Generalized Extreme Value Model Research Articles

What are the best covariates for developing non-stationary rainfall Intensity-Duration-Frequency relationship?

Present infrastructure design is primarily based on rainfall Intensity-Duration-Frequency (IDF) curves with so-called stationary assumption. However, in recent years, the extreme precipitation events are increasing due to global climate change and creating non-stationarity in the series. Based on recent theoretical developments in the Extreme Value Theory (EVT), recent studies proposed a methodology for developing non-stationary rainfall IDF curve by incorporating trend in the parameters of the Generalized Extreme Value (GEV) distribution using Time covariate. But, the covariate Time may not be the best covariate and it is important to analyze all possible covariates and find the best covariate to model non-stationarity. In this study, five physical processes, namely, urbanization, local temperature changes, global warming, El Niño-Southern Oscillation (ENSO) cycle and Indian Ocean Dipole (IOD) are used as covariates. Based on these five covariates and their possible combinations, sixty-two non-stationary GEV models are constructed. In addition, two non-stationary GEV models based on Time covariate and one stationary GEV model are also constructed. The best model for each duration rainfall series is chosen based on the corrected Akaike Information Criterion (AICc). From the findings of this study, it is observed that the local processes (i.e., Urbanization, local temperature changes) are the best covariate for short duration rainfall and global processes (i.e., Global warming, ENSO cycle and IOD) are the best covariate for the long duration rainfall of the Hyderabad city, India. Furthermore, the covariate Time is never qualified as the best covariate. In addition, the identified best covariates are further used to develop non-stationary rainfall IDF curves of the Hyderabad city. The proposed methodology can be applied to other situations to develop the non-stationary IDF curves based on the best covariate.

Flood frequency analysis for alterations of extreme maximum water levels in the Pearl River Delta

Based on the annual maximum water level record spanning about 60 years at 34 gauging stations, we conduct flood frequency analysis in the Pearl River Delta (PRD) with generalized extreme value (GEV) distribution. The GEV model performs sound estimation with the correlation coefficient at most stations larger than 0.99. To better understand the flood risk especially in ungauged regions, the spatial distribution of flood-stage is displayed for three given return periods, 10-year, 20-year and 50-year. Similar pattern can be identified for different return periods except the difference in magnitude, which generally exhibits a decreasing trend from the upper region to offshore area with a scope of about 2–9m. Additionally, a comparison between pre-1980 and post-1980 is carried out to quantify the flood-stage alteration. The results show that most stations display increasing flood risk except for some stations in the upper estuaries, which experienced slight water level decline. Particularly, the lower part of the PRD is vulnerable to the most severe flood-stage increment. The estimated extreme water level increases by 0.35m (14.96%), 0.59m (21.23%) and 1.06m (29.96%) on average at stations corresponding to 10-year, 20-year and 50-year return period, respectively.

A hierarchical Bayesian GEV model for improving local and regional flood quantile estimates

We estimate local and regional Generalized Extreme Value (GEV) distribution parameters for flood frequency analysis in a multilevel, hierarchical Bayesian framework, to explicitly model and reduce uncertainties. As prior information for the model, we assume that the GEV location and scale parameters for each site come from independent log-normal distributions, whose mean parameter scales with the drainage area. From empirical and theoretical arguments, the shape parameter for each site is shrunk towards a common mean. Non-informative prior distributions are assumed for the hyperparameters and the MCMC method is used to sample from the joint posterior distribution. The model is tested using annual maximum series from 20 streamflow gauges located in an 83,000km2 flood prone basin in Southeast Brazil. The results show a significant reduction of uncertainty estimates of flood quantile estimates over the traditional GEV model, particularly for sites with shorter records. For return periods within the range of the data (around 50years), the Bayesian credible intervals for the flood quantiles tend to be narrower than the classical confidence limits based on the delta method. As the return period increases beyond the range of the data, the confidence limits from the delta method become unreliable and the Bayesian credible intervals provide a way to estimate satisfactory confidence bands for the flood quantiles considering parameter uncertainties and regional information. In order to evaluate the applicability of the proposed hierarchical Bayesian model for regional flood frequency analysis, we estimate flood quantiles for three randomly chosen out-of-sample sites and compare with classical estimates using the index flood method. The posterior distributions of the scaling law coefficients are used to define the predictive distributions of the GEV location and scale parameters for the out-of-sample sites given only their drainage areas and the posterior distribution of the average shape parameter is taken as the regional predictive distribution for this parameter. While the index flood method does not provide a straightforward way to consider the uncertainties in the index flood and in the regional parameters, the results obtained here show that the proposed Bayesian method is able to produce adequate credible intervals for flood quantiles that are in accordance with empirical estimates.

