{"finding":{"findingId":"01a0df85-c90c-74b4-b254-5f7cbfcea81d","status":"published","tier":1,"rule":{"name":"forest_loss","version":"1.0","params":{"days":90,"minAlerts":1500,"minConfidence":"high","minHa":150,"minValidPct":60,"ndviDrop":0.1}},"title":"Forest loss, Peixoto de Azevedo, Mato Grosso (Brazil) — tile 2/2: 1181.92 ha in 90 d","summary":"97489 GFW integrated deforestation alerts (≥ high confidence) covering 1181.92 ha between 2026-06-28 and 2026-09-26; AOI alert density is 0.768× its 25 km neighbourhood. Awaiting optical confirmation.","geometry":{"type":"Polygon","coordinates":[[[-53.885,-10.78],[-52.61,-10.78],[-52.61,-9.61],[-53.885,-9.61],[-53.885,-10.78]]]},"bbox":[-53.885,-10.78,-52.61,-9.61],"aoi":{"id":"bra-12-92-2","name":"Peixoto de Azevedo, Mato Grosso (Brazil) — tile 2/2","tags":["forest-hotspot","discovered","bra","brazil","mato-grosso","peixoto-de-azevedo"]},"context":{"enso":{"phase":"El Niño","oni":1.8},"events":[],"baseline":{"metric":"alert ha per deg²","ringKm":25,"aoiValue":792.304,"regionalValue":1031.318,"ratio":0.768},"notes":["Living value (benefit transfer, order of magnitude): 1181.92 ha of tropical forest ≈ $7,940,139/yr alive (band $2,331,928–$30,761,832, 2020 USD), ≈ $794,013,900 over 100 yr ($342,206,903 NPV @ 2 %). Source: Costanza R. et al. et al. — a floor for one kind of value, not a price."]},"blindSpots":["Coverage is 30°N–30°S only — zero alerts outside the tropics means 'not monitored', not 'no loss'.","Cannot distinguish legal clearing, fire, storm damage and selective logging from illegal clearing; attribution needs concession/protected-area layers (M3).","Alerts lag reality by ~1–2 weeks; a burst just before the sweep is undercounted.","Regrowth and plantation harvest cycles look like loss; the baseline ring only partly compensates.","Persistent cloud can delay optical confirmation for months (SAR-only RADD alerts stay 'candidate')."],"observedAt":"2026-09-26T00:00:00Z","createdAt":"2026-09-26T20:54:22.777Z","updatedAt":"2026-09-27T01:33:00.537Z","createdBy":"system:forest_loss@1.0","confirmed":{"at":"2026-09-26T20:54:22.777Z","independence":"sensor","signal":{"id":"s2-ndvi-median-bra-12-92-2-2026-05-14..2026-09-26","kind":"scene","source":"sentinel-2-l2a","datetime":"2026-09-26T00:00:00Z","method":{"name":"eo_compare","version":"1.0","params":{"bbox":[-53.885,-10.78,-52.61,-9.61],"composite":"median","dateA":"2026-05-14","dateB":"2026-09-26","index":"NDVI","width":256}},"summary":"Median-composite NDVI changed -0.104 (2026-05-14 → 2026-09-26); valid pixels 99% / 99%. Scenes: [object Object], [object Object], [object Object], [object Object], [object Object], [object Object]","values":{"deltaNdvi":-0.1042716195887381,"validPctA":99,"validPctB":99}}},"evidence":[{"id":"gfw-integrated-bra-12-92-2-2026-06-28..2026-09-26","kind":"alert","source":"gfw-integrated-alerts","datetime":"2026-09-26T00:00:00Z","href":"https://data-api.globalforestwatch.org/dataset/gfw_integrated_alerts","method":{"name":"forest_alerts","version":"1.0","params":{"bbox":[-53.885,-10.78,-52.61,-9.61],"days":90,"minConfidence":"high"}},"summary":"97489 alerts (≥ high confidence), 1181.92 ha, 