Build the application

## Build
Model compound heatwave and dry air anomalies using ERA5 reanalysis pressure level parameters. Detect prolonged wildfire ignition windows and quantify parametric trigger uncertainty for forestry assets.

## User
Forestry Risk Modeler

## Workflow
1. Retrieve ERA5 2-metre temperature and relative humidity grids via cdsapi.
2. Compute cumulative Vapor Pressure Deficit (VPD) across commercial timber stands.
3. Identify high-risk ignition windows exceeding historical 95th percentile anomalies.
4. Simulate parametric trigger sensitivity and calculate trigger uncertainty bands.

## Interface
Thermal anomaly timeline dashboard with threshold trigger sliders and PDF summary export.

## Data
Execute cdsapi python client requests for ERA5 atmospheric variables or load bundled sample NetCDF files in sandbox.
Starter examples below are synthetic. They are not downloaded records.

## Source
[Copernicus Climate Data Store client](https://raw.githubusercontent.com/ecmwf/cdsapi/master/README.rst)
Executes cdsapi.Client retrieve queries for ERA5 variables including temperature and pressure levels.

## Tools
optuna, cdsapi, pandas, scipy.
Optimizes trigger threshold boundaries to balance contract basis risk and false positive trigger occurrences.
Requires standard Python 3.10+ execution environment with optimization packages and API LLM for report generation.

## Deliverables
Submit runnable source, pinned dependencies, setup commands, a working interface, example outputs, and a recorded end-to-end demo. Include tests and an action log.

## Acceptance
- Calculates Vapor Pressure Deficit from ERA5 temperature and dewpoint values.
- Optuna hyperparameter study converges on optimal threshold parameter bounds.
- Yields confidence intervals capturing trigger misclassification uncertainty.

## Starter Examples
```json
[
  {
    "input": {
      "forest_tract_id": "TIMBER-NORTH-04",
      "era5_max_temp_k": 312.4,
      "dewpoint_temp_k": 281.2,
      "consecutive_extreme_days": 9
    },
    "expected_output": {
      "forest_tract_id": "TIMBER-NORTH-04",
      "vpd_kpa": 3.82,
      "ignition_window_risk": "EXTREME",
      "simulated_trigger_fired": true,
      "confidence_interval": [
        0.88,
        0.96
      ]
    }
  },
  {
    "input": {
      "forest_tract_id": "TIMBER-NORTH-05",
      "era5_max_temp_k": 298.1,
      "dewpoint_temp_k": 289.4,
      "consecutive_extreme_days": 1
    },
    "expected_output": {
      "forest_tract_id": "TIMBER-NORTH-05",
      "vpd_kpa": 0.94,
      "ignition_window_risk": "NOMINAL",
      "simulated_trigger_fired": false,
      "confidence_interval": [
        0.02,
        0.08
      ]
    }
  }
]
```

## Safety
Performs retrospective risk modeling only; provides no real-time life safety evacuation orders or underwriting commitments.
Use mock external writes. Require human review before any real deployment.