Nonstationary modeling of extreme precipitation in China

The statistical methods based on extreme value theory have been traditionally used in meteorology and hydrology for a long time. Due to climate change and variability, the hypothesis of stationarity in meteorological or hydrological time series was usually not satisfied. In this paper, a nonstationary extreme value analysis was conducted for annual maximum daily precipitation (AMP) at 631 meteorological stations over China for the period 1951–2013. Stationarity of all 631 AMP time series was firstly tested using KPSS test method, and only 48 AMP time series showed non-stationarity at 5% significance level. The trends of these 48 nonstationary AMP time series were further tested using M-K test method. There were 25 nonstationary AMP time series mainly distributed in southern and western China showing significant positive trend at 5% level. Another 5 nonstationary AMP time series with significant negative trends were near northern urban agglomeration, Sichuan Basin, and central China. For these nonstationary AMP time series with significant positive or negative trends, the location parameter in generalized extreme value (GEV) distribution was assumed to be time-varying, and the trends were successfully characterized by the nonstationary GEV models. For the remaining 18 nonstationary AMP time series mainly in the eastern portion of China, no significant trend was detected. The correlation analysis showed that only 5 nonstationary AMP time series were significantly correlated with one or two of the four climate indices EASMI, WPI, SOI, and PDO. Then, the location and scale parameters in the GEV distribution were modeled as functions of the significantly correlated climate indices. The modeling results in this study showed that the nonstationary GEV distributions performed better than their stationary equivalents. Finally, 20-year and 50-year return levels of precipitation extremes at all 631 stations were estimated using the best fitting distribution for the year 1961 and 2013, respectively.

A statistical model for karst spring discharge estimation under extensive groundwater development and extreme climate change

ABSTRACTTo acquire better understanding of spring discharge under extreme climate change and extensive groundwater pumping, this study proposed an extreme value statistical decomposition model, in which the spring discharge was decomposed into three items: a long-term trend; periodic variation; and random fluctuation. The long-term trend was fitted by an exponential function, and the periodic variation was fitted by an exponential function whose index was the sum of two sine functions. A general extreme value (GEV) model was used to obtain the return level of extreme random fluctuation. Parameters of the non-linear long-term trend and periodic variation were estimated by the Levenberg-Marquardt algorithm, and the GEV model was estimated by the maximum likelihood method. The extreme value statistical decomposition model was applied to Niangziguan Springs, China to forecast spring discharge. We showed that the modelled spring discharge fitted the observed data very well. Niangziguan Springs discharge is likely to continue declining with fluctuation, and the risk of cessation by August 2046 is 1%. The extreme value decomposition model is a robust method for analysing the nonstationary karst spring discharge under conditions of extensive groundwater development/pumping, and extreme climate changes.Editor D. Koutsoyiannis; Associate editor J. Ward

Modelling nonlinear trend for developing non‐stationary rainfall intensity–duration–frequency curve

ABSTRACTThe infrastructure design is primarily based on rainfall intensity–duration–frequency (IDF) curves, and the existing IDF curves are based on the concept of stationary extreme value theory (EVT) (i.e. the occurrence probability of extreme precipitation is not expected to change significantly over time). But, the extreme precipitation events are increasing due to global climate change and questioning the reliability of our current infrastructure design. Based on recent developments in the EVT, recent studies proposed a method for developing non‐stationary rainfall IDF curve by incorporating linear trend in the location parameter of the generalized extreme value (GEV) distribution. Upon detecting a significant trend in the extreme rainfall series, directly applying the linear trend to develop non‐stationary IDF curves may increase the bias of the non‐stationary model. Hence, it is important to develop non‐stationary GEV model which has less bias than the stationary model by modelling nonlinear trend in the series.In this study, we try to develop non‐stationary GEV models with less bias by modelling nonlinear trend in the series using multi‐objective genetic algorithm (MOGA). In addition, the proposed GEV model is compared with the stationary GEV model and the linear trend‐based non‐stationary GEV model. Furthermore, the Wilmington city and the Hyderabad city non‐stationary IDF curves are developed and compared with stationary IDF curves. From the study results, it is observed that the proposed MOGA‐based method is able to build the good quality and less bias non‐stationary GEV models by modelling nonlinear trend in the series. In addition, it is also observed that the usage of linear trend for modelling non‐stationarity in the time series sometimes increase the bias of non‐stationary model.

