2c:null 2d:["$","title",null,{"children":"Irrigation Efficiency Paper | Karthik Ramaswamy | Observable"}] 32:T55f,(() => { const w = Math.min(width, 1000); const plot = Plot.plot({ width: w, height: Math.round(w * 0.62), margin: 0, inset: 4, projection: maps.projection, style: theme.style, color: {type: "linear", domain: [0.5, 0.9], interpolate: d3.interpolateRgbBasis(theme.seq), legend: true, label: "Mean application efficiency (area-weighted)", tickFormat: ".2f", width: 320}, marks: [ ...maps.base(), Plot.dot(data.grid, {x: "lon", y: "lat", r: maps.cellR(w), symbol: "square", fill: "eff", channels: {Fields: "n", "Irrigated field area (acres)": "acres"}, tip: {format: {...maps.noXY, fill: ".3f", Fields: ",d", "Irrigated field area (acres)": ",d"}}}), Plot.geo(conus.borders, {stroke: "white", strokeOpacity: 0.7, strokeWidth: 0.6}) ] }); return figure("Figure 1. Modeled mean field-level irrigation application efficiency", "Probability-weighted field estimates (CSB field-analysis units, annual crop, irrigation persistence, irrigation-system probabilities and crop × system efficiency coefficients), shown here as area-weighted means in 10 km equal-area cells. On-field efficiency only; conveyance losses are not included. High efficiencies cluster in sprinkler-dominated regions such as the High Plains; lower efficiencies in surface-irrigated regions such as the Mississippi Delta.", plot); })()33:T707,(() => { const total = d3.sum(data.hist, (d) => d.fields); const rows = data.hist.map((d) => ({...d, share: d.fields / total})); const above = (t) => d3.sum(rows.filter((d) => d.x0 >= t), (d) => d.share); const pct = d3.format(".1%"); const ymax = d3.max(rows, (d) => d.share); const plot = Plot.plot({ width: Math.min(width, 900), height: 360, marginLeft: 52, marginTop: 28, marginBottom: 46, style: theme.style, x: {domain: [0.49, 0.9], label: "Modeled application efficiency", tickFormat: ".2f", ticks: 9, labelAnchor: "center", labelOffset: 38}, y: {label: "Share of fields (%)", percent: true, grid: true, nice: true}, color: {...theme.classes, legend: true}, marks: [ Plot.rectY(rows, {x1: "x0", x2: "x1", y: "share", fill: "class", insetLeft: 0.75, insetRight: 0.75, channels: {Fields: "fields", Bin: (d) => `${d.x0.toFixed(2)}–${d.x1.toFixed(2)}`}, tip: {format: {x1: false, x2: false, y: (v) => v.toFixed(2) + "%", Fields: ",d", fill: true}}}), Plot.ruleY([0], {stroke: theme.axis}), Plot.ruleX([0.75, 0.85], {stroke: theme.ink2, strokeDasharray: "4 3"}), Plot.text([ {x: 0.75, label: `${pct(above(0.75))} of fields > 0.75`}, {x: 0.85, label: `${pct(above(0.85))} > 0.85`} ], {x: "x", y: ymax * 1.06, text: "label", textAnchor: "end", dx: -6, fill: theme.ink, fontSize: 12}) ] }); return figure("Figure 2. Distribution of irrigation efficiency across U.S. irrigated fields", `${pct(above(0.75))} of fields exceed 0.75 efficiency, including ${pct(above(0.85))} in the very high class (>0.85); ${pct(1 - above(0.75))} are at or below 0.75. Bars are 0.01-wide, right-closed bins of field efficiency (n = ${d3.format(",")(total)} fields), coloured by efficiency class.`, plot); })()34:T400,(() => { const w = Math.min(width, 1000); const plot = Plot.plot({ width: w, height: Math.round(w * 0.62), margin: 0, inset: 4, projection: maps.projection, style: theme.style, color: {...theme.system, legend: true}, marks: [ ...maps.base(), Plot.dot(data.grid, {x: "lon", y: "lat", r: maps.cellR(w), symbol: "square", fill: "dominant", channels: {"Surface share": "sur", "Sprinkler share": "spr", "Drip share": "drp", Fields: "n"}, tip: {format: {...maps.noXY, fill: true, "Surface share": ".0%", "Sprinkler share": ".0%", "Drip share": ".0%", Fields: ",d"}}}), Plot.geo(conus.borders, {stroke: "white", strokeOpacity: 0.7, strokeWidth: 0.6}) ] }); return figure("Figure 3. Dominant irrigation system type across CONUS", "System with the largest share of probability-weighted irrigated field area in each 10 km equal-area cell (shares use the adjusted field-level system probabilities). Hover a cell for the surface, sprinkler and drip shares.", plot); })()35:T65b,(() => { const fmt = d3.format(","); const plot = Plot.plot({ width: Math.min(width, 860), height: 240, marginLeft: 80, marginTop: 16, marginRight: 200, marginBottom: 46, style: theme.style, x: {domain: [0.45, 0.92], label: "Modeled application efficiency", grid: true, tickFormat: ".2f", labelAnchor: "center", labelOffset: 38}, y: {domain: theme.system.domain, label: null, padding: 0.35}, color: theme.system, marks: [ Plot.ruleY(data.systems, {y: "system", x1: "p05", x2: "p95", stroke: "system", strokeWidth: 2}), Plot.barX(data.systems, {y: "system", x1: "p25", x2: "p75", fill: "system", fillOpacity: 0.3, rx: 4}), Plot.tickX(data.systems, {y: "system", x: "p50", stroke: "system", strokeWidth: 2.5}), Plot.dot(data.systems, {y: "system", x: "mean", r: 6, fill: theme.surface, stroke: "system", strokeWidth: 2.5, channels: {Fields: "fields", Median: "p50", "5th pct": "p05", "25th pct": "p25", "75th pct": "p75", "95th pct": "p95"}, tip: {format: {y: true, x: ".3f", stroke: false, Fields: ",d", Median: ".3f", "5th pct": ".3f", "25th pct": ".3f", "75th pct": ".3f", "95th pct": ".3f"}}}), Plot.text(data.systems, {y: "system", x: "p95", dx: 10, textAnchor: "start", fill: theme.ink2, text: (d) => `mean ${d.mean.toFixed(3)} · ${fmt(d.fields)} fields`}) ] }); return figure("Figure 4. Mean irrigation efficiency by dominant irrigation system", "Open circle: mean field efficiency; vertical tick: median; shaded bar: interquartile range; line: 5th–95th percentile, by the field's dominant (highest-probability) system.", plot); })()36:T8ce,(() => { const order = ["<20 acres", "20–80 acres", "≥80 acres"]; const left = Plot.plot({ width: 360, height: 226, marginLeft: 86, marginRight: 50, marginTop: 30, marginBottom: 46, style: theme.style, x: {domain: [0.76, 0.83], label: "Mean modeled efficiency", grid: true, tickFormat: ".2f", ticks: 4, labelAnchor: "center", labelOffset: 38}, y: {domain: order, label: null, padding: 0.4}, marks: [ Plot.ruleY(data.fieldSize, {y: "size_class", x1: 0.76, x2: "mean_eff", stroke: theme.axis}), Plot.dot(data.fieldSize, {y: "size_class", x: "mean_eff", r: 6, fill: "#2a78d6", channels: {Fields: "fields"}, tip: {format: {x: ".4f", Fields: ",d"}}}), Plot.text(data.fieldSize, {y: "size_class", x: "mean_eff", text: (d) => d.mean_eff.toFixed(3), dx: 12, textAnchor: "start", fill: theme.ink2}) ] }); const long = data.fieldSize.flatMap((d) => theme.system.domain.map((s) => ({size_class: d.size_class, system: s, share: d[s]}))); const right = Plot.plot({ width: 440, height: 226, marginLeft: 86, marginRight: 16, marginTop: 30, marginBottom: 46, style: theme.style, x: {label: "Mean adjusted system probability (%)", percent: true, domain: [0, 100], labelAnchor: "center", labelOffset: 38}, y: {domain: order, label: null, padding: 0.4}, color: {...theme.system, legend: true}, marks: [ Plot.barX(long, Plot.stackX({y: "size_class", x: "share", fill: "system", order: theme.system.domain, insetLeft: 1, insetRight: 1, tip: {format: {x: (v) => v.toFixed(1) + "%", y: true, fill: true}}})), Plot.text(long.filter((d) => d.share >= 0.1), Plot.stackX({y: "size_class", x: "share", z: "system", order: theme.system.domain, text: (d) => `${Math.round(d.share * 100)}%`, fill: "white", fontWeight: 600})) ] }); return figure("Figure 5. Mean modeled efficiency and irrigation-system composition by field-size class", "Left: field-count-weighted mean modeled efficiency by field-size class. Right: mean adjusted probability shares of surface, sprinkler and drip irrigation within each class (<20, 20–80 and ≥80 acres).", html`

