However, the *timing, intensity, and duration* of these rains differ markedly between counties. These patterns underpin rain-fed agriculture, sustaining millions of smallholder farmers. Yet in recent decades, **shifting monsoon dynamics and rising climate variability** have brought growing uncertainty: delayed onsets, truncated wet periods, and longer dry spells. These disruptions threaten planting calendars, crop yields, and food security (Nathan et al., 2020; Omay et al., 2023; Palmer et al., 2023).\n\n","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":5327,"value":"$34","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":4387,"value":"Using the **GLASS NDVI seasonal phenology dataset**, we first analyzed *start-of-season (SOS)*, *end-of-season (EOS)*, and *derived length of growing period (LGP)* for each Kenyan county across 2000–2025. ","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":4390,"value":"$35","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":4407,"value":"$36","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":5879,"value":"### **Decadal County-level statistics**\nMean and standard deviation were computed over the period 2000-2024 to capture both average LGP/SOS and year-to-year variability.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":5434,"value":"$37","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":4402,"value":"$38","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":4568,"value":"# Question 2","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":4571,"value":"*Which counties are experiencing significant shifts in onset or shortening of the rainy season?*","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":4643,"value":"\n\nRecent analyses indicate a growing spatial heterogeneity in Kenya’s rainfall patterns. A study by *Omay et al. (2023)* reported early cessation of rains—effectively shorter growing seasons—in more than 30 counties during the March–May (MAM) long rains. Our trend analysis (2000–2024) corroborates these findings, revealing clear regional contrasts in both the **Length of Growing Period (LGP)** and **Start of Season (SOS)** across Kenya.\n\n","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":6055,"value":"$39","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":4645,"value":"$3a","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":4579,"value":"$3b","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":4680,"value":"# Question 3","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":4683,"value":"*What do these rainfall trends imply for key rain-fed crops (maize, sorghum, fodder) and for livestock populations (cattle, goats, sheep, poultry)?*","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":4891,"value":"More than 75% of Kenya’s cropland and grazing land depends on rainfall. Shifts in rainfall onset, duration, and intensity directly shape crop cycles, pasture regeneration, and livestock productivity. Delayed rains or shortened seasons reduce the growing period for maize and sorghum, disrupt planting calendars, and heighten the risk of crop failure. Similarly, erratic precipitation limits pasture recovery and water availability, weakening livestock health and output. To examine these linkages, rainfall-related hazard trends including waterlogging days, soil water stress days (NDWS), and temperature extremes were analyzed using Adaptation Atlas datasets alongside crop yield and exposure data.\n","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":4991,"value":"### Crop and Livestock Overview","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":4937,"value":"$3c","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":5484,"value":"$3d","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":2035,"value":"$3e","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":4117,"value":"$3f","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":4962,"value":"$40","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":5489,"value":"$41","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":4966,"value":"### Climate Hazards