Tank renovation near Khambata
tank-1 · 21.2144, 79.6752 · PIN 441905
Contributing catchment 753.56 km² · at a 3 m bund it impounds 315 thousand m³ over 25 ha · design monsoon yield 353,190 thousand m³ · soil HSG-D (clay 44%).
In plain words
The largest reservoir in the surveyable set — a tank near Khambata fed by a very large (~750 km²) catchment, reaching ~170 ha at peak on radar and drawing down by summer. It displaces no one. It sits at low confidence only on corroboration: no inventory lake falls within range (nearest ~4.5 km) and the two sensors track weakly (correlation 0.53) and differ on extent.
What to build
Ground-truth that this is a real, renovable tank; if confirmed, desilt and strengthen — no land to acquire.
Why here
The seasonal signal and scale are the biggest here and there is no displacement; the only gap is independent confirmation.
How
A field visit to confirm the tank, its bund and feeders; then desilt and strengthen.
The delta — now vs after
Confirmed, it is the largest single restoration in the programme (~2,600 thousand m³) for Khambata; the blocker is corroboration, not conflict.
What it delivers · indicative
Live capacity from the observed satellite extent (173 ha) x 1.5 m mean depth. Desilting restores the depth lost to siltation — the tank empties to near-nothing by summer today, so recovered depth means water held later and more recharge. Irrigation ~5 TCM/ha; drinking 150 lpcd. Actual yield varies with monsoon reliability, seepage and siltation.
Incremental over today · the net-new gain, from the 4-year record
≈ 2,198 thousand m³ of net-new water · 440 ha irrigable · 39,964 drinking person-years.
recovered draw-down (147 ha of 173 ha peak) x 1.5 m — the water it loses to summer today. The site already carries ≈392 thousand m³ into summer today — that is not counted; only the recovered draw-down is new.
Why this site
An existing water body — desilting and strengthening recovers lost storage at a fraction of new-build cost.
Seasonal water extent · dual-sensor
Radar (S1, 48 passes) fills the monsoon gap that cloud hides from optical (S2, 38 scenes). Radar peak vs summer low: 172.7 → 26.2 ha, a 146.5 ha annual draw-down — the storage renovation recovers.
Area-capacity (DEM, 30 m) · gross storage by bund height
Cross-witness & confidence — independent corroboration
Two sensors track together
Optical (S2) and radar (S1) rise and fall together with correlation r = 0.53 across 38 paired months. Absolute extents match 52% — a close magnitude match strengthens the read on an open water body.
Independent inventory (HydroLAKES)
No HydroLAKES polygon within 800 m (nearest 4,488 m) — likely seasonal water not in the global inventory; treat as unconfirmed.
Land acquisition (submergence)
A 3 m bund would submerge 25 ha — the land to be acquired. Cost & consent scale with this.
⚠ catchment likely includes a through-flowing nala — verify pour point
Low confidence — sensors disagree and no independent lake confirms this site. It must be ground-truthed before any commitment of funds.
Displacement witness — real building footprints
No structures in footprintIn the submergence footprint
0
a renovation floods no new land — desilting restores depth within the existing waterspread, so nothing is displaced.
Nearest structure
154 m
to the closest mapped building.
Within 500 m
19
buildings in the influence ring.
Restoring lost depth within the existing waterspread raises no new water level and acquires no new land, so no one is displaced. Verify bed encroachment on the historical spread on the ground before restoring the full level.
Overture Maps buildings 2026-05-20.0 (union of OSM / Google Open Buildings / Microsoft / Esri), Bhandara admin_code=506. A desktop displacement screen — precedes, not replaces, cadastral (7/12) verification.
Reconnaissance context & access — Ola + Google Places
Nearest habitation
Khambata
436 m away · PIN 441905
Nearest landmark
Bus Stop
Schools within 2 km
1
nearest 1,006 m
Health within 2 km
1
nearest 1,249 m
Google Places (New) searchNearby + Ola Maps Places nearbysearch · reconnaissance-grade. Independent witness to the buildings displacement layer — dense amenities near a site corroborate high displacement. Does not replace cadastral 7/12 records, geotechnical bores or a level survey.
Human pressure — land-cover change (2022 → 2026)
Stable land coverVegetation (NDVI) change
+0.041
95 ha lost · 252 ha gained over 1,613 ha land
Built-up (NDBI) change
63.3 ha
new built-up / bare (construction proxy)
Window
2022-11..2023-02 (8 scenes)
→ 2025-11..2026-02 (8 scenes)
Sentinel-2 L2A dry-season MEDIAN composites (2022-23 vs 2025-26), water-masked; NDVI+NDBI change screen — not cadastral or DEM cut/fill
DPR-grade desktop hydrology from Copernicus GLO-30 (30 m) + ISRIC SoilGrids + ESA WorldCover; water extent from Sentinel-2 L2A MNDWI (optical) and Sentinel-1 GRD VV radiometrically calibrated to σ⁰ (DN²/sigmaNought² from the product calibration LUT, GCP thin-plate-spline geocoded, open water at σ⁰ < −17 dB; thermal-noise floor not subtracted). A desktop assessment that precedes — not replaces — the statutory ground & geotechnical survey.