Google earth Engine PK

🌍 Mapping Soil Health & Water Quality in Punjab using GIS & Remote Sensing 🌾💧

In this study, I used Google Earth Engine (GEE) to analyze the impact of agricultural practices and land use on soil and water quality across Punjab, Pakistan.

🔎 What was done:
✔ Landsat-8 (2015–2020) for soil health indices

🌱 NDVI → Vegetation health & soil fertility

🟤 BSI → Bare soil & land degradation

🧂 SI → Soil salinity risk

✔ Sentinel-2 (2019–2021) for water quality indicators

💧 NDWI → Surface water extent

🌫 NDTI → Water turbidity

🟢 Chl-a → Algal blooms / chlorophyll concentration

🗺 Six maps were generated (NDVI, BSI, SI, NDWI, NDTI, Chl-a) in a 2×3 grid layout, each with legends for better interpretation.

📊 District-level zonal statistics were calculated, and graphs in the Console show spatial variations in these indices across Punjab’s districts.

🚜 These insights help identify hotspots of soil degradation, salinity issues, and declining water quality, which are crucial for:

Sustainable agriculture planning 🌾

Soil conservation programs 🌱

Water resource management 💧

Climate adaptation strategies 🌍

🌍 Indus Basin – Snow, Meltwater & Climate Trends (2010–2025) ❄️💧🌡️

This study analyzes the changing cryosphere and hydrology of the Indus Basin using MODIS Snow Cover and ERA5-Land climate reanalysis data.

🔹 Snow Cover Duration (SCD): Annual days under snow cover, showing variability in seasonal snowpack.
🔹 Snowmelt (mm): Annual meltwater contribution – a lifeline for agriculture and river flows.
🔹 Mean Temperature (°C): Annual averages, revealing warming trends across the basin.
🔹 Combined Trends: An integrated look at snow persistence, meltwater availability, and climate warming.

📊 Results indicate that rising temperatures are directly influencing snow cover and meltwater dynamics, with significant implications for water security, agriculture, and climate resilience in South Asia.

⚡ Data Sources:

NASA MODIS MOD10A1 (Snow Cover)

ECMWF ERA5-Land (Temperature & Snowmelt)

🔍 These insights are vital for climate adaptation, sustainable water management, and policy planning in the Indus Basin.

🌍 Mapping Landcover Diversity Across Pakistan Using ESA WorldCover 2020 & Shannon Index in Google Earth Engine 🌱

This analysis highlights the spatial distribution of landcover classes across Pakistan’s provinces and quantifies ecosystem diversity using the Shannon Diversity Index.

🔎 Key Steps:
1️⃣ Extracted provincial boundaries of Pakistan from the FAO GAUL dataset.
2️⃣ Clipped the ESA WorldCover 2020 global landcover dataset to the provinces.
3️⃣ Computed the Shannon Diversity Index for each province to measure landcover heterogeneity.
4️⃣ Visualized results with:

WorldCover Map: Distribution of forests, croplands, built-up areas, water bodies, etc.

Shannon Diversity Map (Choropleth): Provinces ranked by ecosystem landcover diversity.
5️⃣ Added interactive charts showing:

Shannon Diversity Index per province.

Landcover class areas (hectares) across Pakistan.
6️⃣ Designed a dual-panel map interface with legends, clickable province info, and comparative visualization.

📊 Findings:

Provinces with more mixed landcover types (e.g., forest, cropland, rangeland) have higher Shannon Index values, reflecting greater ecological diversity.

Provinces dominated by one or two landcover classes exhibit lower Shannon Index scores, highlighting limited landcover heterogeneity.

⚡ Why it matters?
Understanding landcover diversity is crucial for biodiversity conservation, sustainable land-use planning, and ecosystem service management. The Shannon Index provides a quantitative way to compare regions and track changes over time.

🛰️ Tools Used:

Google Earth Engine (JavaScript API)

FAO GAUL boundaries

ESA WorldCover v200 (2020)

👉 This workflow can be extended to district-level analysis, multi-year comparison, or integrated with climate/soil datasets for deeper ecological insights.

🌍 Soil Erosion Risk Mapping in Gilgit (USPED Model – 2024 Season)

This study integrates Sentinel-2 (NDVI), SRTM DEM, and the USPED (Unit Stream Power Erosion Deposition) model within Google Earth Engine (GEE) to evaluate erosion risks in Gilgit, Pakistan.

🔎 Workflow Highlights

1️⃣ NDVI → C Factor: Vegetation cover translated into soil protection levels.

2️⃣ DEM → LS Factor: Terrain slope and flow accumulation quantified topographic influence.

3️⃣ USPED Erosion: Combined rainfall erosivity (R), soil erodibility (K), C and LS factors.

4️⃣ Classification: Results categorized into 6 erosion risk levels (No, Low, Moderate, High, Very High, Extreme).

5️⃣ Visualization: Multi-map layout showing LS Factor, Erosion, Erosion Classes, and C Factor with legends.

6️⃣ Quantification: Area statistics calculated and displayed via bar chart.

📊 Key Insights

Areas with steep slopes & low vegetation show high to extreme erosion risks.

Vegetated zones (high NDVI) contribute to lower erosion rates.

Results highlight priority regions for soil conservation and watershed management.

🛰️ Tools: Google Earth Engine + Sentinel-2 + SRTM DEM

📍 Region: Gilgit, Pakistan

📅 Season: May – September 2024

✨ These results support sustainable land management and can guide erosion control practices in fragile mountain ecosystems.