Frequency Analysis of High Flow Extremes in the Yingluoxia Watershed in Northwest China

Statistical modeling of hydrological extremes is significant to the construction of hydraulic engineering. This paper, taking the Yingluoxia watershed as the study area, compares the annual maximum (AM) series and the peaks over a threshold (POT) series in order to study the hydrological extremes, examines the stationarity and independence assumptions for the two series, and discusses the estimations and uncertainties of return levels from the two series using the Generalized Extreme Value (GEV) and Generalized Pareto distribution (GPD) models. For comparison, the return levels from all threshold excesses with considering the extremal index are also estimated. For the POT series, the threshold is selected by examining the mean excess plot and the stability of the parameter estimates and by using common-sense. The serial correlation is reduced by filtering out a set of dependent threshold excesses. Results show that both series are approximately stationary and independent. The GEV model fits the AM series well and the GPD model fits the POT series well. The estimated return levels are fairly comparable for the AM series, the POT series, and all threshold excesses with considering the extremal index, with the difference being less than 10% for return periods longer than 10 years. The uncertainties of the estimated return levels are the highest for the AM series, and next for the POT series and then for all threshold excesses series in turn.

A comparison of U.S. precipitation extremes under RCP8.5 and RCP4.5 with an application of pattern scaling

Precipitation extremes are expected to increase in a warming climate, which may have serious societal impacts. This study uses two initial condition ensembles conducted with the Community Earth System Model (CESM) to investigate potential changes in extreme precipitation under two climate change scenarios over the contiguous United States. We fit non-stationary generalized extreme value (GEV) models to annual maximum daily precipitation simulated from a 30-member ensemble under the RCP8.5 scenario and a 15-member ensemble under the RCP4.5 scenario. We then compare impacts using the 1 % annual exceedance probability (AEP) level, which is the amount of daily rainfall with only a 1 % chance of being exceeded in a given year. Under RCP8.5 between 2005 and 2080, the 1 % AEP level is projected to increase by 17 % on average across the U.S., and up to 36 % for some grid cells. Compared to RCP8.5 in the year 2080, RCP4.5 is projected to reduce the 1 % AEP level by 7 % on average, with reductions as large as 18 % for some grid cells. We also investigate a pattern scaling approach in which we produce predictive GEV distributions of annual precipitation maxima under RCP4.5 given only global mean temperatures for this scenario. We compare results from this less computationally intensive method to those obtained from our GEV model fitted directly to the CESM RCP4.5 output and find that pattern scaling produces reasonable projections.

Atmospheric Predictors for Annual Maximum Precipitation in North Africa

AbstractThe high precipitation variability over North Africa presents a major challenge for the population and the infrastructure in the region. The last decades have seen many flood events caused by extreme precipitation in this area. There is a strong need to identify the most relevant atmospheric predictors to model these extreme events. In the present work, the effect of 14 different predictors calculated from NCEP–NCAR reanalysis, with daily to seasonal time steps, on the maximum annual precipitation (MAP) is evaluated at six coastal stations located in North Africa (Larache, Tangier, Melilla, Algiers, Tunis, and Gabès). The generalized extreme value (GEV) B-spline model was used to detect this influence. This model considers all continuous dependence forms (linear, quadratic, etc.) between the covariates and the variable of interest, thus providing a very flexible framework to evaluate the covariate effects on the GEV model parameters. Results show that no single set of covariates is valid for all stations. Overall, a strong dependence between the NCEP–NCAR predictors and MAP is detected, particularly with predictors describing large-scale circulation (geopotential height) or moisture (humidity). This study can therefore provide insights for developing extreme precipitation downscaling models that are tailored for North African conditions.

