electricsheepafrica/africa-somalia-floods

---
annotations_creators:
- no-annotation
language_creators:
- found
language:
- en
license: cc-by-4.0
multilinguality:
- monolingual
size_categories:
- n<1K
source_datasets:
- original
task_categories:
- tabular-classification
- other
task_ids: []
tags:
- africa
- humanitarian
- hdx
- electric-sheep-africa
- affected-population
- climate-weather
- flooding
- hxl
- som
pretty_name: "Somalia : Floods"
dataset_info:
splits:
- name: train
num_examples: 59
- name: test
num_examples: 14
---

# Somalia : Floods

**Publisher:** OCHA Somalia · **Source:** [HDX](https://data.humdata.org/dataset/somalia-floods) · **License:** `cc-by` · **Updated:** 2025-09-24

---

## Abstract

This dataset is from preliminary reports by the Ministry of Humanitarian Affairs and Disaster Management, South West State and humanitarian partners.

Each row in this dataset represents subnational administrative unit observations. Data was last updated on HDX on 2025-09-24. Geographic scope: **SOM**.

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

---

## Dataset Characteristics

| | |
|---|---|
| **Domain** | Climate and environment |
| **Unit of observation** | Subnational administrative unit observations |
| **Rows (total)** | 74 |
| **Columns** | 20 (11 numeric, 9 categorical, 0 datetime) |
| **Train split** | 59 rows |
| **Test split** | 14 rows |
| **Geographic scope** | SOM |
| **Publisher** | OCHA Somalia |
| **HDX last updated** | 2025-09-24 |

---

## Variables

**Geographic** — `state` (South West , Jubaland, Puntland), `region` (Lower Shabelle, Gedo, Bari), `region_pcodes` (SO23, SO26, SO16), `district` (Afgooye, Caluula, Cadale), `district_pcodes` (SO2302, SO1603, SO2104) and 8 others.

**Outcome / Measurement** — `people_affected` (0, 600 , 1,200 ), `shelters_affected` (range 0.0–2000.0), `farmland_affected` (range 0.0–20733.0).

**Identifier / Metadata** — `esa_source` (HDX), `esa_processed` (2026-04-08).

**Other** — `people_relocated` (range 0.0–120000.0), `dead_livestock` (range 0.0–1000.0).

---

## Quick Start

```python
from datasets import load_dataset

ds = load_dataset("electricsheepafrica/africa-somalia-floods")
train = ds["train"].to_pandas()
test = ds["test"].to_pandas()

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

---

## Schema

| Column | Type | Null % | Range / Sample Values |
|---|---|---|---|
| `state` | object | 0.0% | South West , Jubaland, Puntland |
| `region` | object | 0.0% | Lower Shabelle, Gedo, Bari |
| `region_pcodes` | object | 0.0% | SO23, SO26, SO16 |
| `district` | object | 0.0% | Afgooye, Caluula, Cadale |
| `district_pcodes` | object | 0.0% | SO2302, SO1603, SO2104 |
| `population` | object | 0.0% | 505,857 , 86,601 , 75,977 |
| `people_affected` | object | 0.0% | 0, 600 , 1,200 |
| `people_displaced` | int64 | 0.0% | 0.0 – 220460.0 (mean 12144.0541) |
| `people_relocated` | int64 | 0.0% | 0.0 – 120000.0 (mean 1697.2973) |
| `people_killed_by_floods` | int64 | 0.0% | 0.0 – 20.0 (mean 1.5946) |
| `shelters_affected` | int64 | 0.0% | 0.0 – 2000.0 (mean 63.2297) |
| `shelters_destroyed` | int64 | 0.0% | 0.0 – 6593.0 (mean 286.6081) |
| `latrines_destroyed` | int64 | 0.0% | 0.0 – 6587.0 (mean 214.1622) |
| `water_point_destroyed` | int64 | 0.0% | 0.0 – 669.0 (mean 20.1081) |
| `farmland_affected` | int64 | 0.0% | 0.0 – 20733.0 (mean 561.5676) |
| `brides_destroyed` | int64 | 0.0% | 0.0 – 2.0 (mean 0.0405) |
| `dead_livestock` | int64 | 0.0% | 0.0 – 1000.0 (mean 13.5135) |
| `ofschools_destroyed` | int64 | 0.0% | 0.0 – 72.0 (mean 3.027) |
| `esa_source` | object | 0.0% | HDX |
| `esa_processed` | object | 0.0% | 2026-04-08 |

---

## Numeric Summary

| Column | Min | Max | Mean | Median |
|---|---|---|---|---|
| `people_displaced` | 0.0 | 220460.0 | 12144.0541 | 12.0 |
| `people_relocated` | 0.0 | 120000.0 | 1697.2973 | 0.0 |
| `people_killed_by_floods` | 0.0 | 20.0 | 1.5946 | 0.0 |
| `shelters_affected` | 0.0 | 2000.0 | 63.2297 | 0.0 |
| `shelters_destroyed` | 0.0 | 6593.0 | 286.6081 | 0.0 |
| `latrines_destroyed` | 0.0 | 6587.0 | 214.1622 | 0.0 |
| `water_point_destroyed` | 0.0 | 669.0 | 20.1081 | 0.0 |
| `farmland_affected` | 0.0 | 20733.0 | 561.5676 | 0.0 |
| `brides_destroyed` | 0.0 | 2.0 | 0.0405 | 0.0 |
| `dead_livestock` | 0.0 | 1000.0 | 13.5135 | 0.0 |
| `ofschools_destroyed` | 0.0 | 72.0 | 3.027 | 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`. 9 column(s) with >80% missing values were removed: `roads_destroyed`, `people_reached_fsc`, `people_reached_nutrition`, `people_reached_wash`, `people_reached_health`, `people_reached_education`.... 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 OCHA Somalia and has not been independently validated by ESA.
- Automated cleaning cannot correct for misreported values, definitional inconsistencies, or sampling bias in the original collection.
- Refer to the [original HDX dataset page](https://data.humdata.org/dataset/somalia-floods) for the publisher's own methodology notes and caveats.

---

## Citation

```bibtex
@dataset{hdx_africa_somalia_floods,
title = {Somalia : Floods},
author = {OCHA Somalia},
year = {2025},
url = {https://data.humdata.org/dataset/somalia-floods},
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.*