${left}${right}

`); })()37:T929,(() => { const short = new Map([ ["Corn for grain or seed", "Corn"], ["Alfalfa and alfalfa mixtures (dry hay, greenchop, and silage)", "Alfalfa"], ["Soybeans for beans", "Soybeans"], ["All Cotton", "Cotton"], ["Land in orchards", "Orchards"], ["Wheat for grain or seed", "Wheat"], ["All other hay (dry hay, greenchop, and silage)", "Other hay"], ["All other crops", "All other crops"], ["Pastureland, all types", "Pastureland"], ["Rice", "Rice"], ["Other small grains (barley, oats, rye, etc.)", "Other small grains"], ["Sorghum for grain or seed", "Sorghum"], ["All other crops (remaining)", "Remaining crop groups"]]); const mean = new Map(data.cropMeans.map((d) => [d.crop, d])); const label = (c) => `${short.get(c) ?? c} (${mean.get(c).mean.toFixed(2)})`; const totals = d3.rollup(data.cropClasses, (v) => d3.sum(v, (d) => d.fields), (d) => d.crop); const rows = data.cropClasses.map((d) => ({...d, group: label(d.crop), share: d.fields / totals.get(d.crop)})); const yorder = d3.sort(data.cropMeans, (d) => -d.mean).map((d) => label(d.crop)); const plot = Plot.plot({ width: Math.min(width, 900), height: 26 * yorder.length + 86, marginLeft: 170, marginRight: 16, marginTop: 30, marginBottom: 46, style: theme.style, x: {label: "Share of irrigated fields (%)", percent: true, domain: [0, 100], labelAnchor: "center", labelOffset: 38}, y: {domain: yorder, label: null, padding: 0.18}, color: {...theme.classes, legend: true}, marks: [ Plot.barX(rows, Plot.stackX({y: "group", x: "share", fill: "class", order: theme.classes.domain, insetLeft: 1, insetRight: 1, channels: {Fields: "fields"}, tip: {format: {x: (v) => v.toFixed(1) + "%", y: true, fill: true, Fields: ",d"}}})), Plot.text(rows.filter((d) => d.share >= 0.08), Plot.stackX({y: "group", x: "share", z: "class", order: theme.classes.domain, text: (d) => `${Math.round(d.share * 100)}%`, fill: (d) => d.class.startsWith("Low") ? theme.ink : "white", fontSize: 11})) ] }); return figure("Figure 6. Efficiency-class distribution by crop group", "Share of irrigated fields in each modeled efficiency class, by crop group (the twelve largest groups by field count plus the remaining groups combined). Groups are ordered by mean modeled efficiency, shown in parentheses.", plot); })()38:Tbc6,(() => { const lo = 0.2, hi = 1.8; const rows = data.metered.map((d) => ({...d, x: Math.max(lo, Math.min(hi, d.eta_obs)), clipped: d.eta_obs > hi || d.eta_obs < lo})); const stats = d3.rollups(data.metered, (v) => { const e = v.map((d) => d.eta_model - d.eta_obs); const mx = d3.mean(v, (d) => d.eta_model), my = d3.mean(v, (d) => d.eta_obs); const cov = d3.mean(v, (d) => (d.eta_model - mx) * (d.eta_obs - my)); const vx = d3.mean(v, (d) => (d.eta_model - mx) ** 2), vy = d3.mean(v, (d) => (d.eta_obs - my) ** 2); return {n: v.length, units: new Set(v.map((d) => d.unit_id)).size, bias: d3.mean(e), mae: d3.mean(e, Math.abs), rmse: Math.sqrt(d3.mean(e, (x) => x * x)), ccc: (2 * cov) / (vx + vy + (mx - my) ** 2), clipped: v.filter((d) => d.eta_obs > hi || d.eta_obs < lo).length}; }, (d) => d.basin).map(([basin, s]) => ({basin, ...s})); const f3 = (v) => (v < 0 ? "−" : "+") + Math.abs(v).toFixed(3); const plot = Plot.plot({ width: Math.min(width, 900), height: 450, marginLeft: 62, marginBottom: 48, marginTop: 30, style: theme.style, fx: {label: null, domain: theme.basin.domain, padding: 0.12}, x: {domain: [lo, hi], label: "Observed efficiency (Net ET / metered pumping)", ticks: [0.4, 0.8, 1.2, 1.6], grid: true, tickFormat: ".1f", labelAnchor: "center", labelOffset: 38}, y: {domain: [lo, hi], label: "Field-model application efficiency", ticks: [0.4, 0.8, 1.2, 1.6], grid: true, tickFormat: ".1f", labelAnchor: "center", labelOffset: 46}, color: theme.basin, marks: [ Plot.line([[lo, lo], [hi, hi]], {stroke: theme.ink2, strokeDasharray: "5 4"}), Plot.text(["1:1"], {x: 1.66, y: 1.74, text: (d) => d, fill: theme.ink2}), Plot.dot(rows, {fx: "basin", x: "x", y: "eta_model", fill: "basin", fillOpacity: (d) => (d.clipped ? 0.95 : 0.45), stroke: theme.surface, strokeWidth: 1, r: 4.5, channels: {Unit: "unit_id", Year: "year", "Observed efficiency": "eta_obs", "CSB coverage": "coverage"}, tip: {format: {x: false, fx: false, fill: false, fillOpacity: false, y: ".3f", Year: "d", "Observed efficiency": ".3f", "CSB coverage": ".0%"}}}), Plot.text(stats, {fx: "basin", frameAnchor: "bottom-right", dx: -8, dy: -8, lineHeight: 1.35, fill: theme.ink, text: (s) => [`n = ${s.n} unit-years (${s.units} units)`, `bias = ${f3(s.bias)}`, `MAE = ${s.mae.toFixed(3)}`, `RMSE = ${s.rmse.toFixed(3)}`, `CCC = ${s.ccc.toFixed(2)}`, ...