A simple approach for improving spatial interpolation of rainfall using ANN

Chandrasekaran, S.; Arun, Vijayan; Giridhar, Deeptha

doi:10.1007/s00703-010-0090-z

Cited by 28 publications

(17 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…[8] There is a very extensive literature on methods for the spatial interpolation of rainfall characteristics [Sivapragasam et al, 2010;Hancock and Hutchinson, 2006;Lin and Chen, 2004;Fotheringham et al, 2002;Lanza et al, 2001;Brunsdon et al, 2001;Goovaerts, 2000;Grimes et al, 1999;Bacchi and Kottegoda, 1995;and many others], but probably the best known procedure is classical kriging [e.g., Cressie, 1993] and its many variants. The present paper uses a parametric geostatistical model [Diggle and Ribeiro, 2007;Diggle et al, 1998Diggle et al, , 2003] for spatial prediction.…”

Section: The Parametric Geostatistical Modelmentioning

confidence: 99%

Statistically combining rainfall characteristics estimated from remote‐sensed and rain gauge data sets in the Brazilian Amazon‐Tocantins Basins

Clarke

Buarque

2013

JGR Atmospheres

View full text Add to dashboard Cite

[1] This paper explores the use of a parametric geostatistical model for combining rainfall characteristics derived from rain gauge data with the same characteristics derived from remote-sensed data sets. Hypotheses can then be tested about which predictors significantly increase precision of an estimated characteristic. Although applicable wherever ground-level data and remote-sensed data are to be combined, the statistical procedure set out in the paper is developed for two examples of rainfall characteristics: (i) G, the mean annual rainfall at an ungauged site, conditional on knowledge of two predictor variables T (the mean annual rainfall calculated from the TRMM 3B42 data set for 1998-2009), and C (mean annual rainfall derived from the CMORPH data set for 2003-2009); (ii) the mean annual maximum 1 day rainfall H, interpolated using the same modeling procedure with predictor variables T and C derived from annual maximum 1 day rainfalls in the same remote-sensed data sets. Prediction errors showed no bias, skewness of distribution, or spatial heterogeneity. The model's generality means that it could be used with any predictors other than T and C, possibly derived from other satellite data sets or radar. Provided that predictor variables are correlated with the variable to be predicted, it is not necessary for the model relating them to be fitted using data from identical periods nor for the grid spacing of T and C to be identical. Model performance was evaluated by using a "leave-one-site-out" procedure, which showed that the root mean square error (RMSE) of model predictions at omitted sites was smaller than RMSEs obtained from five other well-known spatial predictors.Citation: Clarke, R. T., and D. C. Buarque (2013), Statistically combining rainfall characteristics estimated from remotesensed and rain gauge data sets in the Brazilian Amazon-Tocantins Basins,

show abstract

Section: The Parametric Geostatistical Modelmentioning

confidence: 99%

Statistically combining rainfall characteristics estimated from remote‐sensed and rain gauge data sets in the Brazilian Amazon‐Tocantins Basins

Clarke

Buarque

2013

JGR Atmospheres

View full text Add to dashboard Cite

show abstract

“…O IPD é indicado para locais com distribuição espacial boa e com alta densidade de estações (Akkala et al, 2010), o que não é o caso do presente trabalho. Além disso, a krigagem, a partir dos dados de entrada, determina os melhores parâmetros para modelagem da função matemática de interpolação, por meio do ajuste do semivariograma, o que resulta em boas estimativas, quando os dados são bem representativos da região a ser modelada (Sivapragasam et al, 2010).…”

Section: Resultsunclassified

“…Quando houver necessidade, indica-se que seja feita a análise de consistência e o preenchimento de falhas das séries históricas de chuva. Finalmente, que se avalie criteriosamente o método utilizado para espacialização dos dados mensais de chuva, nos moldes do que foi mostrado em diversos trabalhos da literatura especializada, com destaque, algumas vezes, para o IPD (Amorim et al, 2008;Ruelland et al, 2008), e frequentemente para a krigagem (Carvalho & Assad, 2005;Siqueira et al, 2007;Ruelland et al, 2008;Silva et al, 2011) e técnicas geoestatísticas de cokrigagem com suporte da topografia (Hong et al, 2005;Viola et al, 2010) ou redes neurais artificiais (Sivapragasam et al, 2010). Outra opção seria a utilização de equações de regressão para a estimativa das precipitações em função das coordenadas locais e de altitude, conforme indicado por Naoum & Tsanis (2003) (2011), que fizeram um estudo criterioso de vários interpoladores para a área em estudo, indicados apenas para esta área.…”

Section: Resultsunclassified

Método para a espacialização dos elementos do balanço hídrico climatológico

Cecílio

Silva

Xavier

et al. 2012

Pesq. agropec. bras.

