Skip to main content
Challenge

Build a STAC Agricultural Drought Image Retriever

Process public STAC earth observation collections to discover imagery over agricultural acreage. Compile optical asset links matching temporal and cloud constraints to enable downstream calculation of vegetative drought anomalies.

Data ScienceHosted by Vera
Challenge brief

What you are building

The core problem, expected build, and operating context for this challenge.

Process public STAC earth observation collections to discover imagery over agricultural acreage. Compile optical asset links matching temporal and cloud constraints to enable downstream calculation of vegetative drought anomalies.

Crop Insurance Portfolio Analyst

Incorporate field polygon coordinates and historic baseline metadata for target agricultural regions.

Query STAC collections for items representing optical surface reflectance imagery over the targeted fields.

Filter and extract STAC asset links matching strict cloud cover limits and temporal windows.

Compile organized image retrieval dossiers to feed downstream agronomic raster processing pipelines.

Command-line interface producing structured JSON dossiers and simple map plots of STAC item footprints.

Query public Earth Search STAC APIs or download mock STAC Item JSON records from Radiant Earth examples. Starter examples below are synthetic. They are not downloaded records.

STAC specification Applies STAC standard query parameters for spatial-temporal extents and parses STAC Item properties to extract asset links.

llamaindex, pystac, geopandas. LlamaIndex manages the orchestration of metadata retrieval, filtering conditions, and structuring the final asset link dossiers. Hosted or local LLM for structuring parameter extraction and formatting final JSON dossiers.

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.

Correctly queries a STAC API utilizing spatial bounding boxes and target temporal windows.

Filters and discards STAC Items that exceed the specified cloud cover threshold.

Generates a valid JSON dossier populated with direct optical asset links for downstream consumption.

```json [ { "input": { "field_id": "FIELD-SOY-99", "target_month": "2023-07", "max_cloud_cover_pct": 20 }, "expected_output": { "field_id": "FIELD-SOY-99", "discovered_items": 2, "asset_links": [ "https://example.stac/item1/B04.tif", "https://example.stac/item2/B04.tif" ], "dossier_status": "READY" } }, { "input": { "field_id": "FIELD-SOY-100", "target_month": "2023-08", "max_cloud_cover_pct": 10 }, "expected_output": { "field_id": "FIELD-SOY-100", "discovered_items": 0, "asset_links": [], "dossier_status": "NO_CLEAR_IMAGERY" } } ] ```

Retrieves image links for discovery purposes only; does not calculate agronomic losses or automate crop insurance claims. Use mock external writes. Require human review before any real deployment.

Rules

Participation guide

Retrieves image links for discovery purposes only; does not calculate agronomic losses or automate crop insurance claims. Use mock external writes and human review.

Delivery guide

How work is evaluated

Evaluation

Correctly queries a STAC API utilizing spatial bounding boxes and target temporal windows. Filters and discards STAC Items that exceed the specified cloud cover threshold. Generates a valid JSON dossier populated with direct optical asset links for downstream consumption.

Submission format

Runnable source, pinned dependencies, setup commands, working interface, example outputs, recorded demo, tests, and action logs.

Datasets

Challenge datasets

Public datasets, plus the private datasets you have access to.

Evaluation rubric

How submissions are scored

These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.

Dimensions
4 scoring checks
Binary
1 pass or fail dimensions
Ordinal
3 scaled dimensions
Dimension 1

Complete workflow

Incorporate field polygon coordinates and historic baseline metadata for target agricultural regions. Query STAC collections for items representing optical surface reflectance imagery over the targeted fields. Filter and extract STAC asset links matching strict cloud cover limits and temporal windows. Compile organized image retrieval dossiers to feed downstream agronomic raster processing pipelines.

Score scale
012
Score 0

Missing or incorrect.

Score 1

Partly complete, with documented gaps.

Score 2

Complete, with reproducible evidence.

Dimension 2

Correctness and evidence

Correctly queries a STAC API utilizing spatial bounding boxes and target temporal windows. Filters and discards STAC Items that exceed the specified cloud cover threshold. Generates a valid JSON dossier populated with direct optical asset links for downstream consumption.

Score scale
012
Score 0

Missing or incorrect.

Score 1

Partly complete, with documented gaps.

Score 2

Complete, with reproducible evidence.

Dimension 3

Review interface

Command-line interface producing structured JSON dossiers and simple map plots of STAC item footprints.

Score scale
012
Score 0

Missing or incorrect.

Score 1

Partly complete, with documented gaps.

Score 2

Complete, with reproducible evidence.

Dimension 4

Safe execution

Retrieves image links for discovery purposes only; does not calculate agronomic losses or automate crop insurance claims.

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Learning goals

What you should walk away with

  • Incorporate field polygon coordinates and historic baseline metadata for target agricultural regions.

  • Query STAC collections for items representing optical surface reflectance imagery over the targeted fields.

  • Filter and extract STAC asset links matching strict cloud cover limits and temporal windows.

  • Compile organized image retrieval dossiers to feed downstream agronomic raster processing pipelines.

Resources and assets

Reference links and supporting material

Dataset notes

Query public Earth Search STAC APIs or download mock STAC Item JSON records from Radiant Earth examples. Supplied starter examples are synthetic, not source extracts.

Challenge setup

Configuration summary for this challenge.

Challenge input

Input details are not configured.

LlamaIndex

Data framework for LLM applications

Evaluated output

No output checks are configured.

Evaluation checks
  • Incorporate field polygon coordinates and historic baseline metadata for target agricultural regions. Query...
  • Correctly queries a STAC API utilizing spatial bounding boxes and target temporal windows. Filters and disc...
  • Command-line interface producing structured JSON dossiers and simple map plots of STAC item footprints.
Configured evaluation evidence
  • 2 public reference cases
Configured runtime
  • 1 configured tool

Summary shows the first task, up to three tools, and up to three evidence items. Open the technical recipe to view all configured items.

View technical recipe

Configured tools

Action Space
  • LlamaIndex · Optional

Evaluation contract

  • Complete workflow · Weight 40
  • Correctness and evidence · Weight 30
  • Review interface · Weight 20
  • Safe execution · Weight 10

Reference cases

  • 2 public reference cases

Workflow tasks

No workflow tasks are configured.

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build a STAC Agricultural Drought Image Retriever