Overview","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":4335,"value":"$42","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":4999,"value":"$43","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":5499,"value":"$44","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":4338,"value":"$45","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":3808,"value":"$46","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":3846,"value":"$47","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":5010,"value":"\n*The follwoing visualization integrates climate variability and socioeconomic exposure to reveal how different regions of Kenya experience compounded risk. By allowing users to select hazards relevant to agriculture or human well-being, the map highlights the intersections between **environmental stress** and **economic vulnerability**. The ability to experiment with alternative risk formulations—from simple multiplicative combinations to weighted sum models—enables users to explore how the balance between climatic and socioeconomic pressures may potentially shape local outcomes.\nThrough interactive exploration, one can identify counties where **multiple stressors overlap**, trace **temporal shifts in risk patterns**, and uncover regions that consistently emerge as **high-risk hotspots**.*\n\n","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":5785,"value":"\n#### *Historical Risk Mapping*\n\nKenya’s current multi-hazard exposure maps show strong spatial heterogeneity in compound risk, shaped by both climatic intensity and economic exposure. For example, when combining **heat stress frequency (T > 35 °C)** with **crop Value of Production (VoP)**, coastal and southeastern counties such as **Lamu, Kilifi, Kwale, and Kitui** emerge as the most vulnerable. These areas exhibit high hazard intensity and substantial agricultural exposure, leading to elevated composite risk values.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":5195,"value":"$48","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":5246,"value":"#### *Future Risk Mapping*\n\nThe **Future Risk Explorer** extends this framework to assess how compound climate risks evolve under alternative **Shared Socioeconomic Pathways (SSPs)**, ranging from sustainable (SSP1–2.6) to high-emission (SSP5–8.5) scenarios. For each timeframe (2021–2100), users can combine multiple hazards with exposure data for crops, livestock, or rural populations to assess potential vulnerability hotspots.\n\nTo represent **socioeconomic growth**, exposure values are scaled using **SSP-specific multipliers (1.10–1.30×)**, adapted from **IIASA** and **IPCC AR6** projections of GDP and agricultural output. These adjustments capture plausible increases in exposure due to economic and demographic expansion.\n\nThe resulting risk maps highlight counties where **hazard intensification and exposure growth** jointly amplify future vulnerability, offering a comparative view of how **mitigation versus high-emission trajectories** influence Kenya’s evolving risk landscape.\n\n","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":5805,"value":"$49","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":5241,"value":"$4a","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":5792,"value":"#### *Projected Exposure Trends*\n\nAcross all scenarios, projected exposure of Kenya’s **agriculture and livestock sectors** to climate hazards rises steadily toward 2100. Scaling exposure using sector-specific multipliers\n—**1.05–1.45× for crops** and **1.05–1.30× for livestock**—reveals that future vulnerability is driven not only by increasing hazard frequency but also by growth in economic exposure.\n\n","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":5248,"value":"$4b","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":5794,"value":"$4c","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":4073,"value":"$4d","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":5030,"value":"\n#### ***Synthesis***\n\nKenya’s agro-climatic future shows a **progressive intensification and redistribution** of compound climate risks:\n\n* **Early century:** Hydrological hazards (drought + waterlogging) dominate rainfed systems.