electricsheepafrica/africa-icpac-geonode-2015-tropical-cyclone-megh-path

---
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
- tabular-regression
task_ids: []
tags:
- africa
- humanitarian
- hdx
- electric-sheep-africa
- geodata
- hazards-and-risk
- som
pretty_name: "2015 Tropical Cyclone Megh Path"
dataset_info:
splits:
- name: train
num_examples: 28
- name: test
num_examples: 7
---

# 2015 Tropical Cyclone Megh Path

**Publisher:** IGAD Climate Prediction and Applications Center (ICPAC) · **Source:** [HDX](https://data.humdata.org/dataset/icpac-geonode-2015-tropical-cyclone-megh-path) · **License:** `cc-by` · **Updated:** 2026-04-05

---

## Abstract

<p>This layer shows the movement path for 2015 Tropical Cyclone Megh. Following Tropical Cyclone Chapala, new tropical cyclone Megh originated from the Arabian Sea causing even more rains in parts of Bari region in Puntland and Somaliland. The storm produced a maximum windspeed of 110knots.</p>

<p>Areas affected included: Af Kalahay, Alula, Bareda, BiyoCade, Boolimoog, Dhurbo, Fagoora, Geesalay, Murcanyo, Sayn Weyn, Sayn Yar, Toxin and Xaabo.</p>
<p>Re-estimated population figures after Tropical Megh, showed 4.9 million people were in need of assistance, 308,700 children under-5 were acutely malnourished, of which 55,800 were severely malnourished and 1.1 million remain in a protracted internal displacement situation.</p>

Each row in this dataset represents geolocated point observations. Data was last updated on HDX on 2026-04-05. Geographic scope: **SOM**.

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

---

## Dataset Characteristics

| | |
|---|---|
| **Domain** | Forced displacement and migration |
| **Unit of observation** | Geolocated point observations |
| **Rows (total)** | 36 |
| **Columns** | 32 (21 numeric, 11 categorical, 0 datetime) |
| **Train split** | 28 rows |
| **Test split** | 7 rows |
| **Geographic scope** | SOM |
| **Publisher** | IGAD Climate Prediction and Applications Center (ICPAC) |
| **HDX last updated** | 2026-04-05 |

---

## Variables

**Geographic** — `cy` (range 5.0–5.0), `yyyymmddhh` (range 2015110518.0–2015111012.0), `lat` (range 12.2–13.7), `long` (range 46.7–63.2), `vmax` (range 20.0–110.0) and 3 others.

**Identifier / Metadata** — `fid` (a__2015_TC_Megh.1, a__2015_TC_Megh.2, a__2015_TC_Megh.21), `fid_1` (range 1.0–36.0), `windcode` (NEQ), `stormname` (MEGH), `esa_source` (HDX) and 1 others.

**Other** — `the_geom` (POINT (54.9 12.6), POINT (50.8 12.2), POINT (58.3 12.8)), `basin` (IO), `tech` (BEST), `tau` (range 0.0–0.0), `mslp` (range 941.0–1007.0) and 13 others.