2026-06-28…2026-09-26.","values":{"alerts":97489,"ha":1181.92,"high_alerts":96217,"high_ha":1166.5,"highest_alerts":1272,"highest_ha":15.42,"living_value_100y_npv2_usd":342206903,"living_value_100y_usd":794013900,"living_value_usd_yr":7940139}},{"id":"gfw-integrated-bra-12-92-2-2026-06-29..2026-09-27","kind":"alert","source":"gfw-integrated-alerts","datetime":"2026-09-27T00:00:00Z","href":"https://data-api.globalforestwatch.org/dataset/gfw_integrated_alerts","method":{"name":"forest_alerts","version":"1.0","params":{"bbox":[-53.885,-10.78,-52.61,-9.61],"days":90,"minConfidence":"high"}},"summary":"96205 alerts (≥ high confidence), 1166.36 ha, 2026-06-29…2026-09-27.","values":{"alerts":96205,"ha":1166.36,"high_alerts":95006,"high_ha":1151.83,"highest_alerts":1199,"highest_ha":14.54,"living_value_100y_npv2_usd":337701703,"living_value_100y_usd":783560600,"living_value_usd_yr":7835606}}],"attribution":null,"narration":{"v":1,"eventId":"01a0e07e-54d3-764d-af7c-f362c9b0c5d6","findingId":"01a0df85-c90c-74b4-b254-5f7cbfcea81d","at":"2026-09-27T01:32:50.771Z","actor":"model:claude-opus-5","prev":"ec2dca76094f8bd1e7feb86e671dcde1dc74c59593c564c45758a083b275bdac","kind":"narrated","text":"Peixoto de Azevedo, Mato Grosso: 97489 forest-loss alerts over 1181.92 ha in 90 days\n\nBetween 2026-06-28 and 2026-09-26, satellite alert systems flagged 97489 high-confidence deforestation alerts across 1181.92 hectares in this tile of Peixoto de Azevedo, Mato Grosso. An independent sensor check backs it up: Sentinel-2 median-composite greenness (NDVI) fell by -0.104 between May and late September, with 99% of pixels usable in both composites, so cloud is not hiding the picture. A second alert pull a day later returned a consistent 96205 alerts over 1166.36 ha.\n\nWhat this is not, or may not be: the alerts do not say what removed the canopy. Fire, storm damage, selective logging, legally permitted clearing and plantation harvest or regrowth cycles all register the same way. The region is in an El Niño phase (ONI 1.8), and seasonal dryness alone can depress NDVI. Alert density here is 0.768x the surrounding 25 km ring — active, but below its neighbourhood.\n\nWhat would change our mind: high-resolution optical or SAR imagery showing burn scars rather than cut edges; a rebound in NDVI in the next composite, indicating seasonal browning or regrowth; or concession and protected-area layers placing the change inside permitted operations.\n\nKey numbers:\n- High-confidence alerts (90 days): 97489 [gfw-integrated-bra-12-92-2-2026-06-28..2026-09-26]\n- Hectares covered by alerts: 1181.92 [gfw-integrated-bra-12-92-2-2026-06-28..2026-09-26]\n- Alerts at highest confidence: 1272 [gfw-integrated-bra-12-92-2-2026-06-28..2026-09-26]\n- Hectares in next-day repeat pull: 1166.36 [gfw-integrated-bra-12-92-2-2026-06-29..2026-09-27]\n- Change in median NDVI (May to Sept): -0.104 [s2-ndvi-median-bra-12-92-2-2026-05-14..2026-09-26]\n- Valid pixels in later composite (%): 99 [s2-ndvi-median-bra-12-92-2-2026-05-14..2026-09-26]\n\nConfidence: medium\nCaveats:\n- Alerts cannot separate legal clearing, fire, storm damage or selective logging from unauthorised clearing; attribution needs concession and protected-area layers.