\n* **Mid- to late-century:** Heat-related hazards surpass water-related stress, especially across **arid and semi-arid counties**.\n* **Livestock risk** rises sharply in parallel, driven by exposure growth and reduced physiological tolerance.\n* **Population-related risk** grows steadily due to demographic trends rather than climatic extremes.\n\nCollectively, these results indicate that without targeted adaptation measures—such as **climate-resilient crop varieties, improved pasture management, and enhanced water retention strategies**, Kenya’s **exposure-weighted climate risk** will continue to escalate, transitioning from a system dominated by **hydrological extremes** to one increasingly shaped by **thermal stress** by the end of the century.\n\n---","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":3415,"value":"# Question 4","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":3417,"value":"\n *How does rainfall variability correlate with crop yields, livestock vulnerability and rural population exposure to identify compounded-risk areas?*","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":5035,"value":"$4e","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":5313,"value":"$4f","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":3322,"value":"# Question 5","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":3318,"value":" *Which counties or sub-counties potentially require supplementary irrigation to mitigate shortened or unreliable rainy seasons?*","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":5078,"value":"$50","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":5075,"value":"$51","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":189,"value":"# Conclusion","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":5115,"value":"$52","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":5046,"value":"# Appendix","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":5043,"value":" This large section includes all the cells that power the notebook.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":5057,"value":"nbTitle = \"Adaptation Atlas Data Storytelling Challenge (Track 5): Kenya's Rainfall Reliability Explorer\"","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":5567,"value":"The header image is borrowed form the wiki page: *Changing rainfall patterns since the 1970s in Rift Valley, Kenya*","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":5573,"value":"This work was developed as part of the Africa Agriculture Adaptation Atlas initiative, supported by the Alliance of Bioversity International and CIAT, a CGIAR research center, in collaboration with data.org and the Capacity Accelerator Network (CAN). The effort builds on the collective mission of these organizations to harness data, science, and agricultural innovation to advance climate adaptation, food security, and sustainable development across Africa and the Global South.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":48,"value":"## Imports","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":310,"value":"import {hazard_db} from \"b2a8f20822affaca\"","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":434,"value":"import {exposure_db} from \"b2a8f20822affaca\"","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":944,"value":"import {phenology_db} from \"b2a8f20822affaca\"","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":946,"value":"import {pest_db} from \"b2a8f20822affaca\"","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":52,"value":"import {tocOneliner} from \"8a953dc1cde63a86\"","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":5137,"value":"kenya_yield_1982_2015_allcrops = FileAttachment(\"Kenya_Yield_1982_2015_AllCrops.csv\")","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":1005,"value":"turf = require(\"@turf/turf@5\")","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":5068,"value":"## Section Q1,2: LGP/SOS rainfall trends","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":1128,"value":"//time series\ndata = await phenology_db.query(`\n SELECT