Public
Edited
Nov 22, 2023
Paused
9 forks
Importers
51 stars
timestamp = d3.min([
`${date.toISOString().slice(0, 11)}${d3.format("02d")(hour)}`,
new Date(+new Date() - 3600 * 1000).toISOString().slice(0, 13)
])
// Note: initially I had a unique URL (a fixed datetime), but for a live map it's better to check
// the API against the _current_ time. Once you do that, it’s a simple change to add a date picker
// and a range slider for the hour. The line below ensures that we don’t look “in the future”. And
// indeed we don’t want the current time precisely, because the API can be a few minutes late—so
// we’ll check one hour before the current time.
url = `https://www.airnowapi.org/aq/data/?startDate=${timestamp}&endDate=${timestamp}¶meters=PM25&BBOX=-130,25,-65,55&dataType=A&format=text/csv&API_KEY=${key}`
function getData(url) {
return d3
.text(url)
.then((d) =>
d3.csvParse("lat,lon,date,indicator,PM2.5,??\n" + d, d3.autoType)
);
}
air = cachedData(url)
// This function wraps the logic that fetches & parses the data with a cache, to animate the map
// a bit faster when we play with the date/time controls.
cachedData = {
const M = new Map();
return async function (url) {
if (M.has(url)) return Promise.resolve(M.get(url));
const c = await getData(url);
M.set(url, c);
return c;
};
}
// Here’s a copy, in case the API stops working
// air = FileAttachment("air.csv").csv({typed: true})
color = ({
label: "PM2.5",
type: "threshold",
domain: [50, 100, 150, 200, 300],
range: [
"rgb(156, 216, 78)", // Good
"rgb(250, 207, 57)", // Moderate
"rgb(249, 144, 73)", // Unhealthy for Sensitive Groups
"rgb(246, 94, 95)", // Unhealthy
"rgb(160, 112, 182)", // Very Unhealthy
"rgb(160, 106, 123)" // Hazardous
]
})
Plot.plot({
color,
marks: [
Plot.dot(air, {
x: "lon",
y: "lat",
fill: "PM2.5",
r: 4,
stroke: "white"
})
]
})
// Plot.dot(air, { x: "lon", y: "lat", fill: "PM2.5", fx: "date" }).plot()
// Group by date, and return the group with the largest key.
recent = d3
.greatest(
d3.group(air, (d) => d.date),
(d) => d[0]
)[1]
.filter((d) => d["PM2.5"] > 0) // remove the points with value -999
visibility().then(() => map(width))
// this is a function because we want to show the map at the top of the page.
map = (width) =>
Plot.plot({
width,
projection: "albers",
color: { ...color, legend: true },
marks: [
// this creates a path to clip the contour mark.
Plot.geo(nation, {
render: (i, s, v, d, c, next) => {
const g = next(i, s, v, d, c).children[0];
return svg`<clipPath id=nation>${g}`;
}
}),
Plot.contour(recent, {
x: "lon",
y: "lat",
fill: "PM2.5",
interpolate: "random-walk",
blur: 3,
thresholds: [0, ...color.domain],
stroke: "currentColor",
render: (i, s, v, d, c, next) => {
const g = next(i, s, v, d, c);
g.setAttribute("clip-path", "url(#nation)"); // add clip
return g;
}
}),
Plot.geo(nation, { stroke: "currentColor" }),
Plot.dot(recent, {
x: "lon",
y: "lat",
fill: "PM2.5",
stroke: "currentColor",
strokeOpacity: 0.8,
strokeWidth: 0.75,
tip: true
}),
Plot.text([timestamp], {
frameAnchor: "top-right",
fontSize: 16,
fontVariant: "tabular-nums"
})
]
})
us = FileAttachment("us-counties-10m.json").json()
nation = topojson.feature(us, us.objects.nation)
key = "62F6D106-6781-4DC5-8A63-005044B666E2" // API key
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.