13.3 Integration of remote sensing and GIS in hydrological modeling

Remote sensing and GIS integration revolutionizes hydrological modeling. By combining satellite data with powerful spatial analysis tools, we can create more accurate, detailed models of water systems. This tech duo helps us track everything from soil moisture to snow cover across vast areas.

The magic happens when we mix different data types and scales. We can zoom in on a small stream or pan out to study entire river basins. This flexibility lets us tackle local water issues and global climate change impacts with the same toolset.

Remote sensing and GIS integration for hydrological modeling

Synergistic benefits of integration

Top images from around the web for Synergistic benefits of integration

• 

  GMD - DECIPHeR v1: Dynamic fluxEs and ConnectIvity for Predictions of HydRology View original

  Is this image relevant?

• 

  Determination of Potential Runoff Coefficient Using GIS and Remote Sensing View original

  Is this image relevant?

• 

  Recharge Potentials Mapping using Remote Sensing and GIS Techniques: Case of Shallow Aquifers in ... View original

  Is this image relevant?

• 

  GMD - DECIPHeR v1: Dynamic fluxEs and ConnectIvity for Predictions of HydRology View original

  Is this image relevant?

• 

  Determination of Potential Runoff Coefficient Using GIS and Remote Sensing View original

  Is this image relevant?

1 of 3

Top images from around the web for Synergistic benefits of integration

• 

  GMD - DECIPHeR v1: Dynamic fluxEs and ConnectIvity for Predictions of HydRology View original

  Is this image relevant?

• 

  Determination of Potential Runoff Coefficient Using GIS and Remote Sensing View original

  Is this image relevant?

• 

  Recharge Potentials Mapping using Remote Sensing and GIS Techniques: Case of Shallow Aquifers in ... View original

  Is this image relevant?

• 

  GMD - DECIPHeR v1: Dynamic fluxEs and ConnectIvity for Predictions of HydRology View original

  Is this image relevant?

• 

  Determination of Potential Runoff Coefficient Using GIS and Remote Sensing View original

  Is this image relevant?

1 of 3

• Remote sensing provides spatially continuous data on various hydrological parameters (precipitation, evapotranspiration, soil moisture, snow cover) used as inputs for hydrological models
• GIS offers powerful tools for managing, analyzing, and visualizing spatial data, enabling the integration of remote sensing data with other geospatial datasets (topography, land use/land cover, soil properties)
• Integration allows for the development of spatially distributed hydrological models that better represent the heterogeneity of hydrological processes across a landscape
• Remote sensing data can be used to calibrate and validate GIS-based hydrological models, improving their accuracy and reliability

Monitoring and assessment at various scales

• Synergistic use of remote sensing and GIS enables the monitoring and assessment of hydrological processes at various spatial and temporal scales, from local to regional and global levels
  • Local scale: high-resolution remote sensing data (LiDAR, UAV imagery) integrated with GIS for detailed hydrological modeling of small watersheds
  • Regional scale: moderate-resolution satellite data (Landsat, Sentinel) combined with GIS for regional hydrological assessments and water resource management
  • Global scale: coarse-resolution satellite data (MODIS, TRMM) integrated with GIS for global hydrological modeling and climate change impact studies

Methodologies for integrating remote sensing data

Preprocessing and downscaling techniques

• Preprocessing of remote sensing data, including geometric and radiometric corrections, ensures compatibility with GIS datasets and hydrological models
  • Geometric corrections: orthorectification, image registration, and georeferencing
  • Radiometric corrections: atmospheric correction, topographic correction, and bidirectional reflectance distribution function (BRDF) correction
• Downscaling techniques match the spatial resolution of coarse-resolution remote sensing data (precipitation estimates from satellite sensors) with GIS-based hydrological models
  • Statistical downscaling: regression-based methods, weather generators, and machine learning algorithms
  • Dynamical downscaling: regional climate models and high-resolution numerical weather prediction models

Parameter estimation and data assimilation

• Methods for estimating hydrological parameters (evapotranspiration, soil moisture) from remote sensing data need to be developed and integrated into GIS-based hydrological models
  • Evapotranspiration estimation: surface energy balance algorithms (SEBAL, METRIC), vegetation index-based methods (NDVI, EVI), and land surface temperature-based approaches (TSEB)
  • Soil moisture estimation: microwave remote sensing (active and passive), thermal infrared remote sensing, and data fusion techniques (SMAP, SMOS)
• Data assimilation techniques (Kalman filtering, particle filtering) incorporate remote sensing observations into GIS-based hydrological models, improving their performance and reducing uncertainties
  • Kalman filtering: extended Kalman filter (EKF), ensemble Kalman filter (EnKF), and unscented Kalman filter (UKF)
  • Particle filtering: sequential importance resampling (SIR) and Markov chain Monte Carlo (MCMC) methods