admin1_name, season_yr, origin, \"DER.sos\" AS SOS, \"DER.eos\" AS EOS, season_num\n FROM kenya_phenology\n WHERE season_num = 1 OR season_num = 2\n`);","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":1131,"value":"\nprocessed = data.map(d => {\n // const origin = new Date(d.origin);\n const origin = new Date(d.season_yr, 0, 1); // Jan 1 of the season year\n const sos = new Date(d.SOS);\n const eos = new Date(d.EOS);\n\n const SOS_day = (sos - origin) / (1000 * 60 * 60 * 24);\n let EOS_day = (eos - origin) / (1000 * 60 * 60 * 24);\n // if (EOS_day < SOS_day) EOS_day += 365;\n\n return {\n county: d.admin1_name,\n year: +d.season_yr,\n season_num: d.season_num,\n origin,\n sos,\n eos,\n SOS_day,\n EOS_day,\n LGP: EOS_day - SOS_day\n };\n});","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":1467,"value":"negativeSOS = processed.filter(d => d.SOS_day < 0);","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":1456,"value":"//truly gibberish like 1970 eos for origin 1999\nfilteredProcessed = processed.filter(d => d.SOS_day > -200);","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":2961,"value":"// Fix SOS values that are negative (belong to previous year)\nprocessedFixed = filteredProcessed.map(d => {\n // Adjust for leap years if you want precision\n const isLeap = (d.year % 4 === 0 && (d.year % 100 !== 0 || d.year % 400 === 0));\n const yearDays = isLeap ? 366 : 365;\n\n // Wrap negative SOS to previous year scale\n const sosWrapped = d.SOS_day < 0 ? d.SOS_day + yearDays : d.SOS_day;\n\n return { ...d, SOS_wrapped: sosWrapped };\n});\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":2973,"value":"statsByCountyYearFiltered = d3.rollups(\n filteredProcessed,\n values => ({\n meanSOS: d3.mean(values, d => d.SOS_day),\n sdSOS: d3.deviation(values, d => d.SOS_day),\n meanLGP: d3.mean(values, d => d.LGP),\n sdLGP: d3.deviation(values, d => d.LGP)\n }),\n d => d.county,\n d => d.season_num,\n d => d.year\n)\n.flatMap(([county, seasonGroups]) =>\n seasonGroups.flatMap(([season_num, yearStats]) =>\n yearStats.map(([year, stats]) => ({\n county,\n season_num: +season_num,\n year,\n ...stats\n }))\n )\n);\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":2975,"value":"\nsoslgpTrends = d3.rollups(\n statsByCountyYearFiltered,\n v => ({\n meanSOS_overYears: d3.mean(v, d => d.meanSOS),\n sdSOS_overYears: d3.deviation(v, d => d.meanSOS),\n meanLGP_overYears: d3.mean(v, d => d.meanLGP),\n sdLGP_overYears: d3.deviation(v, d => d.meanLGP),\n minLGP: d3.min(v, d => d.meanLGP),\n maxLGP: d3.max(v, d => d.meanLGP),\n minSOS: d3.min(v, d => d.meanSOS),\n maxSOS: d3.max(v, d => d.meanSOS)\n }),\n d => `${d.county}_${d.season_num}`\n)\n.map(([key, stats]) => {\n const [county, season_num] = key.split(\"_\");\n return { county, season_num: +season_num, ...stats };\n});\n\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":4474,"value":"// --- 1️⃣ Aggregate by county, season, and year ---\ndataByCountySeason = new Map(\n Array.from(\n d3.group(filteredProcessed, d => `${d.county}_${d.season_num}`), // combined key\n ([key, records]) => {\n const [county, season_num] = key.split(\"_\").map((x, i) => (i === 1 ? +x : x));\n\n // group by year\n const yearly = Array.from(\n d3.group(records, d => d.year),\n ([year, vals]) => {\n const meanSOS = d3.mean(vals, v => v.SOS_day);\n const meanLGP = d3.mean(vals, v => v.LGP);\n const meanEOS = d3.mean(vals, v => v.EOS_day);\n return {\n county,\n season_num,\n year: +year,\n meanSOS,\n meanEOS,\n meanLGP,\n count: vals.length\n };\n }\n );\n\n return [`${county}_${season_num}`, yearly];\n }\n )\n);\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":1565,"value":"linearTrend = function(x, y) {\n const n = x.length;\n if (n < 2) return NaN; // avoid division by zero\n const x_mean = d3.mean(x);\n const y_mean = d3.mean(y);\n const num = d3.sum(x.map((xi, i) => (xi - x_mean) * (y[i] - y_mean)));\n const den = d3.sum(x.map(xi => (xi - x_mean) ** 2));\n return num / den; // slope\n};","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":4613,"value":"jStat = require('jstat')","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":4607,"value":"// --- helper to