get_region labels growing species "declining" — the classifier tests |recentChangePct| and drops the sign #220

New issue

Closed

opened 2026-08-12 23:13:20 +00:00 by coilyco-ops · 0 comments

coilyco-ops commented

2026-08-12 23:13:20 +00:00

Member

QA sweep, Eco via Sirens, day 40, 2026-08-12.

Observed

get_region.speciesRisk puts thriving species in the declining bucket:

Species	Cycle change	Recent change	State
HuckleberrySpecies	+425%	+28.0%	`declining`
TomatoesSpecies	+635%	+74.5%	`declining`
RiceSpecies	+199%	+19.6%	`declining`
BighornSheepSpecies	+5.1%	+24.0%	`declining`
CoyoteSpecies	+15.3%	+23.6%	`declining`
WheatSpecies	+49.8%	−52.4%	`declining`
DaisySpecies	−58.3%	−2.1%	`stable`
CriminiMushroomSpecies	−61.2%	−1.8%	`stable`

Each carries "reason": "Population is declining, but the combined at-risk threshold was not crossed."

Diagnosis

Sort by abs(recentChangePct) and the buckets separate cleanly at the documented 0.15 band:

every declining entry has |recent| ≥ 0.196
every stable entry has |recent| ≤ 0.139

So the rule is abs(recentChangePct) > recentDeclinePct — the sign is discarded, and a population that grew 74% in the recent window is reported as declining.

Expected

declining requires recentChangePct < -threshold. Growth should be stable (or a growing state, if one is wanted). Daisy at −58% over the cycle probably deserves better than stable too.

Impact

This is the field a reader uses to decide what to protect. It currently names the winners.

QA sweep, Eco via Sirens, day 40, 2026-08-12. ## Observed `get_region.speciesRisk` puts thriving species in the `declining` bucket: | Species | Cycle change | Recent change | State | |---|---|---|---| | HuckleberrySpecies | **+425%** | **+28.0%** | `declining` | | TomatoesSpecies | **+635%** | **+74.5%** | `declining` | | RiceSpecies | +199% | +19.6% | `declining` | | BighornSheepSpecies | +5.1% | +24.0% | `declining` | | CoyoteSpecies | +15.3% | +23.6% | `declining` | | WheatSpecies | +49.8% | −52.4% | `declining` | | DaisySpecies | **−58.3%** | −2.1% | `stable` | | CriminiMushroomSpecies | **−61.2%** | −1.8% | `stable` | Each carries `"reason": "Population is declining, but the combined at-risk threshold was not crossed."` ## Diagnosis Sort by `abs(recentChangePct)` and the buckets separate cleanly at the documented 0.15 band: - every `declining` entry has |recent| ≥ 0.196 - every `stable` entry has |recent| ≤ 0.139 So the rule is `abs(recentChangePct) > recentDeclinePct` — the sign is discarded, and a population that grew 74% in the recent window is reported as declining. ## Expected `declining` requires `recentChangePct < -threshold`. Growth should be `stable` (or a `growing` state, if one is wanted). Daisy at −58% over the cycle probably deserves better than `stable` too. ## Impact This is the field a reader uses to decide what to protect. It currently names the winners.

coilyco-ops referenced this issue

2026-08-12 23:34:54 +00:00

Tracking: 2026-08-12 full-surface QA sweep of the eco MCP (22 tools, 23 findings) #240

coilyco-ops referenced this issue

2026-08-12 23:36:09 +00:00

Tracking: 2026-08-12 full-surface QA sweep of the eco MCP (22 tools, 23 findings) #240

coilyco-ops referenced this issue

2026-08-12 23:47:23 +00:00

Tracking: 2026-08-12 full-surface QA sweep of the eco MCP (22 tools, 23 findings) #240

coilyco-ops referenced this issue from a commit

2026-08-13 04:53:24 +00:00

fix(region): classify species decline on the signed change, not its size