\n- Alerts lag events by roughly 1-2 weeks, so a burst just before the cut-off is undercounted.\n- Regrowth and plantation harvest cycles can look like loss; the 25 km baseline ring only partly compensates.\n- The NDVI drop of -0.104 is only marginally past the rule's 0.1 threshold, and El Nino-driven dryness can contribute to seasonal browning.\n- Alert coverage is limited to 30N-30S; absence of alerts elsewhere means 'not monitored', not 'no loss'.\n- Living-value figures in the context notes are order-of-magnitude benefit transfers, not a market price, and are not asserted here.","model":{"id":"claude-opus-5","provider":"anthropic"},"promptSha256":"0d8c8167ac2b3c3b19ba5e814f8bc302a7614f97aa348e1271f27bf60591e323","transcriptSha256":"a9cabce6439b344dc6611f96fb45a83eaf15d056fab180ccc06511a2539a2f49","evidenceRefs":["gfw-integrated-bra-12-92-2-2026-06-28..2026-09-26","gfw-integrated-bra-12-92-2-2026-06-29..2026-09-27","s2-ndvi-median-bra-12-92-2-2026-05-14..2026-09-26"]},"reviews":[{"v":1,"eventId":"01a0e07e-7af4-7d05-a166-744b053173da","findingId":"01a0df85-c90c-74b4-b254-5f7cbfcea81d","at":"2026-09-27T01:33:00.532Z","actor":"model:claude-sonnet-5","prev":"29cfb864e4eb600b032160138c5868edf651e7d6d965dddcd25914c482807e59","kind":"reviewed","decision":"approve","tier":1,"note":"All quoted figures (97489 alerts, 1181.92 ha, 1272 highest-confidence alerts, next-day pull 96205/1166.36, NDVI -0.104, 99% valid pixels) match the evidence exactly.; Narration explicitly reproduces all five blind spots (attribution limits, alert lag, regrowth/harvest confusion, tropics-only coverage, cloud/optical delay) plus adds context on El Niño-driven dryness and the marginal NDVI threshold.; No natural persons are named or identifiable; only a location and satellite data are discussed.; AOI is tagged forest-hotspot/discovered, not control, so this is a genuine monitoring target rather than a calibration control area.","verdict":"publish"}],"notifications":[],"replies":[],"retracted":null,"history":[{"at":"2026-09-26T20:54:22.777Z","kind":"created","status":"candidate","actor":"system:forest_loss@1.0"},{"at":"2026-09-26T20:54:22.777Z","kind":"confirmed","status":"confirmed","actor":"system:forest_loss@1.0"},{"at":"2026-09-27T00:00:53.767Z","kind":"evidence_added","status":"confirmed","actor":"system:forest_loss@1.0"},{"at":"2026-09-27T01:32:50.771Z","kind":"narrated","status":"confirmed","actor":"model:claude-opus-5"},{"at":"2026-09-27T01:33:00.532Z","kind":"reviewed","status":"confirmed","actor":"model:claude-sonnet-5"},{"at":"2026-09-27T01:33:00.537Z","kind":"status_changed","status":"published","actor":"model:claude-sonnet-5"}],"eventCount":6,"lastEventHash":"83dfd45ddbda95c6a0f392f917911ea687927865617404df1b24816dd5537f28"},"events":[{"v":1,"eventId":"01a0df85-c90c-72c8-b254-1599777b904f","findingId":"01a0df85-c90c-74b4-b254-5f7cbfcea81d","at":"2026-09-26T20:54:22.777Z","actor":"system:forest_loss@1.0","prev":null,"kind":"created","rule":{"name":"forest_loss","version":"1.0","params":{"days":90,"minAlerts":1500,"minConfidence":"high","minHa":150,"minValidPct":60,"ndviDrop":0.1}},"title":"Forest loss, Peixoto de Azevedo, Mato Grosso (Brazil) — tile 2/2: 1181.92 ha in 90 d","summary":"97489 GFW