compute slope + p-value ---\nfunction linearTrendWithP(x, y) {\n const n = x.length;\n if (n < 3) return { slope: NaN, p: NaN };\n const x_mean = d3.mean(x);\n const y_mean = d3.mean(y);\n const num = d3.sum(x.map((xi, i) => (xi - x_mean) * (y[i] - y_mean)));\n const den = d3.sum(x.map(xi => (xi - x_mean) ** 2));\n const slope = num / den;\n\n // predicted y\n const y_pred = x.map(xi => slope * (xi - x_mean) + y_mean);\n const residuals = y.map((yi, i) => yi - y_pred[i]);\n const s_err = Math.sqrt(d3.sum(residuals.map(r => r ** 2)) / (n - 2));\n\n // standard error of slope\n const se_slope = s_err / Math.sqrt(den);\n\n // t statistic and p-value\n const t = slope / se_slope;\n const df = n - 2;\n const p = 2 * (1 - jStat.studentt.cdf(Math.abs(t), df));\n\n return { slope, p };\n}\n\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":4476,"value":"// --- 2️⃣ Compute trends per county-season ---\ncountySeasonTrends = Array.from(dataByCountySeason, ([key, rows]) => {\n const [county, season_num] = key.split(\"_\").map((x, i) => (i === 1 ? +x : x));\n const years = rows.map(d => d.year);\n const sos = rows.map(d => d.meanSOS);\n const lgp = rows.map(d => d.meanLGP);\n\n const sosStats = linearTrendWithP(years, sos);\n const lgpStats = linearTrendWithP(years, lgp);\n\n return {\n county,\n season_num,\n sosTrend: sosStats.slope,\n sosP: sosStats.p,\n lgpTrend: lgpStats.slope,\n lgpP: lgpStats.p\n };\n});\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":4478,"value":" significantCountiesSeasonSOS = countySeasonTrends.filter(d => d.season_num === 1 && d.sosTrend > 1 ).map(d => d.county);","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":5070,"value":"## Section Q3,4,5: Hazard x exposure study","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":4332,"value":"function computeTrend(yearly, metric) {\n const x = yearly.map(d => d.year);\n const y = yearly.map(d => d[metric]);\n const n = x.length;\n const xMean = d3.mean(x);\n const yMean = d3.mean(y);\n\n let num = 0, den = 0;\n for (let i = 0; i < n; i++) {\n num += (x[i] - xMean) * (y[i] - yMean);\n den += (x[i] - xMean) ** 2;\n }\n const slope = den === 0 ? 0 : num / den;\n const intercept = yMean - slope * xMean;\n return { slope, intercept };\n}\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":5717,"value":"$53","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":5749,"value":"$54","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":5744,"value":"kenyaExposure = await exposure_db.query(`\n SELECT *\n FROM exposure\n WHERE admin0_name = 'Kenya'\n AND admin1_name IS NOT null\n`);\n// await exposure_db.query(`\n// SELECT admin1_name, crop, SUM(\"value\") AS value\n// FROM crop_vop\n// WHERE admin0_name = 'Kenya'\n// AND admin1_name IS NOT null\n// AND admin2_name IS NOT null\n// AND value IS NOT NULL\n// AND value BETWEEN -1e12 AND 1e12\n// GROUP BY admin1_name, crop\n// ORDER BY admin1_name\n// `);","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":4371,"value":"// 1️⃣ Get rainfall data\nkenyaPTOTFiltered = await hazard_db.query(`\n SELECT admin1_name, season, year, mean AS seasonal_rain\n FROM hazard_monthly_historic\n WHERE admin0_name = 'Kenya'\n AND hazard IN ('PTOT')\n AND admin1_name IS NOT null\n`)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":1638,"value":"// 1️⃣ Get rainfall data\nkenyaRainFiltered = await hazard_db.query(`\n SELECT admin1_name, season, year, mean AS seasonal_rain\n FROM hazard_monthly_historic\n WHERE admin0_name = 'Kenya'\n AND hazard IN ('NDWL0')\n AND admin1_name IS NOT null\n`)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":2129,"value":"// Get historical seasonal NDD for Kenya\nkenyaNDWSFiltered = await hazard_db.query(`\n SELECT admin1_name, season, year, mean AS seasonal_ndws\n FROM hazard_monthly_historic\n WHERE admin0_name = 'Kenya'\n AND hazard = 'NDWS'\n AND admin1_name IS NOT null\n`);\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":2272,"value":"// Get historical seasonal NDD for Kenya\nkenyatempFiltered = await hazard_db.query(`\n SELECT admin1_name, season, year, mean AS seasonal_temp, \n FROM hazard_monthly_historic\n WHERE admin0_name = 