Dong-jun Seo

Kalman filter (KF) and its variants and extensions are wildly used for hydrologic prediction in environmental science and engineering. In many data assimilation applications of Kalman filter (KF) and its variants and extensions, accurate estimation of extreme states is often of great importance. When the observations used are uncertain, however, KF suffers from conditional bias (CB) which results in consistent under- and overestimation of extremes in the right and left tails, respectively. Recently, CB-penalized KF, or CBPKF, has been developed to address CB. In this paper, we present an alternative formulation based on variance-inflated KF to reduce computation and algorithmic complexity, and describe adaptive implementation to improve unconditional performance. For theoretical basis and context, we also provide a complete self-contained description of CB-penalized Fisher-like estimation and CBPKF. The results from one-dimensional synthetic experiments for a linear system with vary...

Hydrologic predictions are subject to various sources of error due to uncertainties in the atmospheric forcing observations and predictions, hydrologic model initial conditions, parameters and structures, and streamflow regulations. To allow risk-based decision making in water resources and emergency management, quantification of predictive uncertainty in streamflow forecasts across short, medium and long ranges is necessary. To obtain reliable predictive uncertainty, it is necessary to account for both input (i.e. atmospheric) and hydrologic uncertainties accurately. To provide uncertainty-quantified streamflow forecast products operationally, the National Weather Service (NWS) Office of Hydrologic Development (OHD) and its partners have been developing a prototype hydrologic ensemble forecast system, the EXperimental Ensemble Forecast System (XEFS). The principal components of the prototype system are currently implemented in the Community Hydrologic Prediction System (CHPS). Test...

ABSTRACT The National Weather Service of the U.S. National Oceanic and Atmospheric Administration (NOAA/NWS) recognizes the need for a continued series of science experiments to guide its research into advanced hydrologic models for river and water resources forecasting. This need is accentuated by NOAA/NWS' recent advance into a broader spectrum of water resources forecasting, to complement its more traditional river and flash flood forecasting mission. To this end, the NOAA/NWS welcomes input and contributions from the national and international hydrologic research community in order to better fulfill its mandate to provide valuable products and services. In February of 2006, the Hydrology Laboratory (HL) of the NOAA /NWS Office of Hydrologic Development launched the second phase of the Distributed Hydrologic Model Intercomparison Project (DMIP 2). DMIP 2 follows the success of DMIP 1, which was completed in 2002. DMIP 1 provided a venue for researchers to compare their models with others and with NWS operational models, given a common set of forcings and verification data. The project attracted participants from 12 institutions based in Denmark, China, Canada, New Zealand, and the U.S. The experiments in DMIP 1 focused on the comparison of lumped and distributed models in hydrologically-simple regions. Models were forced with data used for NWS operational river forecasting. DMIP 1 results were mixed: in some basins, distributed models performed better than the NWS lumped model; in other cases the opposite was true. The DMIP 1 results were formally presented in a special issue of the Journal of Hydrology (Vol. 298, 2004). The scope of DMIP 2 is broader than that in DMIP 1 and is designed around two themes: 1) continued investigation of science questions pertinent to the DMIP 1 test sites, and 2) distributed and lumped model tests in hydrologically complex basins in the Sierra-Nevada mountains in the western U.S. DMIP 2 will benefit from data available from the Oklahoma Mesonet and an intense instrumentation effort in one of the Sierra- Nevada basins. Key science questions to be addressed in DMIP 2 include the following: What is the value of soil moisture observations in the validation of distributed models? How do distributed and lumped models perform given forecast estimates of precipitation? Can distributed models provide improved simulations in mountainous areas given current model forcings? Can appropriate observational network densities be defined in mountainous areas that will lead to improved simulations and forecasts? Can new observations of the rain/snow division provide improved streamflow simulations? Can existing remote-sensor observing platforms be better utilized in providing precipitation estimates in mountainous areas? In this presentation we will review the status of DMIP 2. In addition, we will briefly present updated results of HL's lumped and distributed models for the Oklahoma basins. In particular, we expect to show the effects of using an archive of more recent radar precipitation data for model calibration and verification than was used in DMIP 1. The DMIP 1 precipitation data suffered from known biases, especially in the early part of the calibration period.

ABSTRACT The National Climatic Data Center (NCDC) and the National Weather Service (NWS) have implemented the NWS operational Multi-sensor Precipitation Estimation (MPE) algorithm with the historical NEXRAD data, the Digital Precipitation Array (DPA) products, in a reanalysis mode to develop a data set that is suited for long term climatological applications. The reanalysis is set up in a pilot domain over North and South Carolina for a 10 year period (1996-2007) and includes six WSR-88D sites. In this study we provide an evaluation of the multi-sensor precipitation reanalysis (MPR) over this region. In addition we provide comparisons with the operational Stage IV multi-sensor precipitation estimate. The evaluation of the MPR includes rain gauge (point) and radar-rainfall (pixel) comparisons at several temporal scales. A high density network from the Charlotte-Mecklenburg area (USGS) is used as it has a high temporal resolution (5-min) with a long period of record. Other rain gauge networks are from the North Carolina Mesonet and the U.S Climate Reference Network (USCRN). We present results of this evaluation via standard statistics, i.e. correlation coefficient, bias, and mean squared error. Another method of evaluation presented includes the mean squared error decomposition. In addition, we will investigate non-standard methods of evaluation such as Hovmoller diagrams, gridded correlation functions, and time series analysis.