(s.clipped ? [`${s.clipped} beyond axis, drawn at edge`] : [])].join("\n")}) ] }); return figure("Figure 7. Field-model versus metered-pumping application efficiency, 2018–2022", "Points are quality-controlled, groundwater-only unit-years in Diamond Valley, Nevada, and Harney Basin, Oregon, with at least 70% coverage by analytical CSB units; observed efficiency is Net ET divided by metered pumping (Ott et al., 2024). Darker points lie beyond the axis range and are drawn at its edge. The dashed line is 1:1.", plot); })()39:Ta06,(() => { const lo = 0.45, hi = 0.95; const m = data.huc12; const rank = (vals) => { // average ranks (ties share their mean rank), as in pandas const idx = d3.range(vals.length).sort((a, b) => vals[a] - vals[b]); const r = new Array(vals.length); for (let i = 0; i < idx.length;) { let j = i; while (j + 1 < idx.length && vals[idx[j + 1]] === vals[idx[i]]) j++; for (let k = i; k <= j; k++) r[idx[k]] = (i + j) / 2 + 1; i = j + 1; } return r; }; const e = m.map((d) => d.eta_model - d.eta_usgs); const ra = rank(m.map((d) => d.eta_model)), rb = rank(m.map((d) => d.eta_usgs)); const ma = d3.mean(ra), mb = d3.mean(rb); const spearman = d3.sum(ra, (v, i) => (v - ma) * (rb[i] - mb)) / Math.sqrt(d3.sum(ra, (v) => (v - ma) ** 2) * d3.sum(rb, (v) => (v - mb) ** 2)); const thr = d3.range(lo, hi + 1e-9, 0.01); const plot = Plot.plot({ width: Math.min(width, 620), height: 560, marginLeft: 62, marginBottom: 48, marginTop: 12, marginRight: 16, style: theme.style, x: {domain: [lo, hi], label: "USGS system efficiency (Haynes et al., 2024)", grid: true, ticks: [0.5, 0.6, 0.7, 0.8, 0.9], tickFormat: ".1f", labelAnchor: "center", labelOffset: 38}, y: {domain: [lo, hi], label: "Field-model application efficiency", grid: true, ticks: [0.5, 0.6, 0.7, 0.8, 0.9], tickFormat: ".1f", labelAnchor: "center", labelOffset: 46}, color: {type: "log", interpolate: d3.interpolateRgbBasis(theme.seq), legend: true, label: "HUC12s per cell (log)", tickFormat: ",~d"}, marks: [ Plot.rect(m, Plot.bin({fill: "count"}, {x: {value: "eta_usgs", thresholds: thr}, y: {value: "eta_model", thresholds: thr}, inset: 0.4, tip: true})), Plot.line([[lo, lo], [hi, hi]], {stroke: theme.ink2, strokeDasharray: "5 4"}), Plot.text([{}], {frameAnchor: "bottom-right", dx: -10, dy: -10, lineHeight: 1.35, fill: theme.ink, text: () => [`n = ${d3.format(",")(m.length)} HUC12s`, `bias = ${d3.format("+.3f")(d3.mean(e))}`, `MAE = ${d3.mean(e, Math.abs).toFixed(3)}`, `RMSE = ${Math.sqrt(d3.mean(e, (v) => v * v)).toFixed(3)}`, `Spearman ρ = ${spearman.toFixed(2)}`].join("\n")}) ] }); return figure("Figure 8. HUC12 mean application efficiency, 2018–2020: field model versus USGS", `Field-model mean application efficiency against USGS system efficiency (Haynes et al., 2024) for ${d3.format(",")(m.length)} HUC12s with at least 10 ha of analytical-sample irrigated area. Cell shading gives the number of HUC12s (log scale); the dashed line is 1:1.`, plot); })()3a:T432,(() => { const w = Math.min(width, 1000); const plot = Plot.plot({ width: w, height: Math.round(w * 0.62), margin: 0, inset: 4, projection: maps.projection, style: theme.style, color: {type: "diverging", domain: [-0.15, 0.15], pivot: 0, clamp: true, interpolate: d3.interpolateRgbBasis(theme.diverging), legend: true, label: "Field model minus USGS system efficiency", tickFormat: "+.2f", width: 340}, marks: [ ...maps.base(), Plot.dot(data.huc12, {x: "lon", y: "lat", r: Math.max(1, w / 760), fill: "delta", sort: (d) => Math.abs(d.delta), channels: {HUC12: "huc12", "Field model": "eta_model", USGS: "eta_usgs"}, tip: {format: {...maps.noXY, fill: "+.3f", "Field model": ".3f", USGS: ".3f"}}}) ] }); return figure("Figure 9. HUC12 benchmark residuals, 2018–2020 mean", "Field-model application efficiency minus USGS system efficiency (Haynes et al., 2024) at HUC12 centroids. Red: higher modeled efficiency; blue: lower; colours saturate at ±0.15. Largest residuals are drawn on top.", plot); })()3b:Tbd9,(() => { const fmt = d3.format(","); const fmax = d3.max(data.bins, (d) => d.fields); const color = {type: "log", domain: [1, fmax], interpolate: d3.interpolateRgbBasis(theme.seq)}; const signed = (d) => (d > 0 ? "+" : d < 0 ? "−" : "") + Math.abs(d) + "%"; const panel = (s) => Plot.plot({ width: 320, height: 200, marginLeft: 46, marginTop: 30, marginBottom: 36, marginRight: 8, style: theme.style, x: {domain: [s.xmin, s.xmax], label: "Lift − break-even lift (m)", ticks: 5, labelAnchor: "center"}, y: {domain: [-50, 110], ticks: [-40, 0, 40, 80], tickFormat: signed, label: null, grid: true}, color, marks: [ Plot.rect(data.bins.filter((d) => d.pathway === s.pathway), {x1: "x0", x2: "x1", y1: "y0", y2: "y1", fill: "fields", inset: 0.2, channels: {"Energy change": (d) => `${Math.round(d.y0)} to ${Math.round(d.y1)}%`, "Lift − L*": (d) => `${d.x0.toFixed(1)} to ${d.x1.toFixed(1)} m`}, tip: {format: {x1: false, x2: false, y1: false, y2: false, fill: ",d"}}}), Plot.ruleY([0], {stroke: theme.ink2}), Plot.ruleX([0], {stroke: theme.ink2, strokeDasharray: "4 3"}), Plot.text([s], {frameAnchor: "top-right", dx: -4, dy: 4, lineHeight: 1.3, fontSize: 11, fill: theme.ink, stroke: theme.surface, strokeWidth: 4, text: (d) => `n = ${fmt(d.fields)}; above L* ${d.pct_above_Lstar.toFixed(0)}%\nsign agreement ${d.sign_agreement_pct.toFixed(1)}%`}), Plot.text([s], {frameAnchor: "top-left", dy: -20, dx: -40, fontSize: 12.5, fontWeight: 600, fill: theme.ink, text: (d) => d.pathway}) ] }); const key = html`