Modelling non-stationary annual maximum flood heights in the lower Limpopo River basin of Mozambique

In this article we fit a time-dependent generalised extreme value (GEV) distribution to annual maximum flood heights at three sites: Chokwe, Sicacate and Combomune in the lower Limpopo River basin of Mozambique. A GEV distribution is fitted to six annual maximum time series models at each site, namely: annual daily maximum (AM1), annual 2-day maximum (AM2), annual 5-day maximum (AM5), annual 7-day maximum (AM7), annual 10-day maximum (AM10) and annual 30-day maximum (AM30). Non-stationary time-dependent GEV models with a linear trend in location and scale parameters are considered in this study. The results show lack of sufficient evidence to indicate a linear trend in the location parameter at all three sites. On the other hand, the findings in this study reveal strong evidence of the existence of a linear trend in the scale parameter at Combomune and Sicacate, whilst the scale parameter had no significant linear trend at Chokwe. Further investigation in this study also reveals that the location parameter at Sicacate can be modelled by a nonlinear quadratic trend; however, the complexity of the overall model is not worthwhile in fit over a time-homogeneous model. This study shows the importance of extending the time-homogeneous GEV model to incorporate climate change factors such as trend in the lower Limpopo River basin, particularly in this era of global warming and a changing climate.

A probability distribution model for SCFs in internally ring-stiffened tubular KT-joints of offshore structures subjected to out-of-plane bending loads

The stress concentration factor (SCF) is one of the principal inputs in the fatigue reliability analysis of tubular joints. The SCF shows substantial scatter that highlights the necessity of deriving its governing probability function accurately. In the present research, results from 432 finite element (FE) stress analyses, verified against experimental data, were used to develop a probability distribution model for the SCFs in tubular KT-joints reinforced with internal ring stiffeners subjected to four types of out-of-plane bending (OPB) loads. Based on a FE parametric investigation, a sample database was prepared for the maximum chord-side SCFs of central and outer braces; and density histograms were generated for these samples. Nine different probability density functions (PDFs) were fitted to these histograms. The maximum likelihood method was used to determine the parameters of fitted distributions. In each case, using a developed computer code, Kolmogorov–Smirnov test was applied to assess the goodness of fit. Finally, the Generalized Extreme Value model was selected as the best-fitted distribution and after substituting the values of estimated parameters, six fully defined PDFs were proposed for the maximum chord-side SCFs of central and outer braces in internally ring-stiffened tubular KT-joints subjected to four types of OPB loading.

Evaluation of dynamically downscaled extreme temperature using a spatially-aggregated generalized extreme value (GEV) model

The weather research and forecast (WRF) model downscaling skill in extreme maximum daily temperature is evaluated by using the generalized extreme value (GEV) distribution. While the GEV distribution has been used extensively in climatology and meteorology for estimating probabilities of extreme events, accurately estimating GEV parameters based on data from a single pixel can be difficult, even with fairly long data records. This work proposes a simple method assuming that the shape parameter, the most difficult of the three parameters to estimate, does not vary over a relatively large region. This approach is applied to evaluate 31-year WRF-downscaled extreme maximum temperature through comparison with North American regional reanalysis (NARR) data. Uncertainty in GEV parameter estimates and the statistical significance in the differences of estimates between WRF and NARR are accounted for by conducting a novel bootstrap procedure that makes no assumption of temporal or spatial independence within a year, which is especially important for climate data. Despite certain biases over parts of the United States, overall, WRF shows good agreement with NARR in the spatial pattern and magnitudes of GEV parameter estimates. Both WRF and NARR show a significant increase in extreme maximum temperature over the southern Great Plains and southeastern United States in January and over the western United States in July. The GEV model shows clear benefits from the regionally constant shape parameter assumption, for example, leading to estimates of the location and scale parameters of the model that show coherent spatial patterns.

Statistical Modeling of Extreme Rainfall Processes (SMExRain): A Decision Support Tool for Extreme Rainfall Frequency Analyses

Many probability distributions have been developed to model the extreme rainfall processes. However, there is no general agreement as to which distribution should be used. In the present study, a decision support tool for statistical modeling of extreme rainfall processes (SMExRain) was therefore developed to identify the most suitable distribution that could provide accurate extreme rainfall estimates. More specifically, the proposed tool can be used to assess the descriptive and predictive abilities of ten commonly-used probability models, Beta-K, Beta-P, Generalized Extreme Value (GEV), Generalized Logistic (GLO), Generalized Normal (GNO), Generalized Pareto (GPA), Gumbel, Log-Pearson Type III (LP3), Pearson Type III (P3), and Wakeby, for their accuracy and robustness in the estimation of annual maximum precipitations. The proposed decision support tool was tested using 5-minute, 1-hour, and 24-hour annual maximum precipitation data from a network of 21 raingages located in the Ontario region in Canada. Results based on various numerical and graphical goodness-of-fit criteria have indicated that the GEV, GNO, and P3 models were the best models for describing the distribution of annual maximum precipitations in this region. The GEV distribution, however, was preferred to the GNO and P3 because it requires a simpler parameter estimation method and it was based on a more solid theoretical basis for representing the distribution of extreme random variables. Therefore, the GEV could be recommended as the most suitable model for describing the distribution of annual maximum precipitations in Ontario region based on the proposed decision support tool.

