Principal Web Developer at Freedom of the Press Foundation
Beantown Stomp 2023 CO₂ LevelsAssorted CO₂ ReadingsNYC SARS-CoV-2 Wastewater DataAranet4 Data UtilitiesFreedom of the Press Foundation 2022 All Staff CO₂ LevelsCO₂ Measurements for ESCape at Pinewoods Camp 2022Charting a Case Fatality Rate by WaveCharting a Moving Case Fatality RateWhich states’ worst pandemic days were this week?Covid Cases vs. Deaths and HospitalizationsU.S. Air Travel in a PandemicMassachusetts COVID DataMassachusetts Coronavirus Cases by TownCOVID-19 Infection and Fatality by RaceWhere is COVID-19 rising and falling in the U.S.?How many SARS-CoV-2 tests are we running in the U.S.?
Boston-Area SARS-CoV-2 Wastewater Levels
data = (await d3.csv(DATA_URL)).map(d => ({
northernVariant: parseVariantResults(d['Northern\rVariant\rResults']),
southernVariant: parseVariantResults(d['Southern\rVariant Results']),
...d,
date: d3.utcParse('%m/%d/%Y')(d['Sample Date']),
southern: toIntOrNull(d['Southern\r(copies/mL)']),
northern: toIntOrNull(d['Northern\r(copies/mL)']),
southernAvg: toIntOrNull(d['Southern\r7 day avg']),
northernAvg: toIntOrNull(d['Northern\r7 day avg']),
}))
.map(d => ({
...d,
southernLow: d.southern - toIntOrNull(d['Southern\rLow Confidence\rInterval']),
northernLow: d.northern - toIntOrNull(d['Northern\rLow Confidence\rInterval']),
southernHigh: d.southern + toIntOrNull(d['Southern\rHigh Confidence\rInterval']),
northernHigh: d.northern + toIntOrNull(d['Northern\rHigh Confidence\rInterval']),
}))
.map((d, i, arr) => {
const windowStart = Math.max(0, i - 7)
const windowEnd = i
const window = arr.slice(windowStart, windowEnd)
return {
...d,
southernLowAvg: d3.mean(window.filter(d => d.southernLow), d => d.southernLow),
northernLowAvg: d3.mean(window.filter(d => d.northernLow), d => d.northernLow),
southernHighAvg: d3.mean(window.filter(d => d.southernHigh), d => d.southernHigh),
northernHighAvg: d3.mean(window.filter(d => d.northernHigh), d => d.northernHigh),
}
})
// Unfold the data so each system gets its own datapoints
.map((d, i, arr) => (
[
{
system: 'northern',
date: d.date,
concentration: d.northern,
concentrationAvg: d.northernAvg,
ciHigh: d.northernHigh,
ciLow: d.northernLow,
avgHigh: d.northernHighAvg,
avgLow: d.northernLowAvg,
},
{
system: 'southern',
date: d.date,
concentration: d.southern,
concentrationAvg: d.southernAvg,
ciHigh: d.southernHigh,
ciLow: d.southernLow,
avgHigh: d.southernHighAvg,
avgLow: d.southernLowAvg,
}
]
))
.flat()
data
Type Table, then Shift-Enter. Ctrl-space for more options.