Reading the panels

x: field lift minus break-even lift L* (m)

y: change in pumping energy (%)

dashed line: lift = L*; solid line: no energy change

L* = (η₀h₁ − η₁h₀)/(η₁ − η₀)

${Plot.legend({color: {...color, label: "Fields per cell (log)"}, tickFormat: ",~d", width: 260, style: theme.style})}

`; const panels = d3.sort(data.stats, (d) => d.order).map(panel); return figure("Figure 10. Break-even lift determines the energy consequence of modernization", "Change in pumping energy against field lift minus pathway-specific break-even lift for groundwater-dominated eligible fields (groundwater share ≥ 0.95; baseline-system probability ≥ 0.5). Shading gives the number of fields per cell (log scale). Left of the dashed line, lift is below the break-even lift and conversion increases pumping energy; right of it, energy declines. Annotations give the share of fields above the break-even lift and the agreement between the break-even classification and the full model.", html`

${panels}${key}

`); })()3c:T961,(() => { const order = ["Surface → conventional sprinkler", "Surface → low-pressure sprinkler", "Surface → subsurface drip", "Conventional → low-pressure sprinkler"]; const draw = {Unresolved: 0, Other: 1, "Co-benefit": 2, Penalty: 3}; const mw = Math.max(300, Math.min(480, Math.floor((width - 34 - 20) / 2))); // two maps per row inside the figure padding const panel = (p) => { const sh = new Map(data.shares.filter((d) => d.pathway === p).map((d) => [d.class, d.pct_area])); const sub = `co-benefit ${Math.round(sh.get("Co-benefit") ?? 0)}% · penalty ${Math.round(sh.get("Penalty") ?? 0)}% · unresolved ${Math.round(sh.get("Unresolved") ?? 0)}% of eligible area`; const plot = Plot.plot({ width: mw, height: Math.round(mw * 0.62), margin: 0, inset: 4, projection: maps.projection, style: theme.style, color: theme.outcome, marks: [ ...maps.base(), Plot.dot(data.outcomes.filter((d) => d.pathway === p), {x: "lon", y: "lat", r: Math.max(0.9, mw / 380), fill: "class", sort: (d) => draw[d.class], channels: {HUC12: "huc12", "Eligible area (ha)": "eligible_ha"}, tip: {format: {...maps.noXY, fill: true, "Eligible area (ha)": ",.1f"}}}) ] }); return html`

${p}

${sub}

${plot}

`; }; const legend = Plot.legend({color: {...theme.outcome, tickFormat: (d) => ({"Co-benefit": "Co-benefit (less water, energy and carbon)", Penalty: "Less water, more energy and carbon", Other: "Other (energy-neutral or no material water change)", Unresolved: "Unresolved (<80% agreement)"})[d]}, style: theme.style}); return figure("Figure 11. Robust joint water–energy–carbon outcome of four modernization pathways by HUC12", "Each HUC12 is coloured by the outcome class with the largest eligible irrigated area; a class is robust when it occurs in at least 80% of 200 Monte Carlo realizations, conditional on the baseline irrigation system. Co-benefit: withdrawals, pumping energy and emissions all decline by more than 2%; penalty: withdrawals decline while energy and emissions increase.", legend, html`