The Range Correction Algorithm (RCA) and its companion, the Convective Stratiform Separation Algorithm (CSSA), have been developed by the Office of Hydrologic Development (OHD), NOAA's National Weather Service (NWS). The algorithms mitigate ...

This study performed an inter-comparison analysis for evaluation of multi-level products of the radar-based multi-sensor precipitation estimation (MPE) algorithm. The main objective of this study was to provide the user community and the algorithm developers with some insight on the potential value of increasing degrees of complexities in the estimation algorithm in terms of bias removal and optimal merging with gauge observations. Seven different MPE products were considered for inter-comparison: a gauge-only product (GAGEONLY), a radar-only product (RMOSAIC), a mean-field bias adjusted product (MOSAIC), a local bias-adjusted product (LMOSAIC), two other products that are based on merging the bias-adjusted products with gauge observations (MMOSAIC and LMOSAIC), and a final product (XMRG) that represents the forecaster's selection of one or more of the intermediate products for hydrologic operational purposes. Surface rainfall observations from an independent dense high-quality...

ABSTRACT This paper presents a comprehensive intercomparison analysis of different radar-based multisensor precipitation products generated operationally by the National Weather Service (NWS) Multisensor Precipitation Estimator (MPE) algorithm from the Weather Surveillance Radar-1988 Doppler version and concurrent rain gauge data. The analysis provides close insight into different effects of the increasing degree of complexitzy in the MPE algorithms. First, a gauge-only product produced by the MPE algorithm was assessed. Then six MPE products were analyzed: a radar-only product, a mean-field bias-adjusted product, a local bias-adjusted product, two products that are based on merging the bias-adjusted products with gauge observations, and a final product that includes human intervention by NWS forecasters. Data from a dense, carefully maintained experimental rain gauge cluster are used as an independent surface reference. A number of summary and conditional statistics are applied to the product intercomparisons. The results reported in this paper show that the most effective improvement of the rainfall products comes from applying the mean-field bias adjustment to the radar-only product. The analysis demonstrates that, for the current study site, some best-intended schemes for the optimal merging of radar and rain gauge data processing did not necessarily lead to clear improvements and, in some respects, caused accuracy degradation in the final products. This behavior by the MPE merging schemes is possibly attributed to the rather poor density of operational rain gauges that need to be available in real time for the implementation of such schemes. Future research is required to examine whether this behavior persists in other regions that may have better coverage and availability of operational rain gauges. DOI: 10.1061/(ASCE)HE.1943-5584.0000638. (C) 2013 American Society of Civil Engineers.

When there exist catchment-wide biases in the distributed hydrologic model states, state updating based on streamflow assimilation at the catchment outlet tends to over- and under-adjust model states close to and away from the outlet, respectively. This is due to the greater sensitivity of the simulated outlet flow to the model states that are located more closely to the outlet in the hydraulic sense, and the subsequent overcompensation of the states in the more influential grid boxes to make up for the larger scale bias. In this work, we describe Mean Field Bias (MFB)-aware variational (VAR) assimilation, or MVAR, to address the above. MVAR performs bi-scale state updating of the distributed hydrologic model using streamflow observations in which MFB in the model states are first corrected at the catchment scale before the resulting states are adjusted at the grid box scale. We comparatively evaluate MVAR with conventional VAR based on streamflow assimilation into the distributed S...

Hurricane Harvey was one of the most extreme weather events to occur in Texas, USA; there was a huge amount of urban flooding in the city of Houston and the adjoining coastal areas. In this study, we reanalyze the spatiotemporal evolution of inundation during Hurricane Harvey using high-resolution two-dimensional urban flood modeling. This study’s domain includes the bayou basins in and around the Houston metropolitan area. The flood model uses the dynamic wave method and terrain data of 10-m resolution. It is forced by radar-based quantitative precipitation estimates. To evaluate the simulated inundation, on-site photos and water level observations were used. The inundation extent and severity are estimated by combining the retrieved water depths, images collected from the impacted area, and high-resolution terrain data. The simulated maximum inundation extent, which is frequently found outside of the designated flood zones, points out the importance of capturing multi-scale hydrod...

