Spatial reconstruction of rainfall fields from rain gauge and radar data
01 Jul 2014-Stochastic Environmental Research and Risk Assessment (Springer Berlin Heidelberg)-Vol. 28, Iss: 5, pp 1235-1245
TL;DR: In this article, Monte Carlo Markov Chain (MCMCMC) algorithms were used to calibrate radar measurements via rain gauge data and make spatial predictions for hourly rainfall, by means of a Bayesian hierarchical framework.
Abstract: Rainfall is a phenomenon difficult to model and predict, for the strong spatial and temporal heterogeneity and the presence of many zero values. We deal with hourly rainfall data provided by rain gauges, sparsely distributed on the ground, and radar data available on a fine grid of pixels. Radar data overcome the problem of sparseness of the rain gauge network, but are not reliable for the assessment of rain amounts. In this work we investigate how to calibrate radar measurements via rain gauge data and make spatial predictions for hourly rainfall, by means of Monte Carlo Markov Chain algorithms in a Bayesian hierarchical framework. We use zero-inflated distributions for taking zero-measurements into account. Several models are compared both in terms of data fitting and predictive performances on a set of validation sites. Finally, rainfall fields are reconstructed and standard error estimates at each prediction site are shown via easy-to-read spatial maps.
Citations
More filters
TL;DR: The recent advances on integration of remotely sensed precipitation and soil moisture with rainfall-runoff models for rainfall-driven flood forecasting are reviewed.
Abstract: Fluvial flooding is one of the most catastrophic natural disasters threatening people’s lives and possessions. Flood forecasting systems, which simulate runoff generation and propagation processes, provide information to support flood warning delivery and emergency response. The forecasting models need to be driven by input data and further constrained by historical and real-time observations using batch calibration and/or data assimilation techniques so as to produce relatively accurate and reliable flow forecasts. Traditionally, flood forecasting models are forced, calibrated and updated using in-situ measurements, e.g., gauged precipitation and discharge. The rapid development of hydrologic remote sensing offers a potential to provide additional/alternative forcing and constraint to facilitate timely and reliable forecasts. This has brought increasing interest to exploring the use of remote sensing data for flood forecasting. This paper reviews the recent advances on integration of remotely sensed precipitation and soil moisture with rainfall-runoff models for rainfall-driven flood forecasting. Scientific and operational challenges on the effective and optimal integration of remote sensing data into forecasting models are discussed.
67 citations
46 citations
Cites background from "Spatial reconstruction of rainfall ..."
...Poorer results for daily comparisons were expected, because precipitation data are by nature zero-inflated (Bruno et al., 2014)....
[...]
TL;DR: In this article, the authors applied the Weather Research and Forecasting (WRF) model with 12 designed parameterisation schemes with different combinations of physical parameterisations, including microphysics, radiation, planetary boundary layer (PBL), land-surface model (LSM), and cumulus parameterisations.
31 citations
TL;DR: In this paper, a detailed validation of precipitation grids based on four factors, that is, station density used for grid construction, grid spatial resolution, station altitude, and climate type, is presented.
Abstract: Funding information Consejería de Educación, Junta de Castilla y León, Grant/Award Number: LE240P18; Ministerio de Ciencia e Innovación, Grant/Award Numbers: CGL2016-78702-C2-1-R, PID2019-108470RB-C22, CGL201 Abstract Gridded precipitation datasets have been developed for data assimilation and evaluation tasks of weather and climate models and for climate analyses. Gridded data uncertainty evaluation is crucial to understand the limitations and feasibility. The development of high-resolution daily gridded precipitation datasets is desirable, but several factors need to be considered, namely rain gauge station availability, their spatial distribution, and orographic and climate characteristics of a study area. Quality assessment of gridded datasets can present difficulties when the influence of these factors is not thoroughly analysed. The main objective of this study was a detailed validation of precipitation grids based on four factors, that is, station density used for grid construction, grid spatial resolution, station altitude, and climate type. To this end, 18 grids were built using six spatial resolutions (0.01 , 0.025 , 0.05 , 0.1 , 0.2 and 0.4 ) and three station densities (25, 50 and 75% of all available stations). Results indicate larger differences among the grids as a function of analysed factors. Station density was found to be the main factor, whereas grid spatial resolution had minor importance. However, the latter factor becomes more relevant in areas with strong altitude gradients and when a high station density is available. In addition, weak and moderate precipitation is overestimated on daily grids, whereas heavy precipitation cells are less frequent, reducing data variability. On the contrary, monthly and annual aggregates present less deviation from the observed distribution than daily comparisons. These findings question the applicability of the daily grid datasets for validation studies and climate analysis on a grid cell level.
21 citations
TL;DR: In this article, the spatial covariance structure of the spectral model is equivalent to the well-known Matern covariance model, and the parameters of the Matern model are estimated using high-quality rain gauge data.