Generalized Extreme Value models for count data: Application to worker telecommuting frequency choices

Count models are used for analyzing outcomes that can only take non-negative integer values with or without any pre-specified large upper limit. However, count models are typically considered to be different from random utility models such as the multinomial logit (MNL) model. In this paper, Generalized Extreme Value (GEV) models that are consistent with the Random Utility Maximization (RUM) framework and that subsume standard count models including Poisson, Geometric, Negative Binomial, Binomial, and Logarithmic models as special cases were developed. The ability of the Maximum Likelihood (ML) inference approach to retrieve the parameters of the resulting GEV count models was examined using synthetic data. The simulation results indicate that the ML estimation technique performs quite well in terms of recovering the true parameters of the proposed GEV count models. Also, the models developed were used to analyze the monthly telecommuting frequency decisions of workers. Overall, the empirical results demonstrate superior data fit and better predictive performance of the GEV models compared to standard count models.

Non-stationary frequency analysis of extreme precipitation in South Korea using peaks-over-threshold and annual maxima

The conventional approach to the frequency analysis of extreme precipitation is complicated by non-stationarity resulting from climate variability and change. This study utilized a non-stationary frequency analysis to better understand the time-varying behavior of short-duration (1-, 6-, 12-, and 24-h) precipitation extremes at 65 weather stations scattered across South Korea. Trends in precipitation extremes were diagnosed with respect to both annual maximum precipitation (AMP) and peaks-over-threshold (POT) extremes. Non-stationary generalized extreme value (GEV) and generalized Pareto distribution (GPD) models with model parameters made a linear function of time were applied to AMP and POT respectively. Trends detected using the Mann–Kendall test revealed that the stations showing an increasing trend in AMP extremes were concentrated in the mountainous areas (the northeast and southwest regions) of South Korea. Trend tests on POT extremes provided fairly different results, with a significantly reduced number of stations showing an increasing trend and with some stations showing a decreasing trend. For most of stations showing a statistically significant trend, non-stationary GEV and GPD models significantly outperformed their stationary counterparts, particularly for precipitation extremes with shorter durations. Due to a significant-increasing trend in the POT frequency found at a considerable number of stations (about 10 stations for each rainfall duration), the performance of modeling POT extremes was further improved with a non-homogeneous Poisson model. The large differences in design storm estimates between stationary and non-stationary models (design storm estimates from stationary models were significantly lower than the estimates of non-stationary models) demonstrated the challenges in relying on the stationary assumption when planning the design and management of water facilities. This study also highlighted the need of caution when quantifying design storms from POT and AMP extremes by showing a large discrepancy between the estimates from those two approaches.

A comparative analysis of the UK and Italian small businesses using Generalised Extreme Value models

This paper presents a cross-country comparison of significant predictors of small business failure between Italy and the UK. Financial measures of profitability, leverage, coverage, liquidity, scale and non-financial information are explored, some commonalities and differences are highlighted. Several models are considered, starting with the logistic regression which is a standard approach in credit risk modelling. Some important improvements are investigated. Generalised Extreme Value (GEV) regression is applied in contrast to the logistic regression in order to produce more conservative estimates of default probability. The assumption of non-linearity is relaxed through application of BGEVA, non-parametric additive model based on the GEV link function. Two methods of handling missing values are compared: multiple imputation and Weights of Evidence (WoE) transformation. The results suggest that the best predictive performance is obtained by BGEVA, thus implying the necessity of taking into account the low volume of defaults and non-linear patterns when modelling SME performance. WoE for the majority of models considered show better prediction as compared to multiple imputation, suggesting that missing values could be informative.

