Estimation of SPEI Meteorological Drought Using Machine Learning Algorithms

Accurate estimation of drought events is vital for the mitigation of their adverse consequences on water resources, agriculture and ecosystems. Machine learning algorithms are promising methods for drought prediction as they require less time, minimal inputs, and are relatively less complex than dynamic or physical models. In this study, a combination of machine learning with the Standardized Precipitation Evapotranspiration Index (SPEI) is proposed for analysis of drought within a representative case study in the Tibetan Plateau, China, for the period of 1980–2019. Two timescales of 3 months (SPEI-3) and 6 months (SPEI-6) aggregation were considered. Four machine learning models of Random Forest (RF), the Extreme Gradient Boost (XGB), the Convolutional neural network (CNN) and the Long-term short memory (LSTM) were developed for the estimation of the SPEIs. Seven scenarios of various combinations of climate variables as input were adopted to build the models. The best models were XGB with scenario 5 (precipitation, average temperature, minimum temperature, maximum temperature, wind speed and relative humidity) and RF with scenario 6 (precipitation, average temperature, minimum temperature, maximum temperature, wind speed, relative humidity and sunshine) for estimating SPEI-3. LSTM with scenario 4 (precipitation, average temperature, minimum temperature, maximum temperature, wind speed) was relatively better for SPEI-6 estimation. The best model for SPEI-6 was XGB with scenario 5 and RF with scenario 7 (all input climate variables, i.e., scenario 6 + solar radiation). Based on the NSE index, the performances of XGB and RF models are classified as good fits for scenarios 4 to 7 for both timescales. The developed models produced satisfactory results and they could be used as a rapid tool for decision making by water-managers.

[1]  Tianqi Chen,et al.  XGBoost: A Scalable Tree Boosting System , 2016, KDD.

[2]  Rajesh Kumar,et al.  Comparison of regression and artificial neural network models for estimation of global solar radiations , 2015 .

[3]  Lucas Borges Ferreira,et al.  New approach to estimate daily reference evapotranspiration based on hourly temperature and relative humidity using machine learning and deep learning , 2020 .

[4]  Dawen Yang,et al.  Spatio-temporal variation of drought in China during 1961–2012: A climatic perspective , 2015 .

[5]  Fernando França da Cunha,et al.  Multi-step ahead forecasting of daily reference evapotranspiration using deep learning , 2020, Comput. Electron. Agric..

[6]  Hongfei Zhao,et al.  Evapotranspiration as a response to climate variability and ecosystem changes in southwest, China , 2020, Environmental Earth Sciences.

[7]  M. Janga Reddy,et al.  Ensemble prediction of regional droughts using climate inputs and the SVM–copula approach , 2014 .

[8]  M. Masinde.  Artificial neural networks models for predicting effective drought index: Factoring effects of rainfall variability , 2014, Mitigation and Adaptation Strategies for Global Change.

[9]  P. Feng,et al.  Drought prediction models driven by meteorological and remote sensing data in Guanzhong Area, China , 2020, Hydrology Research.

[10]  Ningbo Cui,et al.  Development of data-driven models for prediction of daily global horizontal irradiance in Northwest China , 2019, Journal of Cleaner Production.

[11]  R. Al-Weshah,et al.  Drought indices and indicators revisited , 2019, Arabian Journal of Geosciences.

[12]  Feng Gao,et al.  The Evaporative Stress Index as an indicator of agricultural drought in Brazil: An assessment based on crop yield impacts , 2016 .

[13]  Rui Mao,et al.  Extreme drought event of 2009/2010 over southwestern China , 2012, Meteorology and Atmospheric Physics.

[14]  Zahraie Banafsheh,et al.  SEASONAL METEOROLOGICAL DROUGHT PREDICTION USING SUPPORT VECTOR MACHINE , 2012 .

[15]  Pilar Paneque.  Drought Management Strategies in Spain , 2015 .

[16]  Najeebullah Khan,et al.  Influence of Surface Water Bodies on the Land Surface Temperature of Bangladesh , 2019 .