Handling missing or cloudy data

• Strategies for handling missing or cloudy remote sensing data (interpolation, data fusion) ensure the continuous availability of input data for GIS-based hydrological models
  • Interpolation methods: inverse distance weighting (IDW), kriging, and spline interpolation
  • Data fusion techniques: multi-sensor data fusion, multi-temporal data fusion, and multi-resolution data fusion

Performance evaluation of integrated approaches

Statistical metrics and validation

• Statistical metrics (root mean square error (RMSE), Nash-Sutcliffe efficiency (NSE), correlation coefficient (R)) assess the agreement between simulated and observed hydrological variables
• Comparison of model outputs with ground-based measurements (streamflow, soil moisture) is essential for validating the performance of integrated remote sensing and GIS approaches
  • Streamflow validation: comparison with gauged river discharge data
  • Soil moisture validation: comparison with in-situ soil moisture measurements from ground-based networks (SCAN, COSMOS)

Sensitivity and uncertainty analysis

• Sensitivity analysis identifies the most influential remote sensing and GIS input parameters on hydrological model outputs, guiding data collection and model refinement efforts
  • Local sensitivity analysis: one-at-a-time (OAT) method and Morris method
  • Global sensitivity analysis: variance-based methods (Sobol' indices) and regression-based methods (standardized regression coefficients)
• Uncertainty analysis techniques (Monte Carlo simulation, Bayesian inference) quantify the uncertainties associated with remote sensing data, GIS datasets, and hydrological model parameters
  • Monte Carlo simulation: random sampling of input parameters and propagation of uncertainties through the model
  • Bayesian inference: posterior probability distributions of model parameters and predictions using Markov chain Monte Carlo (MCMC) sampling

Transferability and robustness assessment

• Transferability of integrated remote sensing and GIS approaches across different regions and scales should be evaluated to assess their robustness and generalizability
  • Cross-validation: leave-one-out (LOO) and k-fold cross-validation techniques
  • Multi-site validation: testing the model performance in different geographical locations and climatic conditions

Case studies of remote sensing and GIS integration

Flood modeling and mapping

• Case studies on the use of remote sensing and GIS for flood modeling and mapping highlight the benefits of integrating high-resolution topographic data, land use/land cover information, and real-time satellite observations for improved flood prediction and risk assessment
  • Integration of LiDAR-derived digital elevation models (DEMs) and high-resolution satellite imagery (WorldView, GeoEye) for urban flood modeling
  • Assimilation of near-real-time satellite precipitation estimates (GPM, TRMM) and radar observations into GIS-based hydrodynamic models for flood forecasting

Drought monitoring and assessment

• Studies on the application of remote sensing and GIS for drought monitoring and assessment demonstrate the value of integrating multi-sensor remote sensing data (precipitation, soil moisture, vegetation indices) with GIS-based hydrological models for early warning and impact assessment
  • Integration of satellite-derived precipitation estimates (CHIRPS, PERSIANN), soil moisture data (SMAP, SMOS), and vegetation indices (NDVI, EVI) for drought severity assessment
  • Assimilation of remote sensing-based drought indicators into GIS-based crop yield models for agricultural drought impact assessment

Groundwater modeling

• Case studies on the use of remote sensing and GIS for groundwater modeling showcase the importance of integrating remote sensing-derived evapotranspiration estimates, GIS-based aquifer properties, and well observations for improved groundwater resource management
  • Integration of MODIS-derived evapotranspiration estimates and GIS-based hydrogeological data for regional groundwater recharge estimation
  • Assimilation of GRACE satellite-derived groundwater storage changes and well observations into GIS-based groundwater flow models for aquifer characterization

Snow hydrology

• Studies on the application of remote sensing and GIS for snow hydrology highlight the benefits of integrating remote sensing-derived snow cover and snow water equivalent data with GIS-based hydrological models for improved snowmelt runoff prediction and water supply forecasting
  • Integration of MODIS-derived snow cover area (SCA) and snow water equivalent (SWE) estimates with GIS-based temperature-index models for snowmelt runoff modeling
  • Assimilation of Sentinel-1 SAR-derived snow depth and Sentinel-2 MSI-derived snow albedo into GIS-based energy balance models for improved snow process representation

Agricultural water management

• Case studies on the use of remote sensing and GIS for agricultural water management demonstrate the value of integrating high-resolution remote sensing data on crop growth and evapotranspiration with GIS-based irrigation scheduling tools for optimizing water use efficiency and crop productivity
  • Integration of Sentinel-2 MSI-derived vegetation indices (NDVI, LAI) and surface energy balance-based evapotranspiration estimates (SEBAL, METRIC) for precision irrigation management
  • Assimilation of UAV-derived high-resolution multispectral imagery and thermal data into GIS-based crop water stress models for site-specific irrigation scheduling