integrated deforestation alerts (≥ high confidence) covering 1181.92 ha between 2026-06-28 and 2026-09-26; AOI alert density is 0.768× its 25 km neighbourhood. Awaiting optical confirmation.","tier":1,"geometry":{"type":"Polygon","coordinates":[[[-53.885,-10.78],[-52.61,-10.78],[-52.61,-9.61],[-53.885,-9.61],[-53.885,-10.78]]]},"bbox":[-53.885,-10.78,-52.61,-9.61],"aoi":{"id":"bra-12-92-2","name":"Peixoto de Azevedo, Mato Grosso (Brazil) — tile 2/2","tags":["forest-hotspot","discovered","bra","brazil","mato-grosso","peixoto-de-azevedo"]},"observedAt":"2026-09-26T00:00:00Z","evidence":[{"id":"gfw-integrated-bra-12-92-2-2026-06-28..2026-09-26","kind":"alert","source":"gfw-integrated-alerts","datetime":"2026-09-26T00:00:00Z","href":"https://data-api.globalforestwatch.org/dataset/gfw_integrated_alerts","method":{"name":"forest_alerts","version":"1.0","params":{"bbox":[-53.885,-10.78,-52.61,-9.61],"days":90,"minConfidence":"high"}},"summary":"97489 alerts (≥ high confidence), 1181.92 ha, 2026-06-28…2026-09-26.","values":{"alerts":97489,"ha":1181.92,"high_alerts":96217,"high_ha":1166.5,"highest_alerts":1272,"highest_ha":15.42,"living_value_100y_npv2_usd":342206903,"living_value_100y_usd":794013900,"living_value_usd_yr":7940139}}],"context":{"enso":{"phase":"El Niño","oni":1.8},"events":[],"baseline":{"metric":"alert ha per deg²","ringKm":25,"aoiValue":792.304,"regionalValue":1031.318,"ratio":0.768},"notes":["Living value (benefit transfer, order of magnitude): 1181.92 ha of tropical forest ≈ $7,940,139/yr alive (band $2,331,928–$30,761,832, 2020 USD), ≈ $794,013,900 over 100 yr ($342,206,903 NPV @ 2 %). Source: Costanza R. et al. et al. — a floor for one kind of value, not a price."]},"blindSpots":["Coverage is 30°N–30°S only — zero alerts outside the tropics means 'not monitored', not 'no loss'.","Cannot distinguish legal clearing, fire, storm damage and selective logging from illegal clearing; attribution needs concession/protected-area layers (M3).","Alerts lag reality by ~1–2 weeks; a burst just before the sweep is undercounted.","Regrowth and plantation harvest cycles look like loss; the baseline ring only partly compensates.","Persistent cloud can delay optical confirmation for months (SAR-only RADD alerts stay 'candidate')."]},{"v":1,"eventId":"01a0df85-e44e-7668-9dc9-a58eddd4b5ae","findingId":"01a0df85-c90c-74b4-b254-5f7cbfcea81d","at":"2026-09-26T20:54:22.777Z","actor":"system:forest_loss@1.0","prev":"95fd555e8536ce717fe88304190416207c33f28050fcb64db4cf4fc8aff810b8","kind":"confirmed","signal":{"id":"s2-ndvi-median-bra-12-92-2-2026-05-14..2026-09-26","kind":"scene","source":"sentinel-2-l2a","datetime":"2026-09-26T00:00:00Z","method":{"name":"eo_compare","version":"1.0","params":{"bbox":[-53.885,-10.78,-52.61,-9.61],"composite":"median","dateA":"2026-05-14","dateB":"2026-09-26","index":"NDVI","width":256}},"summary":"Median-composite NDVI changed -0.104 (2026-05-14 → 2026-09-26); valid pixels 99% / 99%. Scenes: [object Object], [object Object], [object Object], [object Object], [object Object], [object