Getting Started with FAO WaPOR for Water Productivity Analysis

What is WaPOR?

The FAO Water Productivity Open-access portal (WaPOR) provides satellite-derived data on water consumption, biomass production, and water productivity across Africa and the Near East. It covers three spatial resolutions — continental (250 m), national (100 m), and sub-national (30 m for selected areas) — at dekadal, monthly, and annual temporal scales.

The core equation is straightforward:

WP = Y / ETa

Where Y is the harvested yield (kg/ha) and ETa is the actual evapotranspiration (mm). The result gives you kg/m³ — how much crop you produce per cubic metre of water consumed.

Key Data Products

| Product | Code | Resolution | Description | |---------|------|------------|-------------| | Actual Evapotranspiration | ETa | 100 m / 250 m | Water consumed by crops and soil | | Above-Ground Biomass | AGBP | 100 m / 250 m | Total dry matter produced | | Net Primary Production | NPP | 250 m | Carbon fixed by vegetation | | Land Cover | LCC | 100 m | Annual cropland classification | | Phenology | PHE | 100 m | Start/end of growing season |

Step 1 — Access the Data via Python

The wapor Python package (or direct API) makes downloading straightforward. First, register at wapor.apps.fao.org to get your API token.

import requests
import os

API_TOKEN = "your_wapor_api_key"
BASE_URL = "https://io.apps.fao.org/gismgr/api/v1"

def download_wapor_raster(crop_code, year, season, output_dir):
    """
    Download a WaPOR L2 product for a specific year and season.
    crop_code: e.g. 'L2_AETI_A' for annual ETa at 100m
    """
    headers = {"X-GISMGR-API-KEY": API_TOKEN}
    
    # Build query for the specific time period
    params = {
        "year": year,
        "season": season,
        "format": "GeoTIFF",
    }
    
    url = f"{BASE_URL}/catalog/workspaces/WAPOR_2/{crop_code}/rasters"
    response = requests.get(url, headers=headers, params=params)
    response.raise_for_status()
    
    raster_url = response.json()["response"]["downloadUrl"]
    
    # Download the file
    out_path = os.path.join(output_dir, f"{crop_code}_{year}.tif")
    with open(out_path, "wb") as f:
        f.write(requests.get(raster_url).content)
    
    return out_path

Step 2 — Compute Water Productivity

Once you have ETa (in mm) and AGBP (in kg/ha), calculate water productivity per pixel:

import rasterio
import numpy as np

def compute_water_productivity(agbp_path, eta_path, output_path, harvest_index=0.45):
    """
    Compute seasonal water productivity (kg/m³).
    harvest_index: fraction of AGBP that becomes grain yield (wheat ~0.45)
    """
    with rasterio.open(agbp_path) as agbp_src:
        agbp = agbp_src.read(1).astype(np.float32)
        profile = agbp_src.profile
        agbp = np.where(agbp == agbp_src.nodata, np.nan, agbp)

    with rasterio.open(eta_path) as eta_src:
        eta = eta_src.read(1).astype(np.float32)
        eta = np.where(eta == eta_src.nodata, np.nan, eta)

    # Convert ETa from mm to m³/ha (1 mm/ha = 10 m³/ha)
    eta_m3_ha = eta * 10

    # Yield estimate using harvest index
    yield_kg_ha = agbp * harvest_index

    # Water productivity in kg/m³
    wp = np.where(eta_m3_ha > 0, yield_kg_ha / eta_m3_ha, np.nan)

    profile.update(dtype=rasterio.float32, nodata=-9999)
    with rasterio.open(output_path, "w", **profile) as dst:
        dst.write(np.where(np.isnan(wp), -9999, wp).astype(np.float32), 1)

    return wp

Step 3 — Visualise Results in QGIS

After computing water productivity rasters, load them into QGIS and apply a diverging colour ramp (e.g. RdYlGn) to highlight areas of low vs. high water productivity. Key visualisation tips:

• Clip the raster to your irrigation scheme boundary to avoid surrounding dryland areas skewing the statistics
• Use band statistics per season — WP varies significantly between winter and summer crops
• Compare years using the QGIS temporal controller to spot productivity trends

My QGIS Plugin

I packaged this entire workflow — download, compute, visualise — into a QGIS plugin called wapor-water-productivity, released as open source. It automates seasonal WP calculation with a simple UI and outputs publication-ready maps directly from QGIS.

You can install it from the QGIS Plugin Repository (search "WaPOR Water Productivity") or from the GitHub repository.

Key Takeaways

• WaPOR removes the need for costly field experiments to estimate ETa at scale
• The 100 m national products are sufficient for irrigation scheme analysis
• Python + rasterio handles bulk downloads efficiently — 60 dekadal files in under 5 minutes
• Always validate pixel-level WP estimates against field measurements at sampling points

For questions or collaboration on WaPOR-based water productivity assessment, feel free to reach out.