'Kenya'\n AND hazard = 'THI-max'\n AND admin1_name IS NOT null\n`);\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":3368,"value":"kenyatempmaxFiltered = await hazard_db.query(`\n SELECT admin1_name, season, year, mean AS seasonal_tempmax\n FROM hazard_monthly_historic\n WHERE admin0_name = 'Kenya'\n AND hazard IN ('NTx35')\n AND admin1_name IS NOT null\n`)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":3356,"value":"\n// kenyaCropsAll = await exposure_db.query(`\n// SELECT admin1_name, admin2_name, crop, \"value\"\n// FROM crop_vop\n// WHERE admin0_name = 'Kenya'\n// AND admin1_name IS NOT null\n\n// `)//.filter(r => r.admin1_name !== null);\n\nkenyaCropsAll = await exposure_db.query(`\n SELECT admin1_name, crop, SUM(\"value\") AS value\n FROM crop_vop\n WHERE admin0_name = 'Kenya'\n AND admin1_name IS NOT null\n AND admin2_name IS NOT null\n AND value IS NOT NULL\n AND value BETWEEN -1e12 AND 1e12\n GROUP BY admin1_name, crop\n ORDER BY admin1_name\n`);\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":3359,"value":"// kenyaLivestockAll = await exposure_db.query(`\n// SELECT admin1_name, admin2_name, crop, \"value\"\n// FROM livestock_vop\n// WHERE admin0_name = 'Kenya'\n// AND admin1_name IS NOT null\n// AND admin2_name IS NOT null\n// `)//.filter(r => r.admin1_name !== null);\n kenyaLivestockAll = await exposure_db.query(`\n SELECT admin1_name, crop, SUM(\"value\") AS value\n FROM livestock_vop\n WHERE admin0_name = 'Kenya'\n AND admin1_name IS NOT null\n AND admin2_name IS NOT null\n AND value IS NOT NULL\n AND value BETWEEN -1e12 AND 1e12\n GROUP BY admin1_name, crop\n ORDER BY admin1_name\n`);","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":3667,"value":"$55","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":4840,"value":"hazard_historical_selected_dict = ({\n \"NDWL0\": { name: \"Number of Waterlogging Days\", thresholds: [5, 10, 20], colours: [\"#edf8e9\", \"#bae4b3\", \"#74c476\", \"#31a354\", \"#006d2c\"] },\n \"NDWS\": { name: \"Number of Days of Water Stress\", thresholds: [5, 10, 20], colours: [\"#fff5eb\", \"#fd8d3c\", \"#f16913\", \"#d94801\", \"#7f2704\"] },\n \"NTx35\": { name: \"Number of Heat Stress Days T>35 C\", thresholds: [5, 10, 20], colours: [\"#feedde\", \"#fdbe85\", \"#fd8d3c\", \"#e6550d\", \"#a63603\"] },\n \"THI-max\": { name: \"Thermal Humidity Index\", thresholds: [70, 80, 90], colours: [\"#f2f0f7\", \"#cbc9e2\", \"#9e9ac8\", \"#756bb1\", \"#54278f\"] },\n})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":4036,"value":"$56","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":3699,"value":"$57","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":4210,"value":" cropExposure = await exposure_db.query(`\n SELECT admin1_name, ROUND(SUM(TRY_CAST(value AS DOUBLE)/1e6),2) AS exposure_million\n FROM crop_vop\n WHERE admin0_name='Kenya' AND admin1_name IS NOT NULL\n AND TRY_CAST(value AS DOUBLE) IS NOT NULL\n AND TRY_CAST(value AS DOUBLE) BETWEEN -1e12 AND 1e12\n GROUP BY admin1_name\n `); \n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":4214,"value":"livestockExposure = await exposure_db.query(`\n SELECT admin1_name, ROUND(SUM(TRY_CAST(value AS DOUBLE)/1e6),2) AS exposure_million\n FROM livestock_vop\n WHERE admin0_name='Kenya' AND admin1_name IS NOT NULL\n AND TRY_CAST(value AS DOUBLE) IS NOT NULL\n AND TRY_CAST(value AS DOUBLE) BETWEEN -1e12 AND 1e12\n GROUP BY admin1_name\n `);","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":4191,"value":"populationExposure = await exposure_db.query(`\n SELECT admin1_name, ROUND(SUM(TRY_CAST(value AS DOUBLE)/1e6),2) AS exposure_million\n FROM population\n WHERE admin0_name='Kenya'\n AND type='rural'\n AND admin1_name IS NOT NULL\n AND TRY_CAST(value AS DOUBLE) IS NOT NULL\n AND TRY_CAST(value AS DOUBLE) BETWEEN -1e12 AND 1e12\n GROUP BY admin1_name\n `);","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":4274,"value":"compoundedValueslgp = countySeasonTrends.filter(d => d.season_num === 1).map(ct => {\n const cname = ct.county;\n const crop = cropExposure.find(d => d.admin1_name === cname)?.exposure_million ?? 0;\n const livestock = livestockExposure.find(d => d.admin1_name === cname)?.exposure_million ?? 