Research
Research Topics
Hydroscience
Our focus is on providing an accurate description of the links between water, energy, and biogeochemical cycles and atmospheric, ground, and vegetation states to advance the predictability of transient hydrologic processes leading to extreme events (e.g. droughts and flooding). We use an extensive set of tools from in-situ eddy covariance and weather stations, reflectometry soil moisture stations (both continuous and grid spatially distributed), infiltrometers, tensiometers, flowmeters, vegetation sap-flow sensors, field lysimeters, and others combined with remotely-sensed imagery from radars, radiometers, multi-spectral and thermal cameras for terrain and atmospheric property measurements. We combine all this field capacity with a digital infrastructure consisting of software and hardware capable of conducting telemetric missions and post-process data with celerity. Our team has extensive computation capability at ³ÉÈËÍ·Ìõ supercomputer center and 3D printing for sensor design, mounting, and deployment.
Some sub-research topics within this realm are:
- Surface and Root-Zone Soil Moisture and Links with Life Processes, Droughts, and Floods: We seek to understand the processes that control the Spatio-temporal variability and predictability of soil moisture across the globe and the atmospheric, climatic, vegetative and material characteristics and mechanisms that are intrinsically linked to its variability. Our team is conducting multi-sensor observations of surface and root-zone soil moisture from continental to pixel-size distribution array. We account for profiling Delta-T and Campbell Sci soil profiling sensors and a miniaturized dual-polarization L-Band radiometer on board of a copter and process-based, hyper-resolution modeling for multiple-source comparison of estimations.
- Surface Energy Budget, Evapotranspiration, and Decarbonization: We investigate the partitioning of the surface available energy from the global to the plot scales in rural and urban environments and under climate and land cover change conditions. We use a multi-physics model approach that accounts for precipitation (including snow), atmospheric forcing, vegetation cover, type and activity, soil texture and depth, soil moisture storage, and re-distribution, vadose zone energy and water fluxes, vegetation interception and shading, albedo and remote effects on energy sheltering, canopy cover and light extinction, atmospheric and stomatal resistance, soil infiltration and runoff generation mechanisms and all relevant processes at the element scale (e.g. 1 m) to be able to provide estimations of the soil surface and root-zone soil moisture and temperature, sensible heat flux, ground heat flux, latent heat flux. With the help of in-situ eddy correlation, sap flow, and lysimeters stations we provide an accurate description of vegetation processes including transpiration, net, and gross ecosystem productivity, and carbon flows. Information links from multiple observations and modeling outputs are analyzed with machine learning tools.
- Water Resources under Climate Change: Our team seeks to understand and improve our predictions of water stocks and fluxes under an ever-changing environment. We focus on the combination of general circulation models (GCM) under a variety of representative concentration pathways (RCP) with distributed hydrologic and social modeling to understand the complex interactions and drivers of future availability or unavailability of water in different regions of the world. The modeling effort involves water reservoirs and dam operation strategies as well as societal attitudes to water conservation within an optimization framework strategy.
- Flood Forecasting: The mechanisms that produce floods and flash-floods in flat and complex terrain are studied and understood from a mechanistic perspective to be able to find causality and improve predictability. Precipitation from satellite and radar forecasting models are blended with physics-based, distributed high resolution modeling to understand the ways water distributes and triggers enhanced runoff mechanisms in different terrain and vegetation coverage types. Super-computing and cutting-edge visualization tools are used to provide the best picture of the causal factors and limits of predictability.
- Post-fire Hydrology: Wildfires are the new normal in the U.S and the world, in part due to climate change. Through this line of work, we are unveiling the processes that rule watersheds’ response to storm events under new soil and vegetation conditions after wildfires. Since short and longwave surface energy and water fluxes change after the fire, a new, very transient hydrologic condition develops. The new condition is highly unpredictable by the previously-calibrated or steady-state hydrologic models that remiss the changes in terrain and canopy cover. We work to understand the links between fire intensity and coverage from remote sensing and observable macro and microscopic soil and vegetation properties that affect flows of water and soil. The end goal is to develop adequate prediction tools that adapt to the changes triggered by a fire in their boundary and initial conditions.
-
Full-Scale laboratory model of sandbars, slope stability, and mass failures: We study erosion of river sandbars that examines the impact of dam operation criteria on the bank slope stability leading to mass failure through seepage. This research work is developed by means of a full-scale laboratory experiment and simulates the sandbar beaches of the Lower Colorado River Basin in Marble and Grand Canyons. These bars are considered an important resource for this riverine ecosystem that provides habitat for native fish species, a substrate for riparian vegetation, an aeolian source of fine sediment which preserves archaeological sites and campsites for river rafters. Findings reveal that there is not a significant correlation between diurnal stage fluctuation caused by the Glen Dam operation criteria and slope stability. The erosion of sandbars with a slope greater than the equilibrium slope is inevitable and after the bars reach the equilibrium slope, mass failure and seepage erosion cease. These findings are essential for decision-makers and scientists to restrict the Glen Canyon Dam operation rule.