${order.map(panel)}

`); })()3d:T5bb,const data = await (async () => { const typed = {typed: true}; const keepId = (rows, key) => rows.map((d) => { const id = d[key]; d3.autoType(d); d[key] = id; return d; }); const clean = (rows) => rows.map((d) => Object.fromEntries(Object.entries(d).map(([k, v]) => [k.replace(/^/, ""), v]))); const [grid, hist, systems, fieldSize, cropClasses, cropMeans, metered, huc12, bins, stats, outcomes, shares, table1, tableA5] = await Promise.all([ FileAttachment("fig01_03_grid10km.csv").csv(typed), FileAttachment("fig02_distribution.csv").csv(typed), FileAttachment("fig04_system.csv").csv(typed), FileAttachment("fig05_field_size.csv").csv(typed), FileAttachment("fig06_crop_classes.csv").csv(typed), FileAttachment("fig06_crop_means.csv").csv(typed), FileAttachment("fig07_metered_validation.csv").csv(typed), FileAttachment("fig08_09_huc12_benchmark.csv").csv().then((r) => keepId(r, "huc12")), FileAttachment("fig10_breakeven_bins.csv").csv(typed), FileAttachment("fig10_breakeven_stats.csv").csv(typed), FileAttachment("fig11_outcome_classes.csv").csv().then((r) => keepId(r, "huc12")), FileAttachment("fig11_class_shares.csv").csv(typed), FileAttachment("table1_pathways.csv").csv().then(clean), FileAttachment("tableA5_parameters.csv").csv().then(clean) ]); return {grid, hist, systems, fieldSize, cropClasses, cropMeans, metered, huc12, bins, stats, outcomes, shares, table1, tableA5}; })();3e:[{"id":"44d14095282960eb2e339c28a406af1e5b8731b4b07329cdf94cba44dae7abe89fba360547c29c44cc12db34d18ce2ef99d5d28f652ee3be30df18e344b7395a","href":"https://static.observableusercontent.com/files/44d14095282960eb2e339c28a406af1e5b8731b4b07329cdf94cba44dae7abe89fba360547c29c44cc12db34d18ce2ef99d5d28f652ee3be30df18e344b7395a","name":"fig01_03_grid10km.csv","type":"text/csv","lastModified":1790994293018,"size":1468552},{"id":"d50f05bfc3cd51d6b1573d8b1638d8aa9270033ec92a1b9a2e06ad95dcda376fe280d31113b577d582459475e5823b0f981634ea676834b2acde5d83f5df7d54","href":"https://static.observableusercontent.com/files/d50f05bfc3cd51d6b1573d8b1638d8aa9270033ec92a1b9a2e06ad95dcda376fe280d31113b577d582459475e5823b0f981634ea676834b2acde5d83f5df7d54","name":"fig04_system.csv","type":"text/csv","lastModified":1790994293022,"size":212},{"id":"2b0ae085c5ab9d697ad5ea7e3a1abe0927eb88b6f72a94bb574d1bea6c394dd2c7e90c851b11b424b06e1ec03514b250fbec51bcb34e3e905bb7565ce97e83d8","href":"https://static.observableusercontent.com/files/2b0ae085c5ab9d697ad5ea7e3a1abe0927eb88b6f72a94bb574d1bea6c394dd2c7e90c851b11b424b06e1ec03514b250fbec51bcb34e3e905bb7565ce97e83d8","name":"fig05_field_size.csv","type":"text/csv","lastModified":1790994293024,"size":194},{"id":"c33f8c94f972dc68d6c66fb1e934eae8edcab1a0293f986cf54b1b603ecaf322f0222fe9ac2273be40b80064cb52b9eeb088185b28a32dfcf418d5201c51fa30","href":"https://static.observableusercontent.com/files/c33f8c94f972dc68d6c66fb1e934eae8edcab1a0293f986cf54b1b603ecaf322f0222fe9ac2273be40b80064cb52b9eeb088185b28a32dfcf418d5201c51fa30","name":"fig06_crop_classes.csv","type":"text/csv","lastModified":1790994293025,"size":2711},{"id":"cac8c05c904e1ade43257e4779c9ba106b7c54b719b152cd9e3941e3a66d7c9847b36e53449a74c1dcfed5da97930211ee68d586b2dfb9510fe6e0b723cab4af","href":"https://static.observableusercontent.com/files/cac8c05c904e1ade43257e4779c9ba106b7c54b719b152cd9e3941e3a66d7c9847b36e53449a74c1dcfed5da97930211ee68d586b2dfb9510fe6e0b723cab4af","name":"fig07_metered_validation.csv","type":"text/csv","lastModified":1790994293026,"size":29023},{"id":"5934f23dc0ba80e4171d3d0d9475546d200188e09ae9cb85c1658993c334eca098fb5a42d57cec94c7d69716edbd86dd93c939fc3b83a95b046fc27fee272333","href":"https://static.observableusercontent.com/files/5934f23dc0ba80e4171d3d0d9475546d200188e09ae9cb85c1658993c334eca098fb5a42d57cec94c7d69716edbd86dd93c939fc3b83a95b046fc27fee272333","name":"fig08_09_huc12_benchmark.csv","type":"text/csv","lastModified":1790994293043,"size":1116966},{"id":"831c8d8136ee26a01c3ca1d660fe11c5ba2087bae7bf070788b102f92f0975d5390412f60b45a3e39869a9495cc91e9174bb18668ecf8c3d6416e30f14b27996","href":"https://static.observableusercontent.com/files/831c8d8136ee26a01c3ca1d660fe11c5ba2087bae7bf070788b102f92f0975d5390412f60b45a3e39869a9495cc91e9174bb18668ecf8c3d6416e30f14b27996","name":"fig10_breakeven_bins.csv","type":"text/csv","lastModified":1790994293044,"size":61035},{"id":"8a0b5b1799a1977ae5ff4441d794a486bc0d2e8dedfa9617b6a755dcfe4e04f8cc35b77548587e1624de26099f6d3ae97dea685b4001a1d963265b51a0e61c01","href":"https://static.observableusercontent.com/files/8a0b5b1799a1977ae5ff4441d794a486