[17]  Zhou Yuke.  Characterizing the Spatio-Temporal Dynamics and Variability in Climate Extremes Over the Tibetan Plateau during 1960–2012 , 2019, Journal of Resources and Ecology.

[18]  A. Kurban,et al.  Meteorological Drought Analysis in the Lower Mekong Basin Using Satellite-Based Long-Term CHIRPS Product , 2017 .

[19]  T. McKee,et al.  THE RELATIONSHIP OF DROUGHT FREQUENCY AND DURATION TO TIME SCALES , 1993 .

[20]  S. Dech,et al.  The relationship between precipitation anomalies and satellite-derived vegetation activity in Central Asia , 2013 .

[21]  Ravinesh C. Deo,et al.  Input selection and data-driven model performance optimization to predict the Standardized Precipitation and Evaporation Index in a drought-prone region , 2018, Atmospheric Research.

[22]  R. Deo,et al.  Application of the extreme learning machine algorithm for the prediction of monthly Effective Drought Index in eastern Australia , 2015 .

[23]  Qunli Wu,et al.  Daily urban air quality index forecasting based on variational mode decomposition, sample entropy and LSTM neural network , 2019, Sustainable Cities and Society.

[24]  Ravinesh C. Deo,et al.  Global solar radiation prediction by ANN integrated with European Centre for medium range weather forecast fields in solar rich cities of Queensland Australia , 2019, Journal of Cleaner Production.

[25]  Martina Zeleňáková,et al.  Comprehensive Characterization of Droughts in Slovakia , 2017 .

[26]  S. Poornima,et al.  Drought prediction based on SPI and SPEI with varying timescales using LSTM recurrent neural network , 2019, Soft Computing.

[27]  Yu Feng,et al.  Comparison of artificial intelligence and empirical models for estimation of daily diffuse solar radiation in North China Plain , 2017 .

[28]  Wen Wang,et al.  Drought severity change in China during 1961–2012 indicated by SPI and SPEI , 2015, Natural Hazards.

[29]  Y. F. Huang,et al.  Application of artificial intelligence models for the prediction of standardized precipitation evapotranspiration index (SPEI) at Langat River Basin, Malaysia , 2018, Comput. Electron. Agric..

[30]  Lucas Borges Ferreira,et al.  Estimation of reference evapotranspiration in Brazil with limited meteorological data using ANN and SVM – A new approach , 2019, Journal of Hydrology.

[31]  Sergio M. Vicente-Serrano,et al.  The influence of atmospheric circulation at different spatial scales on winter drought variability through a semi‐arid climatic gradient in Northeast Spain , 2006 .

[32]  Zhao-Yue Chen,et al.  Meteorological drought forecasting based on a statistical model with machine learning techniques in Shaanxi province, China. , 2019, The Science of the total environment.

[33]  A. Mokhtar,et al.  Application of neural network and time series modeling to study the suitability of drain water quality for irrigation: a case study from Egypt , 2020, Environmental Science and Pollution Research.

[34]  Jian Ma,et al.  A Deep Neural Network Model for Short-Term Load Forecast Based on Long Short-Term Memory Network and Convolutional Neural Network , 2018, Energies.

[35]  Jian Wang,et al.  Deep learning and its application in geochemical mapping , 2019, Earth-Science Reviews.

[36]  Qiuhong Tang,et al.  Climate change impacts on meteorological, agricultural and hydrological droughts in China , 2015 .

[37]  Shaokun Li,et al.  SPEIPM-based research on drought impact on maize yield in North China Plain , 2015 .

[38]  Hyojoo Son,et al.  A Deep Learning Approach to Forecasting Monthly Demand for Residential–Sector Electricity , 2020, Sustainability.

[39]  Y. Lian,et al.  Drought monitoring utility of satellite-based precipitation products across mainland China , 2019, Journal of Hydrology.

[40]  M. Zhang,et al.  Analysis of drought events and their impacts on vegetation productivity based on the integrated surface drought index in the Hanjiang River Basin, China , 2021 .