---

## Quick Start

```python
from datasets import load_dataset

ds = load_dataset("electricsheepafrica/africa-icpac-geonode-2015-tropical-cyclone-megh-path")
train = ds["train"].to_pandas()
test = ds["test"].to_pandas()

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

---

## Schema

| Column | Type | Null % | Range / Sample Values |
|---|---|---|---|
| `fid` | object | 0.0% | a__2015_TC_Megh.1, a__2015_TC_Megh.2, a__2015_TC_Megh.21 |
| `fid_1` | int64 | 0.0% | 1.0 – 36.0 (mean 18.5) |
| `the_geom` | object | 0.0% | POINT (54.9 12.6), POINT (50.8 12.2), POINT (58.3 12.8) |
| `basin` | object | 0.0% | IO |
| `cy` | int64 | 0.0% | 5.0 – 5.0 (mean 5.0) |
| `yyyymmddhh` | int64 | 0.0% | 2015110518.0 – 2015111012.0 (mean 2015110789.5556) |
| `tech` | object | 0.0% | BEST |
| `tau` | int64 | 0.0% | 0.0 – 0.0 (mean 0.0) |
| `lat` | float64 | 0.0% | 12.2 – 13.7 (mean 12.725) |
| `long` | float64 | 0.0% | 46.7 – 63.2 (mean 54.8417) |
| `vmax` | int64 | 0.0% | 20.0 – 110.0 (mean 71.5278) |
| `mslp` | int64 | 0.0% | 941.0 – 1007.0 (mean 969.3056) |
| `ty` | object | 0.0% | TY, TS, TD |
| `rad` | int64 | 0.0% | 0.0 – 0.0 (mean 0.0) |
| `windcode` | object | 8.3% | NEQ |
| `rad1` | int64 | 0.0% | 0.0 – 60.0 (mean 33.1944) |
| `rad2` | int64 | 0.0% | 0.0 – 55.0 (mean 31.3889) |
| `rad3` | int64 | 0.0% | 0.0 – 60.0 (mean 31.3889) |
| `rad4` | int64 | 0.0% | 0.0 – 60.0 (mean 33.4722) |
| `radp` | int64 | 0.0% | 1008.0 – 1011.0 (mean 1009.0833) |
| `rrp` | int64 | 0.0% | 120.0 – 200.0 (mean 150.8333) |
| `mrd` | int64 | 0.0% | 5.0 – 40.0 (mean 16.6944) |
| `gusts` | int64 | 0.0% | 0.0 – 0.0 (mean 0.0) |
| `eye` | int64 | 0.0% | 0.0 – 0.0 (mean 0.0) |
| `subregn` | object | 5.6% | A |
| `maxseas` | int64 | 0.0% | 0.0 – 0.0 (mean 0.0) |
| `dir` | int64 | 0.0% | 0.0 – 0.0 (mean 0.0) |
| `speed` | int64 | 0.0% | |
| `stormname` | object | 0.0% | MEGH |
| `depth` | object | 5.6% | D, M |
| `esa_source` | object | 0.0% | HDX |
| `esa_processed` | object | 0.0% | |

---

## Numeric Summary

| Column | Min | Max | Mean | Median |
|---|---|---|---|---|
| `fid_1` | 1.0 | 36.0 | 18.5 | 18.5 |
| `cy` | 5.0 | 5.0 | 5.0 | 5.0 |
| `yyyymmddhh` | 2015110518.0 | 2015111012.0 | 2015110789.5556 | 2015110806.0 |
| `tau` | 0.0 | 0.0 | 0.0 | 0.0 |
| `lat` | 12.2 | 13.7 | 12.725 | 12.7 |
| `long` | 46.7 | 63.2 | 54.8417 | 54.9 |
| `vmax` | 20.0 | 110.0 | 71.5278 | 75.0 |
| `mslp` | 941.0 | 1007.0 | 969.3056 | 966.5 |
| `rad` | 0.0 | 0.0 | 0.0 | 0.0 |
| `rad1` | 0.0 | 60.0 | 33.1944 | 30.0 |
| `rad2` | 0.0 | 55.0 | 31.3889 | 30.0 |
| `rad3` | 0.0 | 60.0 | 31.3889 | 30.0 |
| `rad4` | 0.0 | 60.0 | 33.4722 | 30.0 |
| `radp` | 1008.0 | 1011.0 | 1009.0833 | 1009.0 |
| `rrp` | 120.0 | 200.0 | 150.8333 | 155.0 |

---

## 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`. 8 column(s) with >80% missing values were removed: `technum`, `initials`, `seas`, `seascode`, `seas1`, `seas2`.... 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 IGAD Climate Prediction and Applications Center (ICPAC) 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/icpac-geonode-2015-tropical-cyclone-megh-path) for the publisher's own methodology notes and caveats.

---

## Citation

```bibtex
@dataset{hdx_africa_icpac_geonode_2015_tropical_cyclone_megh_path,
title = {2015 Tropical Cyclone Megh Path},
author = {IGAD Climate Prediction and Applications Center (ICPAC)},
year = {2026},
url = {https://data.humdata.org/dataset/icpac-geonode-2015-tropical-cyclone-megh-path},
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.*