bc0d2e8dedfa9617b6a755dcfe4e04f8cc35b77548587e1624de26099f6d3ae97dea685b4001a1d963265b51a0e61c01","name":"fig10_breakeven_stats.csv","type":"text/csv","lastModified":1790994293045,"size":657},{"id":"7edbaae1ebba1e7a88f2aceeddfe1c833410978fdb8381ce5beed08c381e0325458fa9e6ebadd8a0cac3703de11bc676ab3f52461b880d23949efcbb70a2d437","href":"https://static.observableusercontent.com/files/7edbaae1ebba1e7a88f2aceeddfe1c833410978fdb8381ce5beed08c381e0325458fa9e6ebadd8a0cac3703de11bc676ab3f52461b880d23949efcbb70a2d437","name":"fig11_class_shares.csv","type":"text/csv","lastModified":1790994293045,"size":571},{"id":"d6ca0b425b2167d37f921514b1a53f41ef7fda6b56fc1f5b2f8107d752799cf67c488f2632dc3cc6cb51de329754e38bb2f9ce929a5d7c781b65a06138bda4a3","href":"https://static.observableusercontent.com/files/d6ca0b425b2167d37f921514b1a53f41ef7fda6b56fc1f5b2f8107d752799cf67c488f2632dc3cc6cb51de329754e38bb2f9ce929a5d7c781b65a06138bda4a3","name":"fig11_outcome_classes.csv","type":"text/csv","lastModified":1790994293072,"size":3018491},{"id":"c64aeb017eb6828c63dcd2ebae28f3a9bd23fcb56f534a1b8be98dbcea7b1a8344b8c772c5970adca0ead3df2fc1640e4d9e6052a3c81567bf3c914fde27bdd2","href":"https://static.observableusercontent.com/files/c64aeb017eb6828c63dcd2ebae28f3a9bd23fcb56f534a1b8be98dbcea7b1a8344b8c772c5970adca0ead3df2fc1640e4d9e6052a3c81567bf3c914fde27bdd2","name":"table1_pathways.csv","type":"text/csv","lastModified":1790994293072,"size":898},{"id":"3ff63f76a0c6497121c74ec09d17ce7def7774e6e5a7c1c8674c56d3dc384692eb7c9207faad4ee3dd6dfb3c7108d1747b286e3e79f377d8fdb5a1260dfad7b1","href":"https://static.observableusercontent.com/files/3ff63f76a0c6497121c74ec09d17ce7def7774e6e5a7c1c8674c56d3dc384692eb7c9207faad4ee3dd6dfb3c7108d1747b286e3e79f377d8fdb5a1260dfad7b1","name":"tableA5_parameters.csv","type":"text/csv","lastModified":1790994293073,"size":6675},{"id":"9b46d8626837511b359a0658f23853ffd9d7b8fb36b60226902cc82c59b92c82f08c285810620eed7df1ffcd21cbe022d260cc6014f48cc2b241b54aea9388d9","href":"https://static.observableusercontent.com/files/9b46d8626837511b359a0658f23853ffd9d7b8fb36b60226902cc82c59b92c82f08c285810620eed7df1ffcd21cbe022d260cc6014f48cc2b241b54aea9388d9","name":"fig06_crop_means.csv","type":"text/csv","lastModified":1790994435355,"size":557},{"id":"6ce28ebfa8c0265bbe97bb34a75812c1bbc2ae361e84c2236ea8a2654c8572e2a9aa255272aa988d63658a4d4d3d819a530eb3ca8e7f5107f16af8b4e9677107","href":"https://static.observableusercontent.com/files/6ce28ebfa8c0265bbe97bb34a75812c1bbc2ae361e84c2236ea8a2654c8572e2a9aa255272aa988d63658a4d4d3d819a530eb3ca8e7f5107f16af8b4e9677107","name":"fig02_distribution.csv","type":"text/csv","lastModified":1790994706218,"size":1431}] 30:{"id":"5c5f62570968d279","slug":"irrigation-efficiency-paper","title":"Irrigation Efficiency Paper","thumbnail":"51b81baf896ca06d945f38d0e0f84f539a6224d96d722c4de31fa8f39f6705ae","access_level":"public","create_time":"$D2026-10-03T02:08:01.168Z","update_time":"$D2026-10-03T02:39:06.643Z","view_time":null,"first_public_time":"$D2026-10-03T02:08:05.282Z","has_unpublished_edits":null,"fork":false,"stars":0,"starred":false,"owner":{"type":"team","id":"9736ccf33b4dd1b8","login":"karthikramaswamy","name":"Karthik Ramaswamy","avatar_url":null,"tier":"free_2026"},"version":123,"publish_version":123,"role":null,"authored":false,"authors":[{"type":"individual","id":"f4b5ce6952ea85e3","login":"karthikramaswamy","name":"Karthik Ramaswamy","avatar_url":null,"tier":null}],"members":[{"type":"team","id":"9736ccf33b4dd1b8","login":"karthikramaswamy","name":"Karthik Ramaswamy","avatar_url":null,"tier":"free_2026","role":"editor"}],"comments":[],"custom_thumbnail":false,"trashed":false,"lapsed":null,"subscribed":null,"forks":0,"collection":null,"type":"notebook","body":{"titling":"auto","title":"Irrigation Efficiency Paper","cells":[{"id":1,"value":"# Irrigation Efficiency Paper","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":2,"value":"md`Interactive figures (1–11) and tables (Table 1, Supplementary Table A5) for *Field-scale irrigation efficiency estimates reveal water–energy–carbon trade-offs across the United States*. Hover any mark for its values. Numbering follows the current manuscript draft.\n\nData are compact extracts attached to this notebook (*Files* pane), written by \\`Analysis/Observable/01_export_observable_data.py\\` from the analysis outputs. Maps of field-level results (Figures 1 and 3) show area-weighted means of the 1,690,462 analytical fields in 10 km equal-area cells; HUC12 maps show one point per HUC12 centroid.