[41]  Benjamin Lloyd-Hughes,et al.  A spatio‐temporal structure‐based approach to drought characterisation , 2012 .

[42]  Sergio M. Vicente-Serrano,et al.  Contribution of precipitation and reference evapotranspiration to drought indices under different climates , 2015 .

[43]  Nengcheng Chen,et al.  An evaluation of statistical, NMME and hybrid models for drought prediction in China , 2018, Journal of Hydrology.

[44]  J. Im,et al.  Drought assessment and monitoring through blending of multi-sensor indices using machine learning approaches for different climate regions , 2016 .

[45]  Saman Haratizadeh,et al.  CNNpred: CNN-based stock market prediction using a diverse set of variables , 2019, Expert Syst. Appl..

[46]  Eyyup Ensar Başakın,et al.  Drought Analysis with Machine Learning Methods , 2019 .

[47]  S. Vicente‐Serrano,et al.  A Multiscalar Drought Index Sensitive to Global Warming: The Standardized Precipitation Evapotranspiration Index , 2009 .

[48]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[49]  J. Sung,et al.  Appraisal of Drought Characteristics of Representative Drought Indices using Meteorological Variables , 2018 .

[50]  Shamsuddin Shahid,et al.  Rainfall variability and the trends of wet and dry periods in Bangladesh , 2010 .

[51]  Jingyun Fang,et al.  Aboveground biomass in Tibetan grasslands , 2009 .

[52]  J. Chan,et al.  The East Asian summer monsoon: an overview , 2005 .

[53]  Elie Azar,et al.  Evaluation of tree-based ensemble learning algorithms for building energy performance estimation , 2018 .

[54]  C. Agnew.  Using the SPI to identify drought , 2000 .

[55]  María Amparo Gilabert,et al.  Mapping daily global solar irradiation over Spain: A comparative study of selected approaches , 2011 .

[56]  B. Pradhan,et al.  Long lead time drought forecasting using lagged climate variables and a stacked long short-term memory model. , 2020, The Science of the total environment.

[57]  Ravinesh C. Deo,et al.  Prediction of SPEI Using MLR and ANN: A Case Study for Wilsons Promontory Station in Victoria , 2015, 2015 IEEE 14th International Conference on Machine Learning and Applications (ICMLA).

[58]  Jianping Yang,et al.  Temporal-Spatial Variation of Drought Indicated by SPI and SPEI in Ningxia Hui Autonomous Region, China , 2015 .

[59]  M. Mccabe,et al.  Evidence for intensification of meteorological droughts in Oman over the past four decades , 2020 .

[60]  Jaydeo K. Dharpure,et al.  Prediction of Combined Terrestrial Evapotranspiration Index (CTEI) over Large River Basin Based on Machine Learning Approaches , 2021, Water.

[61]  Yali Luo,et al.  Regional atmospheric anomalies responsible for the 2009–2010 severe drought in China , 2011 .

[62]  Li-Chiu Chang,et al.  Explore a deep learning multi-output neural network for regional multi-step-ahead air quality forecasts , 2019, Journal of Cleaner Production.

[63]  M. Bayazit,et al.  Markov Chain Models for Hydrological Drought Characteristics , 2012 .

[64]  Jie Zhou,et al.  Risks to water resources and development of a management strategy in the river basins of the Hengduan Mountains, Southwest China , 2020 .

[65]  Francesco Granata,et al.  Evapotranspiration evaluation models based on machine learning algorithms—A comparative study , 2019, Agricultural Water Management.

[66]  Haiming Xu,et al.  Features of the extremely severe drought in the east of Southwest China and anomalies of atmospheric circulation in summer 2006 , 2011 .

[67]  A. Sutera,et al.  Drought Monitoring and Forecasting at Large Scale , 2007 .

[68]  P. Gowda,et al.  Assessment of the Standardized Precipitation and Evaporation Index (SPEI) as a Potential Management Tool for Grasslands , 2019, Agronomy.