Flood Frequency Analysis Considering Probability Distribution and Return Period under Non-stationary Condition

This study performed the non-stationary flood frequency analysis considering time-varying parameters of a probability density function. Also, return period and risk under non-stationary condition were estimated. A stationary model and three non-stationary models using Generalized Extreme Value(GEV) were developed. The only location parameter was assumed as time-varying parameter in the first model. In second model, the only scale parameter was assumed as time-varying parameter. Finally, the both parameters were assumed as time varying parameter in the last model. Relative likelihood ratio test and Akaike information criterion were used to select appropriate model. The suggested procedure in this study was applied to eight multipurpose dams in South Korea. Using relative likelihood ratio test and Akaike information criterion it is shown that the inflow into the Hapcheon dam and the Seomjingang dam were suitable for non-stationary GEV model but the other six dams were suitable for stationary GEV model. Also, it is shown that the estimated return period under non-stationary condition was shorter than those estimated under stationary condition.

Analyzing streamflow extremes in the upper Ürümqi River with the generalized Pareto distribution

The generalized Pareto distribution (GPD), as one of the most important distributions in statistical theory of extreme value, is often used to explain the probability of extreme events in nature, through setting up a model with the observation points exceeding the threshold. This paper uses GPD model to fit the distributions of the relatively large and small monthly average discharge of Urumqi River in Northwest China and gives the detailed steps of this method. Firstly, the Mean Excess, shape parameter and modified scale parameter plots are applied to determine the thresholds; then the parameters of GPD are estimated by the maximum likelihood method; next the models are diagnosed by the probability and quantile plots; finally the return levels and the corresponding 95 % confidence intervals of the discharge are calculated by the maximum likelihood and profile likelihood methods, respectively. The results show that the return levels of the maximum monthly average discharge with the return periods 10, 25, 50 and 100 years are 35.4, 39.9, 43.2 and 46.3 m3 s−1, respectively, and the return levels of the minimum monthly average discharge with the return periods 10, 25, 50 and 100 years are 0.60, 0.43, 0.30 and 0.18 m3 s−1, respectively. Some comparisons are also made between the generalized extreme value (GEV) and GPD models. The results of these two models are close to each other while the GPD model should be more reliable because it can make use of more information than the GEV model. This paper proposes a complete framework for modeling hydrological data by GPD model.

Comparing different extreme wave analysis models for wave climate assessment along the Italian coast

By utilizing various statistical models to quantify the return levels of extreme significant wave height this study seeks to achieve two objectives: the updating of the state-of-the-art concerning extreme wave analysis along the Italian coast and the creation of a long-term predictive model. To these ends, four different methods widely used in the field of metocean engineering are employed to analyze both buoy data (Rete Ondametrica Nazionale, RON) and wave data obtained by means of dynamical hindcasting techniques such as the forecast/hindcast operational model chain in use at the University of Genoa (www.dicca.unige.it/meteocean). Return levels are estimated by the Goda method, the Generalized Extreme Value (GEV) and the Generalized Pareto Distribution–Poisson point process models and the Equivalent Triangular Storm (ETS) algorithm. All models follow the Peak-Over-Threshold (POT) approach which require an optimal threshold implementation, save for the GEV analysis, which is applied to model significant wave height maxima pertaining to time-blocks. The models exhibit different performance characteristics, presented here and treated in depth. In general, noteworthy versatility characterizes the GPD–Poisson model, which often recovers Goda results, while the GEV and ETS models exhibit limitations in assessment of a variety of wave fields, greatly diversified in a semi-closed basin such as the Mediterranean Sea. Long-term estimates have been provided by means of the most appropriate model selected, thus offering a complete overview of wave climate in the Mediterranean basin based on wave hindcast data.

Spatial modeling of extreme rainfall in northeast Thailand

It is well recognized that the generalized extreme value (GEV) distribution is widely used for any extreme events. This notion is based on the study of discrete choice behavior; however, there is a limit for predicting the distribution at ungauged sites. Hence, there have been studies on spatial dependence within extreme events in continuous space using recorded observations. We model the annual maximum daily rainfall data consisting of 25 locations for the period from 1982 to 2013. The spatial GEV model that is established under observations is assumed to be mutually independent because there is no spatial dependency between the stations. Furthermore, we divide the region into two regions for a better model fit and identify the best model for each region. We show that the regional spatial GEV model reflects the spatial pattern well compared with the spatial GEV model over the entire region as the local GEV distribution. The advantage of spatial extreme modeling is that more robust return levels and some indices of extreme rainfall can be obtained for observed stations as well as for locations without observed data. Thus, the model helps to determine the effects and assessment of vulnerability due to heavy rainfall in northeast Thailand.