H3 Tutorial: Intro to h3-js (v4)Indexing Polygons With H3Cell counts per countryH3 Adaptive GridH3 Index Bit LayoutH3 Hierarchical (Non)ContainmentH3 Deck.gl Heat MapH3 Mapbox Style Heat MapIndexing OrderArea VariationH3 Continuous ElevationH3 Indexes: Cartesian vs SphericalShape simplification with H3H3 Index Inspector"Flat" H3 Grid via Gnomonic ProjectionH3 Radius LookupH3 Tutorial: Suitability AnalysisH3 Tutorial: Using Point LayersH3 Tutorial: Rendering HexagonsH3 Tutorial: Intro to h3-js (v3)H3 Pathfinding
Surface construction with H3
function hexMapForPoints() {
// Calculate the average value for each hex associated with a point
const hexes = new Map();
for (const {lat, lng, value} of aqi) {
const h3Index = h3.geoToH3(lat, lng, h3Resolution);
let d;
if (!hexes.has(h3Index)) {
d = {sum: 0, count: 0, value: 0};
hexes.set(h3Index, d);
} else {
d = hexes.get(h3Index);
}
d.sum += value;
d.count++;
d.value = d.sum / d.count;
}
return hexes;
}
function weightedValue(h3Index, candidates, hexes) {
// calculate the weighted value of the new cell
const weights = candidates.map(c => 1 / Math.pow(h3.h3Distance(c, h3Index), 2));
const weightSum = weights.reduce((sum, weight) => sum + weight, 0);
let value = 0;
for (let i = 0; i < candidates.length; i++) {
value += hexes.get(candidates[i]).value * weights[i] / weightSum
}
return value;
}
function smoothHexes(hexes) {
for (const h3Index of hexes.keys()) {
// add surrounding rings, with less weight in each
const rings = h3.kRingDistances(h3Index, smoothingRadius);
let sum = 0;
let weightSum = 0;
for (let k = 0; k < rings.length; k++) {
const weight = 1 / ((k + 1) * (k + 1));
for (const neighbor of rings[k]) {
const record = hexes.get(neighbor);
if (record) {
sum += record.value * weight;
weightSum += weight;
}
}
}
hexes.set(h3Index, {value: sum / weightSum});
}
}
function runSmooth(hexes) {
for (let i = 0; i < smoothingPasses; i++) smoothHexes(hexes);
}
hexbins = {
const hexes = hexMapForPoints();
return geojsonForHexMap(hexes);
}
bufferedHexbins = {
const hexes = hexMapForPoints();
for (const h3Index of [...hexes.keys()]) {
// Add additional hexagons, up to the smoothing radius, with value 0
for (const neighbor of h3.kRing(h3Index, smoothingRadius)) {
if (!hexes.has(neighbor)) {
hexes.set(neighbor, {value: 0})
}
}
}
runSmooth(hexes);
return geojsonForHexMap(hexes);
}
delaunayHexagons = {
const hexes = hexMapForPoints();
// Calculate the Delaunay triangulation
const delaunay = delaunator.from(aqi, d => d.lng, d => d.lat)
// now fill in the triangles
for (let t = 0; t < delaunay.triangles.length / 3; t++) {
// get the triangle geometry
const triangle = [3 * t, 3 * t + 1, 3 * t + 2]
.map(e => aqi[delaunay.triangles[e]])
.map(({lat, lng}) => [lat, lng]);
// get the hexagons at the vertices
const triangleHexes = triangle.map(([lat, lng]) => h3.geoToH3(lat, lng, h3Resolution));
for (const h3Index of h3.polyfill(triangle, h3Resolution)) {
if (!hexes.has(h3Index)) {
const value = weightedValue(h3Index, triangleHexes, hexes);
hexes.set(h3Index, {value});
}
}
}
runSmooth(hexes);
return geojsonForHexMap(hexes);
}
ancestorHexagons = {
const hexes = hexMapForPoints();
// step up two resolutions to index
const parents = new Map();
for (const h3Index of hexes.keys()) {
const parent = h3.h3ToParent(h3Index, h3Resolution - 2);
let children = parents.get(parent);
if (!children) {
children = [];
parents.set(parent, children);
}
children.push(h3Index);
}
// expand the parents to their children
for (const parent of parents.keys()) {
for (const h3Index of h3.h3ToChildren(parent, h3Resolution)) {
if (!hexes.has(h3Index)) {
const candidates = [].concat(
...h3.kRing(parent, 1).map(parent => parents.get(parent) || [])
);
const value = weightedValue(h3Index, candidates, hexes);
hexes.set(h3Index, {value});
}
}
}
runSmooth(hexes);
return geojsonForHexMap(hexes);
}
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.