About this map
Coverage. Delhi and the NCR towns around it: Gurugram, Faridabad, Noida, Greater Noida and Ghaziabad. Outside Delhi, population is a 2025 estimate rather than Census counts. Colony category: Gurugram's and Faridabad's come from their 2026-27 residential circle rates, placed on MCD's A–H scale using Delhi's own category rates (theirs are newer, so they may read slightly high); Noida and Ghaziabad don't have one yet (their published rate lists are too blurry to read), so the score counts them as Delhi-average on it and marks them partial.
ilaqa rates every ~450 m square of Delhi and the NCR towns on eight factors: colony category, crowding, metro access, greenery, quiet, distance from landfills, distance from industrial areas, and Yamuna flood safety. The score combines them with weights you can change, and shows how a square compares with the rest of Delhi. It's a draft: the factors and defaults are judgement calls, not yet checked against rents.
Population density. Delhi is split into 70 areas (switch on "Census area outlines" to see them), each given exactly its 2011 Census population: the municipal (MCD) part of each assembly constituency, plus NDMC and the Cantonment. The constituency figures add up the Census 2011 populations of their municipal wards, as published by the Delhi State Election Commission (Delhi Gazette, 21 October 2022).
Within each area, people are shared out in proportion to the buildings seen from satellites: how much ground they cover (Microsoft, recent imagery) and how tall they are (EU GHS residential building volume, 2010). It can't see how crowded each home is, so a square's figure is an estimate; the area totals are counts.
This method was chosen over alternatives (hand-estimated neighbourhood densities, building footprint alone) because it agreed best with independent population maps (Meta, WorldPop) and placed the most of Delhi's slum clusters on the 2015 DUSIB list correctly.
Unauthorised colonies (layer): outlines are DDA's PM-UDAY boundary maps (2015 imagery). Colonies DDA lists without a published map, and the 69 "affluent" colonies left out of PM-UDAY (gold), are shown as dots placed by matching their names to OpenStreetMap: filled where the colony's own name was found, hollow where only its locality was. Dots are approximate.
Show on map. Besides the score and density:
- Colony category: MCD's property-tax category (A–H) of the colonies covering a square, from colony outlines (a smaller colony inside a big one like Rohini keeps its own), else nearby colonies, else estimated from colonies within 2 km. NDMC and the Cantonment aren't categorised by MCD.
- Metro access: straight-line distance to the nearest Delhi Metro station; best within 500 m, worst from 3 km.
- Greenery: tree cover within ~700 m (ESA WorldCover 2021); worst at 3%, best from 30%. Grass and scrub are left out: on the city's edge they are mostly vacant or fallow land.
- Quiet: how much of a square is within 120 m of highways, arterial roads or railways, or under IGI's approach paths (strongest at the runway, fading over 10 km).
- Away from landfills: worst within 1 km of a big landfill or fly-ash dump, clear from 3 km.
- Away from industrial areas: share of land within ~700 m mapped as industrial, worst from 50%. Industrial areas that are mostly offices (IT parks, much of Okhla) count less than ones full of factories, warehouses and workshops, judged from the businesses listed in them. It counts a third as much as landfills: "industrial" covers everything from offices to scrap yards.
- Safe from Yamuna floods: risk where ground within 2 km of the Yamuna is below its record 2023 flood level (208.66 m in Delhi; the same height above the river downstream), from the Copernicus elevation model. The Hindon at Ghaziabad and Noida is covered the same way. Monsoon waterlogging isn't covered: elevation data this coarse can't find it.
- Within 1 km: how many schools, restaurants & cafés, gyms & sports facilities, culture & clubs, and healthcare places are within 1 km, from Overture Maps places (Meta and Microsoft sources). More counts with diminishing returns, best at the level of the top 10% of squares. These measure what's there, not how good it is: there are no open ratings.
- Neighbourhood score: the average of all of these plus "less crowding" (no penalty up to 15,000/km², worst from 60,000), weighted for who it's for. Each factor is rescaled to how much it varies across Delhi, so weights are real shares: Balanced is colony category 30% (the closest proxy for overall quality), amenities 20% (the five move together, so they count once), metro 15%, trees 15%, low crowding 10%. Family tilts to schools, trees and space; Young pro to metro, food, gyms and culture; Retiree to healthcare, trees and quiet. Noise, landfills, industrial areas and Yamuna flood risk are dealbreakers: they pull a square down where present instead of being averaged, so restaurants can't make up for a landfill next door. Shown as a percentile: a score of 80 means better than 80% of populated squares. Squares with under 1,000 people/km² (0.5% of Delhi's people, mostly farmland and open land in outer Delhi) aren't scored and show lavender; plain grey map means land left empty (airport, forests, river) or outside Delhi.
The thresholds are judgement calls, and the score measures only these conditions, nothing about who lives there. Not yet included: how good schools, restaurants or gyms are (no open ratings), buses, air quality, monsoon waterlogging, safety.
Colony category (layer): MCD sorts colonies into property-tax categories from A (most valuable, e.g. Vasant Vihar) to H. Dots are 1,391 of its 2,311 colonies, placed by matching their names inside their ward; some sit on their locality rather than the colony itself. Tapping a square lists the nearest categorised colonies. Note that low density doesn't mean a good area: the sparsest colonies are as often category H as A.
Land with no residents (tap it to see which: airport, Ridge forests, Asola sanctuary, biodiversity parks, the Yamuna and a 400 m floodplain strip) is left empty. Squares only partly on such land keep people on the rest, and riverbank settlements are cut out of the floodplain strip.
Sources: Census of India 2011 (ward populations via the Delhi Gazette, 21 Oct 2022; Delhi outline); DataMeet (assembly constituency boundaries); OpenCity (NDMC and Cantonment outlines); OpenStreetMap contributors (base map, non-residential areas); Microsoft Global ML Building Footprints; EU JRC GHS-BUILT-V 2010 (building volume); DDA PM-UDAY colony boundaries (2015 imagery); MCD colony categories (MCD colony directory); ESA WorldCover 2021 (greenery); Copernicus DEM GLO-30 (elevation); Overture Maps places (amenities, CDLA-Permissive-2.0); JRC GHS-POP 2025 (NCR population); NoBroker owner listings (asking rents, for the rent and value views; not part of the score); commute times estimated from OpenStreetMap metro lines and roads at rush-hour speeds (rough, for comparing areas); UP RERA (registered housing projects in Noida, Greater Noida and Ghaziabad: status, declared completion and extensions, complaints, as listed by the regulator; placed by the regulator's map point, a matching society name in OpenStreetMap or Overture, or the sector in the address, so some positions are approximate; projects never affect scores).