Optimizing the Indus Basin Irrigation System and reservoir operations using remotely sensed snow surface properties in the ParBal model

This project is funded by NASA Research Opportunities in Space and Earth Sciences NNH16ZDA001N-GEO A.50 Group on Earth Observations Work Programme


Karl Rittger (NSIDC), Edward Bair (UCSB), Mary J. Brodzik (NSIDC), William Doan (USACE)

Project Summary

This project will use sophisticated research algorithms to create essential water variables (EWVs) for snow and glacier ice. The EWVs will be analyzed as indicators to long-term trends in the Indus River basin that spans the countries of Pakistan, Afghanistan, India, and China presenting transboundary issues. We expect to generate, archive, and distribute from the University of Colorado Boulder PetaLibary the EWVs including but not limited to snow cover fraction, snow albedo, and snow water equivalent (SWE) based on remote sensing and modeling. We will use optical remote sensing to derive snow surface properties relying on spectral mixture analysis instead of simpler normalized difference snow index products. We will reduce the uncertainty in these remotely sensed products using a filtering system developed at the National Snow and Ice Data Center to improve the daily estimates from the Moderate Resolution Imaging Spectrometer (MODIS) removing issues related to misidentified cloud cover, off-nadir views, and other data errors. The system produces fully gridded complete time series of snow surface properties and separates these into snow on ice, exposed glacier ice and snow on land for a better understanding of surface water melt inputs to streamflow.

Of the EWVs, estimates of SWE are considered most useful because they relate snow pack to water quantity, an important input for water managers for both distribution and planning purposes. We have developed the Parallel Energy Balance model (ParBal) that can be used along with satellite based EWVs to estimate maximum seasonal SWE and hourly snow and ice melt in mountainous terrain without the use of in situ observations. We will combine historical ParBal estimates of SWE with a new high-resolution passive microwave data set and physiographic variables using an innovative neural network to create near-real time (NRT) estimates of snow and ice melt. These melt estimates will serve as input into the US Army Corps of Engineers (USACE) Hydrologic Engineering Center’s Hydrologic Modeling System (HEC-HMS) Model. These in turn will provide reservoir inflow hydrographs to the HEC-ResSim Model reservoir simulation, which is specifically designed to duplicate and improve on the existing water management system comprising the Tarbela and Mangla Reservoirs, two major water resources in Pakistan.

Our proposed work to support the NASA Earth Sciences Division and Applied Sciences will connect this research with a current government-to-government effort between the United States Agency for International Development (USAID), USACE, and the Government of Pakistan (GoP) to provide actionable information for water resource planning. That collaboration currently focuses on flood stage forecasting, disaster risk reduction and reservoir management by linking water managers in the US who operate major US water systems and develop state-of-the-art water resource software with their Pakistani counterparts. We will focus on providing indicators for long-term trends of snow and ice and the most accurate NRT estimates of snow and ice melt to help provide better water management in Pakistan. This project will support Pakistan in providing internal and regional stability by advancing their development and management of critical water resources. The 2017-2019 GEO Work Programme identifies EWVs as part of the (GEOGLOWS) framework, and GEOGLOWS leadership has identified snow cover and snow pack as primary EWVs in activities for 2016. The focus on EWVs in this project fulfills the GEOGLOWS work program framework coordinating water through links in data, information, knowledge, applications, and policy. Our application of snow and ice satellite Earth observations to the cross-cutting phenomena of climate and weather address the GEO’s Societal Benefit Areas through the government-to-government partnership in optimizing the Tarbela and Mangla reservoirs and the Indus River Basin Irrigation System (IBIS).


Bair, E. H., A. Abreu Calfa, K. Rittger, and J. Dozier. 2018. Using machine learning for real-time estimates of snow water equivalent in the watersheds of Afghanistan. The Cryosphere, 12(5), 1579-1594, doi:10.5194/tc-12-1579-2018.

Data Access

Contact Karl Rittger at Karl.Rittger@colorado.edu.