electricsheepafrica/africa-temperature-change

Africa Historical Temperature Change Dataset (1961-2019)

Dataset Description

This dataset captures how land surface temperatures have changed across 53 African countries and 6 regional aggregates over the past six decades (1961–2019). Rather than raw temperatures, it provides temperature anomalies — the difference between a location's observed temperature and a long-term baseline (1951–1980).

Anomalies are the clearest way to see climate shifts: they remove seasonal and geographic offsets so you can compare trends across places and over time.

Key Statistics

• Total Records: 2,006 temperature series
• Countries: 53 African nations
• Regional Aggregates: 6 (Africa, Eastern, Middle, Northern, Southern, Western Africa)
• Temporal Coverage: 1961–2019 (59 years)
• Time Periods: 17 (monthly, seasonal, and annual)
• File Formats: CSV (0.74 MB), Parquet (0.45 MB)

Data Source and Methodology

The records are derived from the widely used NASA GISTEMP product and repackaged by FAOSTAT into analysis-ready tables. That means you get scientifically vetted measurements presented in tidy and wide formats, so you can jump straight into visualization, statistical modeling, or combining these values with economic, health, land-use, or emissions data using standard ISO3 country codes.

Baseline Period: 1951–1980 (temperatures are expressed as anomalies relative to this baseline)

Why This Matters

Trends in temperature anomalies reveal:

• Where warming is greatest across the African continent
• Where variability is increasing and climate instability is emerging
• How quickly climates are shifting — information vital for climate science, policy planning, risk assessment, and public storytelling

Because values are anomaly-based and span 1961–2019, the dataset supports:

• ✅ Robust trend estimation (e.g., degrees per decade)
• ✅ Extreme-event context
• ✅ Longitudinal comparisons between regions
• ✅ Climate change impact assessments

Who Benefits Most

• Data scientists building climate models and forecasts
• Climate researchers studying African warming patterns
• Students learning about climate change through real data
• Journalists creating data-driven climate stories
• NGOs assessing climate risks and adaptation needs
• Civic technologists powering maps, dashboards, and visualizations
• Policy makers needing evidence for climate action

Key Practical Advantages

✅ Ready-to-use: Tidy and wide formats for Pandas, R, and visualization tools
✅ Consistency: ISO3 codes included to avoid name-matching problems when joining with other country-level datasets
✅ Comparability: Anomaly values referenced to a standard baseline (1951–1980), enabling fair comparisons across countries and decades
✅ Depth: Annual coverage from 1961 through 2019 (monthly and seasonal variants available for higher temporal resolution)
✅ Quality: Scientifically vetted NASA GISTEMP data, processed by FAOSTAT

Dataset Structure

Columns

Column	Type	Description
Area Code	integer	Numeric code for the area/country
Area	string	Country or region name
Months Code	integer	Numeric code for the time period
Months	string	Time period (e.g., "January", "Year", "DecJanFeb")
Element Code	integer	Numeric code for the data element
Element	string	Data element type ("Temperature change" or "Standard Deviation")
Unit	string	Unit of measurement (°C)
Y1961–Y2019	float	Temperature anomaly values for each year (59 columns)

Data Elements

1. Temperature change: The actual temperature anomaly in °C relative to 1951–1980 baseline
2. Standard Deviation: Statistical measure of variability for each measurement

Temporal Coverage

17 time periods per country:

• 12 monthly measurements: January through December
• 4 seasonal aggregates:
  • DecJanFeb (Winter/Summer depending on hemisphere)
  • MarAprMay (Spring/Autumn)
  • JunJulAug (Summer/Winter)
  • SepOctNov (Autumn/Spring)
• 1 annual aggregate: Year

Geographic Coverage

Regional Aggregates (6)

• Africa (continent-wide)
• Eastern Africa
• Middle Africa
• Northern Africa
• Southern Africa
• Western Africa

Countries (53)

North Africa (5): Algeria, Egypt, Libya, Morocco, Tunisia

West Africa (15): Benin, Burkina Faso, Gambia, Ghana, Guinea, Guinea-Bissau, Liberia, Mali, Mauritania, Niger, Nigeria, Senegal, Sierra Leone, Togo, Cape Verde

East Africa (18): Burundi, Comoros, Djibouti, Eritrea, Ethiopia, Kenya, Madagascar, Malawi, Mauritius, Mozambique, Rwanda, Seychelles, Somalia, South Sudan, Sudan, Tanzania (United Republic of), Uganda, Zambia, Zimbabwe

Central Africa (8): Angola, Cameroon, Central African Republic, Chad, Congo, Democratic Republic of the Congo, Equatorial Guinea, Gabon

Southern Africa (7): Botswana, Eswatini, Lesotho, Namibia, South Africa, and others covered in regional aggregates

Data Loading

Using Pandas (Parquet - Recommended)

import pandas as pd

# Load Parquet (faster and more efficient)
df = pd.read_parquet("hf://datasets/electricsheepafrica/africa-temperature-change/africa_temperature_change.parquet")

print(f"Loaded {len(df):,} temperature records")
print(df.head())

Using Pandas (CSV)

import pandas as pd

# Load CSV
df = pd.read_csv("hf://datasets/electricsheepafrica/africa-temperature-change/africa_temperature_change.csv")

# View structure
print(df.columns)

Using Hugging Face Datasets

from datasets import load_dataset

# Load the dataset
dataset = load_dataset("electricsheepafrica/africa-temperature-change")
df = dataset['train'].to_pandas()