Object]","values":{"deltaNdvi":-0.1042716195887381,"validPctA":99,"validPctB":99}},"independence":"sensor"},{"v":1,"eventId":"01a0e032-47c9-7b17-8db9-ddffa4e2b92f","findingId":"01a0df85-c90c-74b4-b254-5f7cbfcea81d","at":"2026-09-27T00:00:53.767Z","actor":"system:forest_loss@1.0","prev":"3630fe3ec6db5ca8cca7dcf31e13170f789c055c273acb10bf4d9d8e225aee23","kind":"evidence_added","evidence":[{"id":"gfw-integrated-bra-12-92-2-2026-06-29..2026-09-27","kind":"alert","source":"gfw-integrated-alerts","datetime":"2026-09-27T00:00:00Z","href":"https://data-api.globalforestwatch.org/dataset/gfw_integrated_alerts","method":{"name":"forest_alerts","version":"1.0","params":{"bbox":[-53.885,-10.78,-52.61,-9.61],"days":90,"minConfidence":"high"}},"summary":"96205 alerts (≥ high confidence), 1166.36 ha, 2026-06-29…2026-09-27.","values":{"alerts":96205,"ha":1166.36,"high_alerts":95006,"high_ha":1151.83,"highest_alerts":1199,"highest_ha":14.54,"living_value_100y_npv2_usd":337701703,"living_value_100y_usd":783560600,"living_value_usd_yr":7835606}}]},{"v":1,"eventId":"01a0e07e-54d3-764d-af7c-f362c9b0c5d6","findingId":"01a0df85-c90c-74b4-b254-5f7cbfcea81d","at":"2026-09-27T01:32:50.771Z","actor":"model:claude-opus-5","prev":"ec2dca76094f8bd1e7feb86e671dcde1dc74c59593c564c45758a083b275bdac","kind":"narrated","text":"Peixoto de Azevedo, Mato Grosso: 97489 forest-loss alerts over 1181.92 ha in 90 days\n\nBetween 2026-06-28 and 2026-09-26, satellite alert systems flagged 97489 high-confidence deforestation alerts across 1181.92 hectares in this tile of Peixoto de Azevedo, Mato Grosso. An independent sensor check backs it up: Sentinel-2 median-composite greenness (NDVI) fell by -0.104 between May and late September, with 99% of pixels usable in both composites, so cloud is not hiding the picture. A second alert pull a day later returned a consistent 96205 alerts over 1166.36 ha.\n\nWhat this is not, or may not be: the alerts do not say what removed the canopy. Fire, storm damage, selective logging, legally permitted clearing and plantation harvest or regrowth cycles all register the same way. The region is in an El Niño phase (ONI 1.8), and seasonal dryness alone can depress NDVI. Alert density here is 0.768x the surrounding 25 km ring — active, but below its neighbourhood.\n\nWhat would change our mind: high-resolution optical or SAR imagery showing burn scars rather than cut edges; a rebound in NDVI in the next composite, indicating seasonal browning or regrowth; or concession and protected-area layers placing the change inside permitted operations.\n\nKey numbers:\n- High-confidence alerts (90 days): 97489 [gfw-integrated-bra-12-92-2-2026-06-28..2026-09-26]\n- Hectares covered by alerts: 1181.92 [gfw-integrated-bra-12-92-2-2026-06-28..2026-09-26]\n- Alerts at highest confidence: 1272 [gfw-integrated-bra-12-92-2-2026-06-28..2026-09-26]\n- Hectares in next-day repeat pull: 1166.36 [gfw-integrated-bra-12-92-2-2026-06-29..2026-09-27]\n- Change in median NDVI (May to Sept): -0.104 [s2-ndvi-median-bra-12-92-2-2026-05-14..2026-09-26]\n- Valid pixels in later composite (%): 99 [s2-ndvi-median-bra-12-92-2-2026-05-14..2026-09-26]\n\nConfidence: medium\nCaveats:\n- Alerts cannot separate legal clearing, fire, storm damage or selective logging from unauthorised clearing; attribution needs concession and protected-area layers.