`","pinned":false,"hidden":false,"mode":"ts","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":9,"value":"$32","pinned":false,"hidden":false,"mode":"ts","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":10,"value":"$33","pinned":false,"hidden":false,"mode":"ts","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":11,"value":"$34","pinned":false,"hidden":false,"mode":"ts","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":12,"value":"$35","pinned":false,"hidden":false,"mode":"ts","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":13,"value":"$36","pinned":false,"hidden":false,"mode":"ts","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":14,"value":"$37","pinned":false,"hidden":false,"mode":"ts","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":15,"value":"$38","pinned":false,"hidden":false,"mode":"ts","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":16,"value":"$39","pinned":false,"hidden":false,"mode":"ts","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":17,"value":"$3a","pinned":false,"hidden":false,"mode":"ts","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":18,"value":"$3b","pinned":false,"hidden":false,"mode":"ts","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":19,"value":"$3c","pinned":false,"hidden":false,"mode":"ts","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":20,"value":"figure(\"Table 1. Expected national change in withdrawals, pumping energy and emissions by modernization pathway\",\n \"2018–2020 baseline, irrigation consumptive use held fixed. Changes are relative to the baseline system and are expectations over baseline-system identity (weighted by its adjusted probability) for fields with probability ≥ 0.5 that meet the crop and field-area criteria; negative values are decreases. Pathways are alternatives for overlapping fields and are not additive. Outcome-class shares are of eligible irrigated area for the robust class conditional on the baseline system (≥ 80% of 200 realizations). Pumping depths from Fan et al. (2013).\",\n Inputs.table(data.table1, {layout: \"auto\", rows: 8, select: false, width: {Pathway: 250}}))","pinned":false,"hidden":false,"mode":"ts","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":21,"value":"(() => {\n const rows = [];\n let group = null;\n for (const d of data.tableA5) {\n if (!d[\"Value (range)\"]) { group = d.Parameter; continue; }\n rows.push({Group: group, ...d});\n }\n return figure(\"Supplementary Table A5. Parameters of the pumping-energy, emissions and modernization-pathway models\",\n \"Central values with the minimum–maximum of the triangular distributions used in the Monte Carlo analysis in parentheses. Basis: Literature, taken or derived from the cited source; Data, field-, subregion- or State-specific input data; Assumption, set by the authors; Analysis choice, decision rule of the pathway analysis. Use the column headers to sort.\",\n Inputs.table(rows, {layout: \"auto\", rows: 45, select: false, width: {Group: 170, Parameter: 210, \"Source and notes\": 360}}));\n})()","pinned":false,"hidden":false,"mode":"ts","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":22,"value":"md`## Appendix: data and helpers\nShared colours, the figure frame, data loading (the attached CSVs) and the map projection used by the cells above.`","pinned":false,"hidden":false,"mode":"ts","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":3,"value":"const theme = ({\n ink: \"#0b0b0b\", ink2: \"#52514e\", muted: \"#898781\", grid: \"#e1e0d9\", axis: \"#c3c2b7\", surface: \"#fcfcfb\", land: \"#f3f2ee\",\n system: {domain: [\"Surface\", \"Sprinkler\", \"Drip\"], range: [\"#eb6834\", \"#2a78d6\", \"#1baf7a\"]},\n seq: [\"#cde2fb\", \"#86b6ef\", \"#3987e5\", \"#1c5cab\", \"#0d366b\"],\n classes: {domain: [\"Low (≤0.65)\", \"Moderate (0.65–0.75)\", \"High (0.75–0.85)\", \"Very high (>0.85)\"],\n range: [\"#86b6ef\", \"#3987e5\", \"#1c5cab\", \"#0d366b\"]},\n diverging: [\"#184f95\", \"#6da7ec\", \"#f0efec\", \"#ec8a89\", \"#b8322f\"],\n outcome: {domain: [\"Co-benefit\", \"Penalty\", \"Other\", \"Unresolved\"], range: [\"#2a78d6\", \"#eb6834\", \"#1baf7a\", \"#c3c2b7\"]},\n basin: {domain: [\"Diamond Valley, NV\", \"Harney Basin, OR\"], range: [\"#2a78d6\", \"#eb6834\"]},\n style: {background: \"#fcfcfb\", color: \"#0b0b0b\", fontFamily: \"system-ui, -apple-system, 'Segoe UI', sans-serif\", fontSize: \"12px\", overflow: \"visible\"}\n});","pinned":false,"hidden":false,"mode":"ts","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":4,"value":"const figure = (title, caption, ...content) => html`

