Forecasting SPEI and SPI Drought Indices Using the Integrated Artificial Neural Networks

Máca, Petr; Pech, Pavel

doi:10.1155/2016/3868519

Cited by 45 publications

(20 citation statements)

References 48 publications

(63 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Unfortunately, these authors do not refer the errors to any classification. Maca and Pech [29], analyzing the forecast of SPI using two types of neural network models, found similar MAE and RMSE values. The performances of the different wavelet models for forecasting meteorological drought-identified by SPI in southeastern part of East Azerbaijan province, Iran-were evaluated by comparing RMSE and R 2 [27].…”

Section: Spi Value Forecastmentioning

confidence: 80%

“…The performances of all models were compared using the root mean squared error (RMSE), the mean absolute error (MAE), the coefficient of determination (R 2 ), and a measure of persistence. Maca and Pech [29] compared forecast of drought indices based on two different models of artificial neural networks. The analyzed drought indices were the SPI and the standardized precipitation evaporation index (SPEI), which were derived for the period of 1948-2002 on two U.S. catchments.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Categorical Forecast of Precipitation Anomaly Using the Standardized Precipitation Index SPI

Labedzki

2017

Water

View full text Add to dashboard Cite

Abstract:In the paper, the verification of forecasts of precipitation conditions measured by the standardized precipitation index (SPI) is presented. For the verification of categorical forecasts, a contingency table was used. Standard verification measures were used for the SPI value forecast. The 30-day SPI, moved every 10 days by 10 days, was calculated in 2013-2015 from April to September on the basis of precipitation data from 35 meteorological stations in Poland. Predictions of the 30-day SPI were created in which precipitation was forecasted for the next 10 days (the SPI 10-day forecast) and 20 days (the SPI 20-day forecast). For both the 10 and 20 days, the forecasts were skewed towards drier categories at the expense of wet categories. There was a good agreement between observed and 10-day forecast categories of precipitation. Less agreement is obtained for 20-day forecasts-these forecasts evidently "over-dry" the assessment of precipitation anomalies. The 10-day SPI value forecast accuracy is very good or good depending on the performance measure, whereas accuracy of the 20-day forecast is unsatisfactory. Both for the SPI categorical and the SPI value forecast, the 10-day SPI forecast is trustworthy and the 20-day forecast should be accepted with reservation and used with caution.

show abstract

Section: Spi Value Forecastmentioning

confidence: 80%

Section: Introductionmentioning

confidence: 99%

Categorical Forecast of Precipitation Anomaly Using the Standardized Precipitation Index SPI

Labedzki

2017

Water

View full text Add to dashboard Cite

show abstract

“…The fixed effects in this model are the consecutive day of the year (DOY) ranging between 1 and 365, and expressing the cyclic and periodic change of the seasons, the monitoring day of the analyzed time-span (DAY), expressing the general growth trend over the five years and ranging between 1 (01/01/2009) and 1825 (31/12/2013), the Standardized Precipitation-Evapotranspiration Index (SPEI), the maximum temperature (T mx ), the minimum temperature (T mn ) and the precipitation (P) of the reference period of sampling. The random effect denoted by r. SPEI variable was obtained from raw temperature and precipitation data using the SPEI package [51] available for R statistical language [50]. The SPEI was calculated with monthly resolution using the Hargreaves equation for reference evapotranspiration (ET0):…”

Section: Statistical Analysis and Modelling Methodsmentioning

confidence: 99%

Exploring Nonlinear Intra-Annual Growth Dynamics in Fagus sylvatica L. Trees at the Italian ICP-Forests Level II Network

Ferrara

Marchi

Fabbio

et al. 2019

Forests

View full text Add to dashboard Cite

The European beech (Fagus sylvatica L.) is a widely distributed tree species across Europe, highly sensitive to climate change and global warming. This study illustrates results of a 5-year monitoring time period from eight sites of the ICP-Forests Level II (intensive monitoring network) along the Italian latitudinal gradient. The tree-level relationship between tree growth dynamics and environmental factors, including seasonal climate fluctuations were investigated by means of tree-level Generalized Additive Mixed Models (GAMMs). Model results revealed that climate was responsible for just a portion of the variability in beech growth dynamics. Even if climatic predictors were highly significant in almost all sites, the model explained nearly 30% of the total variance (with just a maximum value of 71.6%), leaving the remaining variance unexplained and likely connected with forest management trajectories applied to each site (e.g., aged coppice and fully grown high forest). Climate change scenarios were then applied to predict site-specific future responses. By applying climate change scenarios, it was predicted that central and northern Italy would face similar climatic conditions to those currently detected at southern latitudes. A special case study was represented by VEN1 plot (Veneto, Northern Italy) whose current and future climate regimes were grouped in a unique and separated cluster.

show abstract

“…Most of the approaches are based on the use of single variables of either precipitation or vegetation conditions. Examples of predictions based on precipitation data are either based on SPI as is the case in Ali [2][3][4][5][6][7][8][9][10][11][12][13][14][15][16], [1,2,20,24] others define a super index of drought indices in the approach of [23] and [25] that define Multi-variate standardised dry index (MSDI) and Drought defining Index (DDI) respectively. The use of vegetation conditions in [21] in a forecast study stands-out in its use of 11 attributes to predict vegetation conditions.…”

Section: Meterological Droughtmentioning

confidence: 99%

A Mixed Model Approach to Vegetation Condition Prediction Using Artificial Neural Networks (ANN): Case of Kenya’s Operational Drought Monitoring

et al. 2019

View full text Add to dashboard Cite

Droughts, with their increasing frequency of occurrence, continue to negatively affect livelihoods and elements at risk. For example, the 2011 in drought in east Africa has caused massive losses document to have cost the Kenyan economy over $12bn. With the foregoing, the demand for ex-ante drought monitoring systems is ever-increasing. The study uses 10 precipitation and vegetation variables that are lagged over 1, 2 and 3-month time-steps to predict drought situations. In the model space search for the most predictive artificial neural network (ANN) model, as opposed to the traditional greedy search for the most predictive variables, we use the General Additive Model (GAM) approach. Together with a set of assumptions, we thereby reduce the cardinality of the space of models. Even though we build a total of 102 GAM models, only 21 have R 2 greater than 0.7 and are thus subjected to the ANN process. The ANN process itself uses the brute-force approach that automatically partitions the training data into 10 sub-samples, builds the ANN models in these samples and evaluates their performance using multiple metrics. The results show the superiority of 1-month lag of the variables as compared to longer time lags of 2 and 3 months. The champion ANN model recorded an R 2 of 0.78 in model testing using the out-of-sample data. This illustrates its ability to be a good predictor of drought situations 1-month ahead. Investigated as a classifier, the champion has a modest accuracy of 66% and a multi-class area under the ROC curve (AUROC) of 89.99%

show abstract

Forecasting SPEI and SPI Drought Indices Using the Integrated Artificial Neural Networks

Cited by 45 publications

References 48 publications

Categorical Forecast of Precipitation Anomaly Using the Standardized Precipitation Index SPI

Categorical Forecast of Precipitation Anomaly Using the Standardized Precipitation Index SPI

Exploring Nonlinear Intra-Annual Growth Dynamics in Fagus sylvatica L. Trees at the Italian ICP-Forests Level II Network

A Mixed Model Approach to Vegetation Condition Prediction Using Artificial Neural Networks (ANN): Case of Kenya’s Operational Drought Monitoring

Contact Info

Product

Resources

About