Example Usage

Basic Analysis

import pandas as pd
import matplotlib.pyplot as plt

# Load data
df = pd.read_parquet("hf://datasets/electricsheepafrica/africa-temperature-change/africa_temperature_change.parquet")

# Filter for annual temperature changes only
annual_temps = df[(df['Months'] == 'Year') & (df['Element'] == 'Temperature change')]

# Get year columns
year_cols = [col for col in df.columns if col.startswith('Y')]
years = [int(col[1:]) for col in year_cols]

print(f"Countries covered: {annual_temps['Area'].nunique()}")
print(f"Years covered: {len(years)} ({min(years)}-{max(years)})")

Visualize Temperature Trends

# Plot temperature change for Egypt
egypt = annual_temps[annual_temps['Area'] == 'Egypt']
temps = egypt[year_cols].values[0]

plt.figure(figsize=(12, 6))
plt.plot(years, temps, linewidth=2)
plt.title('Egypt Annual Temperature Anomaly (1961-2019)', fontsize=14, fontweight='bold')
plt.xlabel('Year')
plt.ylabel('Temperature Anomaly (°C)')
plt.grid(True, alpha=0.3)
plt.axhline(y=0, color='r', linestyle='--', alpha=0.5)
plt.show()

Regional Comparison

# Compare regional temperature trends
regional_data = annual_temps[annual_temps['Area'].str.contains('Africa')]

plt.figure(figsize=(14, 8))
for _, row in regional_data.iterrows():
    area = row['Area']
    temps = row[year_cols].values
    plt.plot(years, temps, label=area, linewidth=2, marker='o', markersize=3)

plt.title('African Regional Temperature Anomalies (1961-2019)', fontsize=14, fontweight='bold')
plt.xlabel('Year')
plt.ylabel('Temperature Anomaly (°C)')
plt.legend()
plt.grid(True, alpha=0.3)
plt.axhline(y=0, color='r', linestyle='--', alpha=0.5)
plt.show()

Calculate Warming Rates

import numpy as np
from scipy import stats

# Calculate warming rate (°C per decade) for each country
warming_rates = []

for _, row in annual_temps.iterrows():
    area = row['Area']
    temps = row[year_cols].values.astype(float)
    
    # Linear regression: temperature vs time
    slope, intercept, r_value, p_value, std_err = stats.linregress(years, temps)
    
    # Convert to degrees per decade
    rate_per_decade = slope * 10
    
    warming_rates.append({
        'Country': area,
        'Warming Rate (°C/decade)': rate_per_decade,
        'R²': r_value**2,
        'P-value': p_value
    })

warming_df = pd.DataFrame(warming_rates).sort_values('Warming Rate (°C/decade)', ascending=False)
print("\nTop 10 Fastest Warming Areas:")
print(warming_df.head(10))

Reshape to Long Format for Analysis

# Convert wide format to long format for easier analysis
long_data = []

for _, row in df.iterrows():
    for year_col in year_cols:
        year = int(year_col[1:])
        value = row[year_col]
        
        long_data.append({
            'Area': row['Area'],
            'Month': row['Months'],
            'Element': row['Element'],
            'Year': year,
            'Value': value
        })

df_long = pd.DataFrame(long_data)

# Now you can easily filter and aggregate
recent_decade = df_long[df_long['Year'] >= 2010]
avg_temp_change = recent_decade.groupby('Area')['Value'].mean().sort_values(ascending=False)

Use Cases

1. Climate Trend Analysis

• Identify warming hotspots across Africa
• Calculate decadal warming rates
• Compare regional temperature trajectories
• Detect acceleration in warming trends

2. Correlation Studies

• Link temperature changes to emissions data
• Correlate with deforestation rates
• Analyze relationships with public health outcomes
• Study impacts on agricultural productivity

3. Data Visualization & Storytelling

• Create interactive temperature maps
• Build climate dashboards for African countries
• Generate climate change infographics
• Produce data journalism pieces

4. Machine Learning Applications

• Time series forecasting of future temperatures
• Clustering countries by warming patterns
• Anomaly detection for extreme climate events
• Predictive modeling of climate impacts

5. Policy and Risk Assessment

• Support climate adaptation planning
• Inform disaster risk reduction strategies
• Guide agricultural policy decisions
• Assess climate vulnerability by region

Data Limitations

• Temporal Coverage: Data ends in 2019; does not include most recent years
• Spatial Resolution: Country and regional level only; no sub-national detail
• Data Type: Temperature anomalies only; does not include precipitation, humidity, or other climate variables
• Missing Data: Some years may have missing values for certain countries or time periods
• Baseline: All anomalies are relative to 1951–1980 baseline

Citation

If you use this dataset in your research or projects, please cite:

@dataset{africa_temperature_change_2025,
  title={Africa Historical Temperature Change Dataset (1961-2019)},
  author={Electric Sheep Africa},
  year={2025},
  publisher={Hugging Face},
  source={NASA GISTEMP via FAOSTAT},
  url={https://huggingface.co/datasets/electricsheepafrica/africa-temperature-change}
}

Original Data Sources:

• NASA GISTEMP: https://data.giss.nasa.gov/gistemp/
• FAOSTAT Environment Database: https://www.fao.org/faostat/

License

This dataset is released under the Creative Commons Attribution 4.0 International (CC BY 4.0) license.

Original data courtesy of NASA GISTEMP and FAO.

Contact

For questions, issues, or suggestions regarding this dataset, please open an issue on the Hugging Face dataset repository.

---

Last Updated: November 2025
Data Period: 1961–2019
Source: NASA GISTEMP via FAOSTAT
Maintained by: Electric Sheep Africa