We present a statistical procedure that generates reliable short-term streamflow ensemble forecasts from operational single-value forecasts. Referred to as the Hydrologic Model Output Statistics (HMOS) streamflow ensemble processor, the procedure uses linear regression between observed and forecast streamflows in the normal space that minimizes a weighted sum of the root mean square error and the error in the probability distribution

ABSTRACT For monitoring and prediction of water-related hazards in urban areas such as flash flooding, high-resolution hydrologic and hydraulic modeling is necessary. Because of large sensitivity and scale dependence of rainfall–runoff models to errors in quantitative precipitation estimates (QPE), it is very important that the accuracy of QPE be improved in high-resolution hydrologic modeling to the greatest extent possible. With the availability of multiple radar-based precipitation products in many areas, one may now consider fusing them to produce more accurate high-resolution QPE for a wide spectrum of applications. In this work, we formulate and comparatively evaluate four relatively simple procedures for such fusion based on Fisher estimation and its conditional bias-penalized variant: Direct Estimation (DE), Bias Correction (BC), Reduced-Dimension Bias Correction (RBC) and Simple Estimation (SE). They are applied to fuse the Multisensor Precipitation Estimator (MPE) and radar-only Next Generation QPE (Q2) products at the 15-min 1-km resolution (Experiment 1), and the MPE and Collaborative Adaptive Sensing of the Atmosphere (CASA) QPE products at the 15-min 500-m resolution (Experiment 2). The resulting fused estimates are evaluated using the 15-min rain gauge observations from the City of Grand Prairie in the Dallas–Fort Worth Metroplex (DFW) in north Texas. The main criterion used for evaluation is that the fused QPE improves over the ingredient QPEs at their native spatial resolutions, and that, at the higher resolution, the fused QPE improves not only over the ingredient higher-resolution QPE but also over the ingredient lower-resolution QPE trivially disaggregated using the ingredient high-resolution QPE. All four procedures assume that the ingredient QPEs are unbiased, which is not likely to hold true in reality even if real-time bias correction is in operation. To test robustness under more realistic conditions, the fusion procedures were evaluated with and without post hoc bias correction of the ingredient QPEs.

ABSTRACT The availability of new sources of comprehensive, high resolution physiographical information shifted watershed modeling toward more physically-based approaches. As a result, model parameterization can be more closely linked to basin physical properties. However, because of simulation and measurement scale differences as well as model and input data uncertainties, different transformation functions between measured properties and actual model parameters/constants (so called a priori parameters) are required. Derivation of a priori parameters is critical in distributed hydrological modeling when parameters have to be specified for a large number of computational elements. Two approaches are most commonly used: establishing direct physically-based relationships, and statistical regionalization. Both approaches have disadvantages because of poorly defined physical relationships or non-uniqueness of calibration results. This presentation evaluates these approaches for the Hydrology Laboratory Distributed Modeling System. Parameterization results for distributed water balance and channel routing models suggest that combining these two approaches can improve an estimation procedure.

... The Hydrologic Research Laboratory has developed a prototype algorithm, Flash Flood Potential (FFP), that compares WSR-88D rainfall estmates with "flash flood guidance" (FFG) rainfall that is computed daily by the NWS River Forecast Center's operational hydrologic models ...

... Tuteja, NK, Shin, D., Laugesen, R., Shao, Q., Wang, E., Kuczera, G., Kavetski, D., Evin, G., Thyer, M., Khan, U., Li, M., Zheng, H ... Werner, K., Brandon, D., Clark, M., Gangopadhyay, S. (2004) Climate index weighting schemes for NWS ESP-based seasonal volume forecasts. ...

... Dong-Jun Seo. Department of Civil Engineering, Massachusetts Institute of Technology, Cambridge. ... Citation: Bras, RL, and D.-J. Seo (1987), Irrigation control in the presence of salinity: Extended linear quadratic approach, Water Resour. ...

Indicator cokriging (Journel 1983) is examined as a tool for real-time estimation of rainfall from rain gage measurements. The approach proposed in this work obviates real-time estimation of real-time statistics of rainfall by using ensemble or climatological statistics exclusively, and reduces computational requirements attendant to indicator cokriging by employing only a few auxiliary cutoffs in estimation of conditional probabilities. Due

ABSTRACT Simple and ordinary kriging, or SK and OK, respectively, represent the best linear unbiased estimator in the unconditional sense in that they minimize the unconditional (on the unknown truth) error variance and are unbiased in the unconditional mean. However, because the above properties hold only in the unconditional sense, kriging estimates are generally subject to conditional biases that, depending on the application, may be unacceptably large. For example, when used for precipitation estimation using rain gauge data, kriging tends to significantly underestimate large precipitation and, albeit less consequentially, overestimate small precipitation. In this work, we describe an extremely simple extension to SK or OK, referred to herein as conditional bias-penalized kriging (CBPK), which minimizes conditional bias in addition to unconditional error variance. For comparative evaluation of CBPK, we carried out numerical experiments in which normal and lognormal random fields of varying spatial correlation scale and rain gauge network density are synthetically generated, and the kriging estimates are cross-validated. For generalization and potential application in other optimal estimation techniques, we also derive CBPK in the framework of classical optimal linear estimation theory.

... is that, because the DA run is based on internally updated stated variables, it may not be straightforward to relate to the ... Reed, S., Victor Koren, Michael Smith, Ziya Zhang, Fekadu Moreda, Dong-Jun Seo and DMIP Participants, Overall distributed model intercomparison project ...

... J. Hydrometeor, 4, 627-641. Real-Time Variational Assimilation of Hydrologic and Hydrometeorological Data into Operational Hydrologic Forecasting. Dong-Jun Seo, Victor Koren, and Neftali Cajina Hydrology Laboratory, National Weather Service, Silver Spring, Maryl. Abstract. ...