${title}

\n ${content}\n

${caption}

\n`;","pinned":false,"hidden":false,"mode":"ts","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":5,"value":"$3d","pinned":false,"hidden":false,"mode":"ts","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":6,"value":"import * as topojson from \"npm:topojson-client\";","pinned":false,"hidden":false,"mode":"ts","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":7,"value":"const conus = await (async () => {\n const us = await d3.json(\"https://cdn.jsdelivr.net/npm/us-atlas@3/states-10m.json\");\n const drop = new Set([\"02\", \"15\", \"60\", \"66\", \"69\", \"72\", \"78\"]);\n const geoms = us.objects.states.geometries.filter((g) => !drop.has(g.id));\n return {\n land: topojson.merge(us, geoms),\n borders: topojson.mesh(us, {type: \"GeometryCollection\", geometries: geoms}, (a, b) => a !== b)\n };\n})();","pinned":false,"hidden":false,"mode":"ts","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":8,"value":"const maps = ({\n // EPSG:5070 (NAD83 / Conus Albers): the projection the 10 km grid was built in\n projection: {type: \"conic-equal-area\", parallels: [29.5, 45.5], rotate: [96, 0], domain: conus.land},\n base: () => [\n Plot.geo(conus.land, {fill: theme.land, stroke: theme.axis, strokeWidth: 0.7}),\n Plot.geo(conus.borders, {stroke: theme.grid, strokeWidth: 0.6})\n ],\n // radius of a square dot whose side spans one 10 km cell at map width w (CONUS spans ~4.65e6 m)\n cellR: (w) => ((w - 8) / 4.65e6) * 1e4 / Math.sqrt(Math.PI) * 1.2,\n noXY: {x: false, y: false, r: false, symbol: false}\n});","pinned":false,"hidden":false,"mode":"ts","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"}],"files":"$W3e","stdlib":"2"}} 31:"$30:body"