View full text Add to dashboard Cite

Resumo -O objetivo deste trabalho foi desenvolver e avaliar um método de espacialização dos elementos do balanço hídrico climatológico (BHC). O método utiliza modelos digitais do terreno que contêm a distribuição espacial dos elementos climáticos de entrada do BHC. A avaliação foi feita com dados reais de BHC referentes a 110 estações pluviométricas do Estado do Espírito Santo. Comparou-se o desempenho do método proposto com o dos seguintes interpoladores espaciais: krigagem e inverso de uma potência da distância. O método proposto apresentou melhores estimativas espaciais de todos os elementos do balanço hídrico -evapotranspiração real, excedente, deficiência e disponibilidade hídrica.Termos para indexação: deficiência hídrica, excedente hídrico, interpolação, krigagem. Method for spatialization of the climatic water balance elementsAbstract -The objective of this work was to develop and evaluate a method for the spatial distribution of the climatic water balance (CWB) elements. The method applies terrain digital models containing the spatial distribution of climate elements used as input data of CWB. Evaluation was made with existing CWB data from 110 pluviometric stations of Espírito Santo state, Brazil. The proposed method's performance was compared to the one of the following spatial interpolators: kriging and inverse distance weighting. The proposed method presented best predictions of the spatial distribution of all CWB elements -actual evapotranspiration, water deficit, water surplus and water availability.

show abstract

“…Akkala et al (2010), ANN interpolators work well with sparse data irregularly distributed, just as for the data presented (Figure 1). The ANNs, in order to have better performance, need consistent training and the data-set used must represent the nuances of the terrain to be modeled (Teegavarapu 2007, Miranda et al 2009, Sivapragasam et al 2010, as was the case in this study.…”

Section: Dif = Percentage Difference Of the Values Computed For Eachmentioning

confidence: 99%

“…The ANNs are cited as an alternative resource for estimating climatic variables that may replace the traditional interpolation methods (Białobrzewski 2008, Sivapragasam et al 2010, including rainfall erosivity indexes as assessed by Moreira et al (2006) for São Paulo State, Brazil.…”

Section: Introductionmentioning

confidence: 99%

Assessing rainfall erosivity indices through synthetic precipitation series and artificial neural networks

Cecílio

Moreira

Pezzopane

et al. 2013

An. Acad. Bras. Ciênc.

View full text Add to dashboard Cite

The rainfall parameter that expresses the capacity to promote soil erosion is called rainfall erosivity (R), and is commonly represented by the indexes EI 30 and KE>25. The calculations of these indexes requires pluviographical records, that are difficult to obtain in Brazil. This paper describes the use of synthetic rainfall series to compute EI 30 and KE>25 in Espírito Santo State (Brazil). Artificial neural networks (ANNs) were also developed to spatially interpolate R values in Espírito Santo. EI 30 and KE>25 indexes values were close to those calculated on a homogeneous area according to the similarity of rainfall distribution; indicating the applicability of the use of synthetic rainfall series to estimate the R factor. ANNs had a better performance than Inverse Distance Weighted and Kriging to spatially interpolate rainfall erosivity values in the State of Espírito Santo.

show abstract

A simple approach for improving spatial interpolation of rainfall using ANN

Cited by 28 publications

References 13 publications

Statistically combining rainfall characteristics estimated from remote‐sensed and rain gauge data sets in the Brazilian Amazon‐Tocantins Basins

Statistically combining rainfall characteristics estimated from remote‐sensed and rain gauge data sets in the Brazilian Amazon‐Tocantins Basins

Método para a espacialização dos elementos do balanço hídrico climatológico

Assessing rainfall erosivity indices through synthetic precipitation series and artificial neural networks

Contact Info

Product

Resources

About