Abstract: It is challenging to model a precipitation field due to its intermittent and highly scale-dependent nature. Many models of point rain rates or areal rainfall observations have been proposed and studied for different time scales. Among them, the spectral model based on a stochastic dynamical equation for the instantaneous point rain rate field is attractive, since it naturally leads to a consistent space–time model. In this paper, we note that the spatial covariance structure of the spectral model is equivalent to the well-known Matern covariance model. Using high-quality rain gauge data, we estimate the parameters of the Matern model for different time scales and demonstrate that the Matern model is superior to an exponential model, particularly at short time scales.
20 citations
Cites methods from "Spatial reconstruction of rainfall ..."
...Much progress has been made for precipitation modeling; Onof et al. (2000) reviewed the development of Poissoncluster processes, Bruno et al. (2014) investigated how to calibrate radar measurements via rain gauge data, Oliveira (2004) constructed separate models for rainfall occurrences and the…...
[...]
References
More filters
TL;DR: In this paper, the authors consider the problem of comparing complex hierarchical models in which the number of parameters is not clearly defined and derive a measure pD for the effective number in a model as the difference between the posterior mean of the deviances and the deviance at the posterior means of the parameters of interest, which is related to other information criteria and has an approximate decision theoretic justification.
Abstract: Summary. We consider the problem of comparing complex hierarchical models in which the number of parameters is not clearly defined. Using an information theoretic argument we derive a measure pD for the effective number of parameters in a model as the difference between the posterior mean of the deviance and the deviance at the posterior means of the parameters of interest. In general pD approximately corresponds to the trace of the product of Fisher's information and the posterior covariance, which in normal models is the trace of the ‘hat’ matrix projecting observations onto fitted values. Its properties in exponential families are explored. The posterior mean deviance is suggested as a Bayesian measure of fit or adequacy, and the contributions of individual observations to the fit and complexity can give rise to a diagnostic plot of deviance residuals against leverages. Adding pD to the posterior mean deviance gives a deviance information criterion for comparing models, which is related to other information criteria and has an approximate decision theoretic justification. The procedure is illustrated in some examples, and comparisons are drawn with alternative Bayesian and classical proposals. Throughout it is emphasized that the quantities required are trivial to compute in a Markov chain Monte Carlo analysis.
11,691 citations
TL;DR: The Global Precipitation Climatology Project (GPCP) version 2 Monthly Precise Analysis as discussed by the authors is a merged analysis that incorporates precipitation estimates from low-orbit satellite microwave data, geosynchronous-orbit-satellite infrared data, and rain gauge observations.
Abstract: The Global Precipitation Climatology Project (GPCP) Version 2 Monthly Precipitation Analysis is described. This globally complete, monthly analysis of surface precipitation at 2.5 degrees x 2.5 degrees latitude-longitude resolution is available from January 1979 to the present. It is a merged analysis that incorporates precipitation estimates from low-orbit-satellite microwave data, geosynchronous-orbit-satellite infrared data, and rain gauge observations. The merging approach utilizes the higher accuracy of the low-orbit microwave observations to calibrate, or adjust, the more frequent geosynchronous infrared observations. The data set is extended back into the premicrowave era (before 1987) by using infrared-only observations calibrated to the microwave-based analysis of the later years. The combined satellite-based product is adjusted by the raingauge analysis. This monthly analysis is the foundation for the GPCP suite of products including those at finer temporal resolution, satellite estimate, and error estimates for each field. The 23-year GPCP climatology is characterized, along with time and space variations of precipitation.
4,951 citations
"Spatial reconstruction of rainfall ..." refers background in this paper
...The precipitation accumulation time is defined according to different specific purposes: yearly or monthly rainfall amounts are used for climate research (Adler et al. 2003), daily and hourly measurements are the starting point for flood monitoring (Cooley et al....
[...]
4,476 citations
"Spatial reconstruction of rainfall ..." refers background in this paper
...A useful index for assessing the appropriateness of rainfall probability predictions is the Brier Score (Brier 1950):...
[...]
Book•
01 Jan 1997
TL;DR: In this article, an advanced-level introduction to geostatistics and Geostatistical methodology is provided, including tools for description, quantitative modeling of spatial continuity, spatial prediction, and assessment of local uncertainty and stochastic simulation.
Abstract: This book provides an advanced-level introduction to geostatistics and geostatistical methodology. The discussion includes tools for description, quantitative modeling of spatial continuity, spatial prediction, and assessment of local uncertainty and stochastic simulation. It also details the theoretical background underlying most GSLIB programs.
4,274 citations
TL;DR: This paper presents a meta-modelling framework that automates the very labor-intensive and therefore time-heavy and expensive process of manually cataloging and estimating the effects of noise in a discrete-time model.
Abstract: 1. Exploratory Data Analysis 2. The Random Functions Model 3. Inference and Modeling 4. Local Estimation: Accounting for a Single Attribute 5. Local Estimation: Accounting for Secondary Information 6. Assessment of Local Uncertainty 7. Assessment of Spatial Uncertainty 8. Summary
3,651 citations