electricsheepafrica/africa-climada-crop-production-dataset

---
annotations_creators:
- no-annotation
language_creators:
- found
language:
- en
license: cc-by-4.0
multilinguality:
- monolingual
size_categories:
- 1K<n<10K
source_datasets:
- original
task_categories:
- other
task_ids: []
tags:
- africa
- humanitarian
- hdx
- electric-sheep-africa
- geodata
- hazards-and-risk
- humanitarian-response-plan-hrp
- hxl
- afg
- bfa
- bdi
- cmr
- caf
pretty_name: "Crop production: Humanitarian Response Plan (HRP) Countries Exposure Data for Disaster Risk Assessment"
dataset_info:
splits:
- name: train
num_examples: 2650
- name: test
num_examples: 662
---

# Crop production: Humanitarian Response Plan (HRP) Countries Exposure Data for Disaster Risk Assessment

**Publisher:** ETH Zürich - Weather and Climate Risks · **Source:** [HDX](https://data.humdata.org/dataset/climada-crop-production-dataset) · **License:** `cc-by` · **Updated:** 2025-09-03

---

## Abstract

Historical and twenty-first century crop production in tons. Global gridded (4km resolution) crop yield simulations for maize, rice, soybean, and wheat, encompassing an ensemble of transient yield simulation output from eight global gridded crop models driven by bias-corrected output from five global climate models, as facilitated by the Inter-Sectoral Impact Model Intercomparison Project (ISIMIP, isimip.org)

Each row in this dataset represents first-level administrative unit observations. Data was last updated on HDX on 2025-09-03. Geographic scope: **AFG, BFA, BDI, CMR, CAF, TCD, COL, COD, and 15 others**.

*Curated into ML-ready Parquet format by [Electric Sheep Africa](https://huggingface.co/electricsheepafrica).*

---

## Dataset Characteristics

| | |
|---|---|
| **Domain** | Climate and environment |
| **Unit of observation** | First-level administrative unit observations |
| **Rows (total)** | 3,313 |
| **Columns** | 9 (3 numeric, 6 categorical, 0 datetime) |
| **Train split** | 2,650 rows |
| **Test split** | 662 rows |
| **Geographic scope** | AFG, BFA, BDI, CMR, CAF, TCD, COL, COD, and 15 others |
| **Publisher** | ETH Zürich - Weather and Climate Risks |
| **HDX last updated** | 2025-09-03 |

---

## Variables

**Geographic** — `country_name` (Nigeria, Afghanistan, Colombia), `admin1_name` (Centre, Sucre, North), `latitude` (range -25.96–51.4062), `longitude` (range -81.55–98.7083).

**Outcome / Measurement** — `value` (range 0.0–1521161858.0).

**Identifier / Metadata** — `esa_source` (HDX), `esa_processed` (2026-04-04).

**Other** — `aggregation` (sum), `indicator` (crop-production.mai.noirr.USD, crop-production.mai.firr.USD, crop-production.whe.noirr.USD).

---

## Quick Start

```python
from datasets import load_dataset

ds = load_dataset("electricsheepafrica/africa-climada-crop-production-dataset")
train = ds["train"].to_pandas()
test = ds["test"].to_pandas()

print(train.shape)
train.head()
```

---

## Schema

| Column | Type | Null % | Range / Sample Values |
|---|---|---|---|
| `country_name` | object | 0.0% | Nigeria, Afghanistan, Colombia |
| `admin1_name` | object | 0.0% | Centre, Sucre, North |
| `latitude` | float64 | 0.0% | -25.96 – 51.4062 (mean 13.6832) |
| `longitude` | float64 | 0.0% | -81.55 – 98.7083 (mean 14.9992) |
| `aggregation` | object | 0.0% | sum |
| `indicator` | object | 0.0% | crop-production.mai.noirr.USD, crop-production.mai.firr.USD, crop-production.whe.noirr.USD |
| `value` | float64 | 0.0% | 0.0 – 1521161858.0 (mean 7654114.5305) |
| `esa_source` | object | 0.0% | HDX |
| `esa_processed` | object | 0.0% | 2026-04-04 |

---

## Numeric Summary

| Column | Min | Max | Mean | Median |
|---|---|---|---|---|
| `latitude` | -25.96 | 51.4062 | 13.6832 | 10.5663 |
| `longitude` | -81.55 | 98.7083 | 14.9992 | 24.9758 |
| `value` | 0.0 | 1521161858.0 | 7654114.5305 | 0.0 |

---

## Curation

Raw data was downloaded from HDX via the CKAN API and converted to Parquet. Column names were lowercased and standardised to snake_case. Common missing-value markers (`N/A`, `null`, `none`, `-`, `unknown`, `no data`, `#N/A`) were unified to `NaN`. 3 column(s) were cast from string to numeric or datetime based on parse-success rate (>85% threshold). The dataset was split 80/20 into train and test partitions using a fixed random seed (42) and saved as Snappy-compressed Parquet.

---

## Limitations

- Data originates from ETH Zürich - Weather and Climate Risks and has not been independently validated by ESA.
- Automated cleaning cannot correct for misreported values, definitional inconsistencies, or sampling bias in the original collection.
- This dataset spans 23 countries; geographic and methodological inconsistencies across national boundaries may affect cross-country comparability.
- Refer to the [original HDX dataset page](https://data.humdata.org/dataset/climada-crop-production-dataset) for the publisher's own methodology notes and caveats.

---

## Citation

```bibtex
@dataset{hdx_africa_climada_crop_production_dataset,
title = {Crop production: Humanitarian Response Plan (HRP) Countries Exposure Data for Disaster Risk Assessment},
author = {ETH Zürich - Weather and Climate Risks},
year = {2025},
url = {https://data.humdata.org/dataset/climada-crop-production-dataset},
note = {Repackaged for machine learning by Electric Sheep Africa (https://huggingface.co/electricsheepafrica)}
}
```

---

*[Electric Sheep Africa](https://huggingface.co/electricsheepafrica) — Africa's ML dataset infrastructure. Lagos, Nigeria.*