Data Analytics and Multi-scale Predictions


Columbia Water Center is a leader in predicting seasonal hydroclimate forecasts and associated risk analyses. These forecasts are used for a variety of applications, including integrating climate forecasts into water allocation procedures for urban, industrial and agricultural consumers.

It is critical to develop strategies for climate change adaptation and variability through climate scenarios for the short- and long-term that accurately represent uncertainties. Columbia Water Center scientists use numerical and statistical methods that take into account the high dimensionality of the problem. These methods consider spatial variability across a region or the globe, and correctly represent the associated uncertainty inherent to data and models. Columbia Water Center has developed large-scale simulation and optimization decision-making tools that introduce novel water allocation and risk management ideas and are supported by quantitative analysis.

Managing Climate Risks through Long Lead Forecasts



Columbia Water Center collaborates with International Research Institute for Climate and Society (IRI) at Columbia University, which is a major global leader in Climate Risk Management and Sustainable Development from interdisciplinary and applied perspectives. Specifically, for climate risk management, these research groups have pioneered techniques for seasonal to inter-annual climate and hydrologic forecasts. The teams have also developed a participatory and adaptive management system for reservoir operation and planning. This includes a mix of short- and long-term tradable water supply contracts with specified reliability that are derived using seasonal climate forecasts and multi-decadal simulations. These are used as a vehicle for water allocation, and as an insurance mechanism in the event of yield failure. Applications of this approach have been implemented in N. E. Brazil and the Philippines.

Projects and Impact

To respond to ever-increasing globalization and climate risks, Columbia Water Center has engaged in projects around the world that strive to improve prediction and risk models by utilizing available data and quantitative analysis. Learn more about each of the Center’s related projects below:

Predictions, Surface Energy, Plants: Understanding the Hydrologic Cycle

Above_the_CloudsCWC research is working to make fundamental advances in our ability to comprehend the complexities of climate by using probabilistic cloud modeling to properly characterize precipitation and solar radiation.


Neural Network Soil Moisture Retrieval


The surface soil moisture state can be inferred by using remote sensing data from Multi-Wavelength Satellite Observations in a process known as retrieval.Neural Network Soil Moisture Retrieval is able to provide global soil moisture estimates at daily or sub-daily resolution. 


Water-Agriculture-Livelihood Security in India


Comprising a mere 1.57 percent of India’s total geographical area, the state of Punjab produces 12 percent of India’s 234 million tons of food grain, and nearly 40 and 60 percent of the wheat and rice that buffer the nation’s central pool for maintaining food stocks. However, Punjab’s agricultural success is currently threatened by unsustainable irrigation practices and a rapidly                                                                 dropping water table. 

Decadal Prediction and Stochastic Simulation of Hydroclimate Over Monsoonal Asia


Monsoons drive seasonal variations. Yet climate change threatens to disrupt the regular, alternating periods of rain and arid dryness.Managing water scarcity is a critical challenge for many Asian nations with similar climates. 



Utility-based Water Risk Indices

640px-Wasserhahn This project explores opportunities for robust, no regret, decision making for drought standard operating procedures and planning in a practical framework relevant to utilities in the northeastern region of the United States. 



The Everglades


The Everglades is one of the largest and most diverse wetland ecosystems in the United States. The pressures of agriculture and development, however, have greatly degraded the system. Today the Everglades is half the size it was 100 years ago, and many keystone species are threatened.  

Brazil Allocation


The Ceará allocation project provides new methods for season-ahead climate prediction along with techniques for translating and presenting forecasting information to citizen committees to facilitate the most efficient allocation of water and the greatest possible cooperation between different sectors of the economy. 



For additional relevant papers, click here. If you have any questions or would like to correspond with the Columbia Water Center on flood- or disaster-related research, please contact Paulina Concha.