Bhārata Strata · Maharashtra · Ratnagiri · morphological twins
Places like cell 4920000513
The MH-36 locality cells most similar to this Ratnagiri cell across terrain, built form, land cover, climate dynamics and connectivity — ranked by a GUM-confidence-weighted (R24) distance, with the temporal block down-weighted. Each twin is a real cell with its own witnessed dossier.
Twins found
18
across 2 distinct districts
Index confidence
0.91
GUM mean confidence of the anchor (R24)
Dimensions compared
45
morphology-v2 feature dimensions
Closest twin distance
0.182
lower = more alike (weighted)
The twins, closest first
#1 · Ratnagiricell 4920000613
distance 0.182
#2 · Ratnagiricell 4920000614
distance 0.221
#3 · Ratnagiricell 4920001164
distance 0.222
#4 · Sindhudurgcell 4950003256
distance 0.224 · cross-district
#5 · Ratnagiricell 4920002763
distance 0.244
#6 · Ratnagiricell 4920002915
distance 0.249
#7 · Ratnagiricell 4920001390
distance 0.249
#8 · Ratnagiricell 4920000673
distance 0.25
#9 · Sindhudurgcell 4950003403
distance 0.259 · cross-district
#10 · Ratnagiricell 4920001347
distance 0.259
#11 · Ratnagiricell 4920002869
distance 0.26
#12 · Ratnagiricell 4920000814
distance 0.263
#13 · Ratnagiricell 4920003258
distance 0.263
#14 · Ratnagiricell 4920001119
distance 0.264
#15 · Ratnagiricell 4920001487
distance 0.265
#16 · Sindhudurgcell 4950003209
distance 0.265 · cross-district
#17 · Ratnagiricell 4920001573
distance 0.266
#18 · Ratnagiricell 4920000659
distance 0.267
witness: find_twins — morphology-twin-index-v2 (R24, GUM-confidence-weighted, magnitude-aware, temporal block down-weighted). Distance is the weighted feature distance; lower is more alike. Cross-district twins are the discriminative signal.
Re-witness these twins
curl -s https://api.gridrock.ai/mcp -H 'Content-Type: application/json' \
-H 'Accept: application/json, text/event-stream' \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"find_twins",
"arguments":{"admin_code":492,"cluster_id":4920000513}}}'Scope: the live MH-36 cube. Twins are computed from the sha-verified morphology-v2 vector; the engine supplies no goodness/suitability ranking — only morphological similarity.