milestones = holidays.inRange(...d3.extent(data, d => d.date)).map(h => ({
...h,
date: d3.utcDay.floor(h.date), // Convert dates to UTC
}))
RECENT_DATE_RE = /Biobot Data - samples submitted through (\d{1,2}\/\d{1,2}\/\d{4})/
mwra_page = (await fetch(CORS_PROXY + 'https://www.mwra.com/biobot/biobotdata.htm')).text()
latestSiteUpdate = {
try {
return d3.utcParse('%m/%d/%Y')(mwra_page.match(RECENT_DATE_RE)[1])
} catch {
return null
}
}
SYSTEMS_BY_TOWN = new Map([
['northern', ['Wilmington', 'Reading', 'Bedford', 'Burlington', 'Woburn', 'Stoneham', 'Lexington', 'Winchester', 'Melrose', 'Arlington', 'Medford', 'Malden', 'Waltham', 'Belmont', 'Somerville', 'Everett', 'Revere', 'Chelsea', 'Watertown', 'Cambridge', 'Winthrop']],
['both', ['Newton', 'Boston', 'Brookline', 'Milton']],
['southern', ['Ashland', 'Framingham', 'Natick', 'Wellesley', 'Needham', 'Dedham', 'Quincy', 'Westwood', 'Canton', 'Randolph', 'Braintree', 'Weymouth', 'Hingham', 'Norwod', 'Canton', 'Walpole', 'Stoughton', 'Holbrook', 'Norwood']]
])
systemForTown = t => {
const system = Array.from(SYSTEMS_BY_TOWN.entries())
.find(([system, towns]) => towns.includes(t))
if (!system) return null
return system[0]
}
includedTowns = topojson.merge(
TOWNS,
TOWNS.objects.towns.geometries.filter(t => !!systemForTown(t.properties.TOWN2))
)
mapFillScale = (system, both, none) => {
const colorScale = chart.scale('color')
const color = colorScale.apply(system)
if (color) return color
if (system === 'both') return both
return none
}
colorScale = chart.scale('color')
bothTexture = textures.lines()
.size(8)
.strokeWidth(2.5)
.stroke(colorScale.range[0])
.background(colorScale.range[1])
map = {
const mapWidth = Math.min(600, width)
const map = d3.select(DOM.svg(mapWidth, mapWidth))
map.attr('style', 'max-width: 100%')
const reprojection = d3.geoIdentity().fitExtent([[20, 20], [mapWidth - 20, mapWidth - 20]],includedTowns)
const path = d3.geoPath(reprojection)
map.call(bothTexture)
const townShapes = map.append('g')
.selectAll()
.data(topojson.feature(TOWNS, TOWNS.objects.towns).features)
.join('path')
.attr('d', path)
.attr('fill', d => mapFillScale(systemForTown(d.properties.TOWN2), bothTexture.url(), '#EEE'))
.attr('stroke', '#FFF')
return map
}
latestDate = data => data.filter(d => !!d.concentrationAvg).slice(-1)[0].date
capitalize = s => s.substring(0, 1).toUpperCase() + s.substring(1)
toIntOrNull = d => d === "" ? null : +d
pluralize = (word, count, plural) => count === 1 ? word : (plural || word + 's')
parseVariantResults = str => {
if (str === '') return
const [variant, pctStr] = str.split('\r')
return { [variant]: +pctStr.substring(0, pctStr.length - 1) / 100 }
}
formatRatio = (n, formats = {}) => {
const { numberFormat, lowerFormat, higherFormat, exactString } = Object.assign({}, {
numberFormat: d3.format('d'),
higherFormat: n => `↑${n}× over`,
lowerFormat: n => `↓${n}× below`,
exactString: 'At'
}, formats)
if (n < 0) throw('Ratio cannot be smaller than 0')
if (n < 1) return lowerFormat(numberFormat(1 / n))
if (n > 1) return higherFormat(numberFormat(n))
return exactString
}
holidays = import('https://cdn.skypack.dev/@18f/us-federal-holidays@4.0.0?min')
pctThroughYear = (date) => {
const year = date.getFullYear()
const isLeapYear = (0 == year % 4) && (0 != year % 100) || (0 == year % 400)
return d3.utcDay.count(d3.utcYear.floor(date), d3.utcDay.ceil(date)) / (isLeapYear ? 366 :365)
}
dateFromPct = (pct) => {
const dayOfYear = pct * 365 // will be slightly inaccurate for leap years, fine for our purposes
return d3.utcDay.offset(d3.utcYear.floor(new Date()), dayOfYear)
}
One platform to build and deploy the best data apps
Experiment and prototype by building visualizations in live JavaScript notebooks. Collaborate with your team and decide which concepts to build out.
Use Observable Framework to build data apps locally. Use data loaders to build in any language or library, including Python, SQL, and R.
Seamlessly deploy to Observable. Test before you ship, use automatic deploy-on-commit, and ensure your projects are always up-to-date.