Brian E. Skahill

Advanced Concepts on Remote Sensing of Precipitation at Multiple Scales

ADVANCED CONCEPTS ON REMOTE SENSING OF PRECIPITATION AT MULTIPLE SCALES by S oroosh S orooshian , A mir A gha K ouchak , P hillip A rkin , J ohn E ylander , E fi F oufoula -G eorgiou , R ussell H armon , J an M. H. H endrickx , B isher I mam , R obert K uligowski , B rian S kahill , and G ail S kofronick -J ackson Overview of Recommendations (i) Uncertainty of merged products and multisensor observations warrants a great deal of research. Quantification of uncertainties and their propa- gation into combined products is vital for future development. (ii) Future improvements in satellite-based precipi- tation retrieval algorithms will rely on more in- depth research on error properties in different climate regions, storm regimes, surface condi- tions, seasons, and altitudes. Given such infor- mation, precipitation algorithms for retrieval, AFFILIATIONS : S orooshian , A gha K ouchak , I mam —University of California, Irvine, Irvine, California; A rkin —University of Maryland, College Park, Maryland; E ylander —U.S. Army Engineer Research and Development Center, Hanover, New Hampshire; F oufoula -G eorgiou —University of Minnesota, Minneapolis, Minnesota; H armon —Army Research Laboratory, Durham, North Carolina; H endrickx —New Mexico Tech, Socorro, New Mexico; K uligowski —NOAA/NESDIS/ STAR, Camp Springs, Maryland; S kahill —U.S. Army Engineer Research and Development Center, Vicksburg, Mississippi; S kofronick -J ackson —NASA GSFC, Greenbelt, Maryland CORRESPONDING AUTHOR : Soroosh Sorooshian, Department of Civil & Environmental Engineering, University of California, Irvine, Irvine, CA 92697 E-mail: soroosh@uci.edu DOI:10.1175/2011BAMS3158.1 In final form 18 April 2011

More efficient PEST compatible model independent model calibration

This article describes some of the capabilities encapsulated within the Model Independent Calibration and Uncertainty Analysis Toolbox (MICUT), which was written to support the popular PEST model independent interface. We have implemented a secant version of the Levenberg-Marquardt (LM) method that requires far fewer model calls for local search than the PEST LM methodology. Efficiency studies on three distinct environmental model structures (HSPF, FASST, and GSSHA) show that we can find comparable local minima with 36-84% fewer model calls than a conventional model independent LM application. Using the secant LM method for local search, MICUT also supports global optimization through the use of a slightly modified version of a stochastic global search technique called Multi-Level Single Linkage [Rinnooy Kan, A.H.G., Timmer, G., 1987a. Stochastic global optimization methods, part I: clustering methods. Math. Program. 39, 27-56; Rinnooy Kan, A.H.G., Timmer, G., 1987b. Stochastic global optimization methods, part ii: multi level methods. Math. Program. 39, 57-78.]. Comparison studies with three environmental models suggest that the stochastic global optimization algorithm in MICUT is at least as, and sometimes more efficient and reliable than the global optimization algorithms available in PEST.

Advancing the remote sensing of precipitation

Author(s): Sorooshian, S; Aghakouchak, A; Arkin, P; Eylander, J; Foufoula-Georgiou, E; Harmon, R; Hendrickx, JMH; Imam, B; Kuligowski, R; Skahill, B; Skofronick-Jackson, G | Abstract: Satellite-based global precipitation data has addressed the limitations of rain gauges and weather radar systems in forecasting applications and for weather and climate studies. Inspite of this ability, a number of issues that require the development of advanced concepts to address key challenges in satellite-based observations of precipitation were identified during the Advanced Concepts Workshop on Remote Sensing of Precipitation at Multiple Scales at the University of California. These include quantification of uncertainties of individual sensors and their propagation into multisensor products warrants a great deal of research. The development of metrics for validation and uncertainty analysis are of great importance. Bias removal, particularly probability distribution function (PDF)-based adjustment, deserves more in-depth research. Development of a near-real-time probabilistic uncertainty model for satellitebased precipitation estimates is highly desirable.

Gridded Surface Subsurface Hydrologic Analysis Modeling for Analysis of Flood Design Features at the Picayune Strand Restoration Project

Abstract : The Picayune Strand Restoration Project is one of many components of the Comprehensive Everglades Restoration Project (CERP) intended to restorenearly 700 hectares of a failed residential development in southwestern Collier County, FL, to its predevelopment wetland conditions. A detailed analysis was performed to derive a restoration plan that will achieve this goal. As required by the Water Resources Development Act (WRDA) 2000,the U.S. Army Corps of Engineers (USACE) is required to ensure that no component of CERP results in an effective taking of land by adversely impacting the level of flood protection of adjacent landowners. To ensure the current level of flood protection is maintained, a hydrologic model was developed to assess the potential for flooding and to refine the proposed flood mitigation features. The USACE physically based Gridded Surface Subsurface Hydrologic Analysis (GSSHA) model was selected for this effort. The GSSHA model simulates fully coupled rainfall distribution, extraction, retention, overland flow, and one-dimensional channel flow. Models of varying resolution were developed from existing and proposed design data and were initially populated with parameter values from a previous hydrodynamic modeling effort. Parameters were then tuned to observed stage and flow data using the Secant Levenberg-Marquardt method, a nonlinear least squares minimization computer-based local search method. The calibrated model is capable of reproducing canal flows, canal stages, and overland stages with very high Nash Sutcliffe Forecast Efficiencies, generally 0.9 or higher. Subsequent uncertainty analysis allowed water stages to be estimated with 95% certainty. Modeling and uncertainty analysis results allowed for refinement of the proposed flood mitigation features.

The Effect of Soil State Predictions on Soil Strength

Soil strength depends on the state of the ground as well as soil type. We have developed both a 1-D and pseudo 3-D SVAT (Soil-Vegetation-Atmosphere Transfer) model, FASST (Fast All-season Soil STrength). FASST predicts soil temperature, moisture, ice content and strength as a function of depth as well as snow accumulation/depletion and surface icing. We implemented a series of parametric tests with the 1-D model to develop the understanding necessary to predict soil strength as a function of soil type, soil hydraulic properties and strength parameterization.