\n- Alerts lag events by roughly 1-2 weeks, so a burst just before the cut-off is undercounted.\n- Regrowth and plantation harvest cycles can look like loss; the 25 km baseline ring only partly compensates.\n- The NDVI drop of -0.104 is only marginally past the rule's 0.1 threshold, and El Nino-driven dryness can contribute to seasonal browning.\n- Alert coverage is limited to 30N-30S; absence of alerts elsewhere means 'not monitored', not 'no loss'.\n- Living-value figures in the context notes are order-of-magnitude benefit transfers, not a market price, and are not asserted here.","model":{"id":"claude-opus-5","provider":"anthropic"},"promptSha256":"0d8c8167ac2b3c3b19ba5e814f8bc302a7614f97aa348e1271f27bf60591e323","transcriptSha256":"a9cabce6439b344dc6611f96fb45a83eaf15d056fab180ccc06511a2539a2f49","evidenceRefs":["gfw-integrated-bra-12-92-2-2026-06-28..2026-09-26","gfw-integrated-bra-12-92-2-2026-06-29..2026-09-27","s2-ndvi-median-bra-12-92-2-2026-05-14..2026-09-26"]},{"v":1,"eventId":"01a0e07e-7af4-7d05-a166-744b053173da","findingId":"01a0df85-c90c-74b4-b254-5f7cbfcea81d","at":"2026-09-27T01:33:00.532Z","actor":"model:claude-sonnet-5","prev":"29cfb864e4eb600b032160138c5868edf651e7d6d965dddcd25914c482807e59","kind":"reviewed","decision":"approve","tier":1,"note":"All quoted figures (97489 alerts, 1181.92 ha, 1272 highest-confidence alerts, next-day pull 96205/1166.36, NDVI -0.104, 99% valid pixels) match the evidence exactly.; Narration explicitly reproduces all five blind spots (attribution limits, alert lag, regrowth/harvest confusion, tropics-only coverage, cloud/optical delay) plus adds context on El Niño-driven dryness and the marginal NDVI threshold.; No natural persons are named or identifiable; only a location and satellite data are discussed.; AOI is tagged forest-hotspot/discovered, not control, so this is a genuine monitoring target rather than a calibration control area.","verdict":"publish"},{"v":1,"eventId":"01a0e07e-7af9-7e65-ae27-63ad6ef4105c","findingId":"01a0df85-c90c-74b4-b254-5f7cbfcea81d","at":"2026-09-27T01:33:00.537Z","actor":"model:claude-sonnet-5","prev":"afdf8ef15bfdc660ee67e1909c686d2eff84801aae405df5978f4513e38227ea","kind":"status_changed","from":"confirmed","to":"published","reason":"autonomous publish: narrated by model:claude-opus-5, reviewed by model:claude-sonnet-5 (policy 2026-09-26-autonomous)","gates":{"narratedBy":"model:claude-opus-5","reviewedBy":["model:claude-sonnet-5"],"policy":"2026-09-26-autonomous"}}],"inclusion":{"index":470,"leafHash":"105e4c8c93bba67aa52c05ccf5853ff95abcbf4053726ce7ce0240c1f133939e","proof":["8fba3cc06a0534b3053bfdca843a59850e1cadabb49b06f59c4aae37a8d46140","70d1bd2f4705b9daca729cc3b67b37380e6c782418aac0d90955574a8e391e1c","f0cad838e2998ef4a2feb947956b9a647120887f5efc8a2475d530fa473a85d4","493a714a42dd17792681ec1368d8c70cc4ff7767c1aa2584aaa0f3deb8ef6917","f565081aeeeac1e7bf708524afa860f70d7cc95211e3d20c39f88709df60ffcc","cfd8a9689a682dbadf13f73837442d7833d9a97eb16baacdedd92a148869879c","2ac34e8c0b554e5c34b3c90bfdddd542109474aaf07521b16dfe8be684c3edae","2334f041f7aa83a5ef6099e0c9511b6242e982238cdf9f30561e5bf0cb2526cc"],"size":474,"root":"3d5c0e5366efedaf637a782ef6bf63ddd427e2416062effeb6f59afde41a817d"}}