Autonomous Systems Applied to Water Mediated Environments
We develop autonomous boats, Unmanned Aerial Systems (UAS) to study fluvial geomorphology and earth science. The use of autonomous systems significantly reduced the time of data collection and human power hours, compared to traditional methods, becoming an alternative for collecting and studying river systems and water bodies. We are currently leading a project that consists of the development of an autonomous boat, using an Ardupilot platform, integrated with a multibeam sonar to measure bathymetry, velocity, and discharge in rivers. We implement a cost-effective alternative to retrieve submerged topography and quantify water storage in river systems and reservoirs using UAS and echo-sounding. Structure from Motion (SfM) Photogrammetry technique is used to reconstruct the DEMs by overlapping multiple 2D image sequences acquired from different viewpoints and a small single.
This research also innovates in the utility of adaptive path planning and sampling to improve decision making and information compression into intelligent systems for effective and efficient spatial reconnaissance. The research objectives are tackled through computer modeling with real-world observations and then tested in our robotic systems for surveying submerged features under mission time, energy, memory, and global positioning system constraints. This data collection also serves as secondary data to construct the computational domain and to validate physically-based models.
Large Eddy Simulation (DES) Models of Turbulence, Sediment, and Riverbed Evolution
We are researching the physics of turbulent flow using a physically-based, eddy-resolving model in large-scale rivers. In this study, we developed one of the first massively parallel, high-performance computational models at the river reach scale. We applied Large Eddy Simulation (LES) techniques in which turbulence structures are directly calculated. This type of analysis gives physical insight into the turbulent flow patterns and coherent turbulence structures generated at a free shear layer downstream of flow separation zones. Our results show that these turbulence structures play a fundamental role in controlling the deposition and erosion rates at the riverbed and banks. Sharp meanders, channel constrictions, many engineering structures, vegetation, and certain types of bedforms all-cause flow separation, secondary circulation, and free shear layers. One of the gaps in the scientific knowledge of fluvial systems is the low predictive capabilities of some available three and two-dimensional quasi-steady and steady models in complex river settings. This three-dimensional, parallelized, turbulence resolving model can capture the instantaneous turbulent structures, sediment transport, and geomorphologic changes. Results show realistic predictions of the geomorphologic changes in complex river systems featured by secondary flows and massive flow separation.
Video 1. Detached Eddy Simulation of a 1.4-km transect of the Colorado River along Grand Canyon, Arizona. It is shown instantaneous contours of Q-criterion displayed by the velocity magnitude taken during 1000 seconds of simulation. This three-dimensional eddy-resolving model was developed in OpenFOAM environment.
Video 2. This video is simulated using an eddy-resolving model developed in OpenFOAM. The non-hydrostatic component of the pressure is also shown. The length of mean surface velocity vectors ranges from 0 to 4.5 m/s, and the length of the near-bed velocity vectors are scaled five times
Global Precipitation Measurement and Predictions
We are helping design the newest precipitation retrieval algorithms of the U.S. National Weather Service radars and satellites from radar echoes and brightness temperatures through the use of data science and machine learning tools. With the help of learners, we are identifying common error structures from a sea of available data and simultaneous atmospheric regime and ground observations. Super-computing modeling and data handling is conducted in this research.
TIN-based Real-time Integrated Basin Simulator (tRIBS)
tRIBS is a physically-based distributed hydrological model developed at the Ralph M. Parsons Laboratory, Massachusetts Institute of Technology. It uses spatially distributed atmospheric, land cover, soil, and topographic parameters for each Voronoi cell to reproduce the energy, and water distribution within the basin. A Triangular Irregular Network (TIN) represents accurately the terrain variability, vegetation, and soil distribution. More information could be found on the website ().
Visual tRIBS
VisualtRIBS is developed by Tri Pham utilizing Python and tested using benchmark simulations on the model website (). The current version of the software has the capability to visualize all the model outputs and export the figures to user-desired format from serial and parallel simulations. Source code and a pre-compiled version of the software are hosted here for public access. The pre-compiled version runs in a GNU/LINUX operative system, while the webmaster source code can be run on your local machine creating an environment in Python with the required packages using Conda or Miniconda. The webmaster version of visual tRIBS is also hosted in a Heroku server ().
Web Master Version
Once downloaded the VisualtRIBS_WebBases-master.zip file, move to the desired location and unzip it, the folder is around 520 kB. If you have already Anaconda or miniconda in your computer, just follow the VisualtRIBS user guide document.
Examples from the program are illustrated using a 1998 simulation at Peacheater Creek, a tributary to the Illinois River in northeastern Oklahoma. The following image is the Actual Evaporation at Peacheater Creek, OK with Red-Blue color scheme.
Pre-compiled Version (GNU/LINUX)
Download the five (5) VisualtRIBS_Precompiled zip files and move them to the desired folder. Then, open a terminal to combine the files into a single zip file.
Then unzip the single zip, and the result will be a 285.3 MB folder.
unzip VisualtRIBS_Precompiled_all.zip
Open the folder and double-click on VisualTribsv1.1 executable file. Then, the pre-compiled version of Visual Tribs is ready to use (see image below):