0;\n const population = populationExposure.find(d => d.admin1_name === cname)?.exposure_million ?? 0;\n\n return {\n county: cname,\n compoundedCrop: (ct.lgpTrend) * crop,\n compoundedLivestock: (ct.lgpTrend) * livestock,\n compoundedPopulation: (ct.lgpTrend) * population\n };\n});\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":4290,"value":"compoundedValuessos = countySeasonTrends.filter(d => d.season_num === 1).map(ct => {\n const cname = ct.county;\n const crop = cropExposure.find(d => d.admin1_name === cname)?.exposure_million ?? 0;\n const livestock = livestockExposure.find(d => d.admin1_name === cname)?.exposure_million ?? 0;\n const population = populationExposure.find(d => d.admin1_name === cname)?.exposure_million ?? 0;\n\n return {\n county: cname,\n compoundedCrop: (ct.sosTrend) * crop,\n compoundedLivestock: (ct.sosTrend) * livestock,\n compoundedPopulation: (ct.sosTrend) * population\n };\n});\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":2800,"value":"// kenyaPopulation = await exposure_db.query(`\n// SELECT admin1_name, admin2_name, type, \"value\"\n// FROM population\n// WHERE admin0_name = 'Kenya'\n// AND admin1_name IS NOT null\n// `)\n\nkenyaPopulation = await exposure_db.query(`\n SELECT admin1_name, type, SUM(\"value\") AS value\n FROM population\n WHERE admin0_name = 'Kenya'\n AND admin1_name IS NOT null\n AND admin2_name IS NOT null\n AND value IS NOT NULL\n AND value BETWEEN -1e12 AND 1e12\n GROUP BY admin1_name, type\n ORDER BY admin1_name\n`);\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":39,"value":"### Default Adaptation Atlas Styling\nThis can be found [here](https://observablehq.com/d/8a953dc1cde63a86) to be modified and adjusted","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":10,"value":"import {atlas_stylesheet} from \"8a953dc1cde63a86\"","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":36,"value":"atlas_stylesheet","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":74,"value":"### Adaptation Atlas Boundaries","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":81,"value":"admin_boundaries = {\n let a0_topoj = await FileAttachment(\"atlas_gaul24_a0_africa_simple-vlowres@1.topojson\").json();\n let a1_topoj = await FileAttachment(\"atlas_gaul24_a1_africa_simple-vlowres.topojson\").json();\n let a2_topoj = await FileAttachment(\"atlas_gaul24_a2_africa_simple-lowres.topojson\").json();\n\n let a0_geojson = topojson.feature(\n a0_topoj,\n a0_topoj.objects[\"atlas_gaul24_a0_africa\"]\n );\n let a1_geojson = topojson.feature(\n a1_topoj,\n a1_topoj.objects[\"atlas_gaul24_a1_africa\"]\n );\n let a2_geojson = topojson.feature(\n a2_topoj,\n a2_topoj.objects[\"atlas_gaul24_a2_africa\"]\n );\n\n return {\n adm0: a0_geojson,\n adm1: a1_geojson,\n adm2: a2_geojson,\n }\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":232,"value":"watershed_boundaries = {\n let ws_topoj = await FileAttachment(\"africa_wmobb_basins.topojson\").json()\n let ws_geoj = topojson.feature(\n ws_topoj,\n ws_topoj.objects[\"africa_wmobb_basins\"]\n );\n return {\n watershed: ws_geoj\n }\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":337,"value":"kenya_boundaries = admin_boundaries.adm1.features.filter(\n d => d.properties.iso3 === \"KEN\"\n)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":473,"value":"function buildKenyaGeojson(hazard_data) {\n const features = kenya_boundaries.map(f => JSON.parse(JSON.stringify(f)));\n\n features.forEach(f => {\n const code = f.properties.gaul1_code;\n let match = null;\n\n if (code != null) {\n match = hazard_data.find(r => String(r.gaul1_code) === String(code));\n }\n\n if (!match) {\n const name = (f.properties.admin1_name || \"\").trim().toLowerCase();\n match = hazard_data.find(r => (r.admin1_name || \"\").trim().toLowerCase() === name);\n }\n\n f.properties.hazard_value = match ? match.value : 0; // use 0 if not found\n });\n\n return { type: \"FeatureCollection\", features };\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":5,"value":"$58","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"}],"files":"$W59","stdlib":"1"}} 31:"$30:body"