i-FAMS

FLOOD EXTENT (USING OPTICAL & SAR)

OPTICAL :
Flood extent is calculated using Landsat 8 [ee.ImageCollection('LANDSAT/LC08/C02/T1')] satellite imagery from Google Earth Engine and the NDWI. It filters the imagery to the pre-flood period, computes the NDWI, and applies a threshold to map and visualize the water extent. Visiting period of this satellite is 16 days. Data for this satellite is available from 2012 onwards.

SAR :
Flood extent is calculated using Sentinel-1 SAR data[ee.ImageCollection('COPERNICUS/S1_GRD')] in Google Earth Engine by filtering images based on specific criteria, reducing noise with a median filter, and identifying water bodies using a threshold. Visiting period of this satellite is 12 days. Data for this satellite is available from 2016 onwards.

FLOOD DEPTH (USING FW-DET)

Flood depth calculation is using the FW-DET (Flood Water Depth Estimation Tool)method with ArcPy. It starts by setting the workspace to a geodatabase and loading a DEM and flood extent shapefile. This method calculates the flood boundary through focal statistics and slope filtering. The flood depth is then computed by subtracting the clipped DEM from the cost allocation raster and is further smoothed with a low-pass filter. Data for this calculation is available from 2012 onwards, as it uses optical satellite.

FLOOD AGE (USING HAND METHOD)

It calculates flood age using satellite imagery and elevation data. It identifies water bodies from Landsat images and generates flood extent. This flood extent, alongside elevation data (HAND - Height Above Nearest Drainage), is used to estimate flood duration. The method assumes a fixed duration for the month and scales this estimate based on normalized elevation values. The resulting output is a raster map indicating the estimated duration of flooding across the study area, crucial for flood impact assessment and planning mitigation strategies. Data for this calculation is available from 2012 onwards, as it uses optical satellite.

FLOOD AGE (USING MODIS)

This method leverages Google Earth Engine and geemap to analyze flood events in India sourced from the Global Flood Database (GLOBAL_FLOOD_DB dataset) using MODIS data. It filters and merges flood events, then clips the merged data to a specified shapefile. The resulting map visualizes the aggregated duration of these flood events, offering insight into their spatial distribution and severity in the region. Available data ranges from 2000-2018.

PRECIPITATION (USING GPM DATA)

This processes daily rainfall data from a NetCDF (Network Common Data Form) file sourced from the Global Precipitation Measurement (GPM) dataset. It extracts rainfall values for a specified time range, sets spatial dimensions. The output results are saved and displayed on the map. Additionally, it prepares the GPM data for creating a time series chart to visualize temporal variations in rainfall. Data is available from 2010-2021.

LAND USE LAND COVER (USING ML ALGORITHM)

Google Earth Engine (GEE) is used to classify land use and land cover (LULC) using Sentinel-2 and LANDSAT-8,9 satellite data. It begins by defining a region of interest and a time period. It filters satellite images, applies cloud masking, and scales the data for accurate analysis. Spectral indices like NDVI and NDWI are computed and added to enhance classification accuracy. It then samples labeled data from GOOGLE_DYNAMICWORLD_V1 (2015 onwards) and ESA WorldCover (2020-2021), trains a Random Forest classifier, and produces a classified map.

WATER SHED (USING HYDROSHEDS)

This is a wider card with supporting text below as a natural lead-in to additional content. This content is a little bit longer.