The Routing Application for Parallel computation of Discharge (RAPID, here) is a river routing model. Given surface and groundwater inflow to rivers, this model can compute the flow and volume of water everywhere in river networks made out of many thousands of reaches. The design of RAPID allows it to be adapted to any river network, if given basic connectivity information. RAPID uses a matrix version of the Muskingum method, and has an automated parameter estimation procedure that allows finding optimal model parameters based on available gauge measurements.
This model uses the Fortran programming language and can be run on personal computers, as well as on massively parallel supercomputers, with demonstrated fixed-size parallel speedup. RAPID has the ability to run and/or optimize model parameters on any sub-basin included in its computing domain. If major manmade infrastructures are present on the river network, RAPID allows users to easily substitute upstream flows measured by gauges within its simulations of river flow and its optimization of parameters. If information concerning water withdrawals or return flows is available, RAPID can remove or add the corresponding flows from its computations as well.
The most unique feature of RAPID is its ability to use mapped water bodies (that accurately describe surface hydrography) as computational elements, compared to a more classic and coarser approach that uses a gridded representation of river networks. Such mapped hydrographic datasets were historically based on maps created by government survey agencies from in-situ observations, and are now increasingly generated from remotely sensed observations. Other features that make RAPID singular include its efficient use of parallel computing and its automated parameter estimation procedure. Additionally, RAPID has been open-source from inception, which has motivated increasing community use, and allowed for steady growth of its user base.
The development of RAPID has been ongoing for almost a decade. The main improvement made here (see figure) over previous versions of the software are in code modifications, allowing for the decrease of model setup time from hours to mere seconds when addressing continental-scale applications. This means that RAPID can now start producing results virtually instantly after launching the executable. The lengthy setup times had previously been a major impediment to large-scale applications (continental to global simulations) of RAPID.
In essence, RAPID is designed to help with water management, particularly during periods of flood or drought that are key to water management in many parts of the world, including the western U.S. The capacity to simulate the flow of water in large river networks in a timely manner has implications for water management at continental to global scales.
This work was done by Cédric H. David of Caltech for NASA's Jet Propulsion Laboratory. This software is available for license through the Jet Propulsion Laboratory, and you may request a license at: here. NPO-49809