[69]  Jia Guo,et al.  Construction of a drought monitoring model using deep learning based on multi-source remote sensing data , 2019, Int. J. Appl. Earth Obs. Geoinformation.

[70]  Muhammed A. Hassan,et al.  Exploring the potential of tree-based ensemble methods in solar radiation modeling , 2017 .

[71]  Akimasa Sumi,et al.  A diagnostic study of the impact of El Niño on the precipitation in China , 1999 .

[72]  Jürgen Vogt,et al.  TOWARD GLOBAL DROUGHT EARLY WARNING CAPABILITY Expanding International Cooperation for the Development of a Framework for Monitoring and Forecasting , 2013 .

[73]  J. Adamowski.  Development of a short-term river flood forecasting method for snowmelt driven floods based on wavelet and cross-wavelet analysis , 2008 .

[74]  Anming Bao,et al.  Spatial and temporal characteristics of droughts in Central Asia during 1966-2015. , 2018, The Science of the total environment.

[75]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[76]  Warren M. Washington,et al.  Observed high-altitude warming and snow cover retreat over Tibet and the Himalayas enhanced by black carbon aerosols , 2014 .

[77]  L. Yanjun,et al.  Analysis of Drought Evolvement Characteristics Based on Standardized Precipitation Index in the Huaihe River Basin , 2012 .

[78]  Dennis P. Lettenmaier,et al.  Soil Moisture Drought in China, 1950–2006 , 2011 .

[79]  S. Hülsmann,et al.  Possible climate change/variability and human impacts, vulnerability of drought-prone regions, water resources and capacity building for Africa , 2016 .

[80]  Z. Fan,et al.  Spatiotemporal variability of reference evapotranspiration and its contributing climatic factors in Yunnan Province, SW China, 1961–2004 , 2012, Climatic Change.

[81]  A. Huete,et al.  An improved SPEI drought forecasting approach using the long short-term memory neural network. , 2021, Journal of environmental management.

[82]  R. Yao,et al.  Drought characteristics and its impact on changes in surface vegetation from 1981 to 2015 in the Yangtze River Basin, China , 2019, International Journal of Climatology.

[83]  L. Thompson,et al.  Different glacier status with atmospheric circulations in Tibetan Plateau and surroundings , 2012 .

[84]  Omid Bozorg-Haddad,et al.  Soil moisture simulation using hybrid artificial intelligent model: Hybridization of adaptive neuro fuzzy inference system with grey wolf optimizer algorithm , 2019, Journal of Hydrology.

[85]  Shahaboddin Shamshirband,et al.  A comparison of the performance of some extreme learning machine empirical models for predicting daily horizontal diffuse solar radiation in a region of southern Iran , 2017 .

[86]  Jiansheng Wu,et al.  Trend analysis of vegetation dynamics in Qinghai–Tibet Plateau using Hurst Exponent , 2012 .

[87]  Richard G. Allen,et al.  Using the FAO-56 dual crop coefficient method over an irrigated region as part of an evapotranspiration intercomparison study. , 2000 .

[88]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[89]  Ju-Young Shin,et al.  Stochastic simulation on reproducing long-term memory of hydroclimatological variables using deep learning model , 2020 .

[90]  Y. Hao,et al.  Different responses of soil hydrolases and oxidases to extreme drought in an alpine peatland on the Qinghai-Tibet Plateau, China , 2020 .

[91]  Ebrahim Eslami,et al.  Using a deep convolutional neural network to predict 2017 ozone concentrations, 24 hours in advance , 2020, Neural Networks.

[92]  Yan Song,et al.  Intelligent Fault Diagnosis Method Based on Full 1-D Convolutional Generative Adversarial Network , 2020, IEEE Transactions on Industrial Informatics.

[93]  E. Harsányi,et al.  Spatial–Temporal Evolution of Drought Characteristics Over Hungary Between 1961 and 2010 , 2020, Pure and Applied Geophysics.

[94]  D. A. Sachindra,et al.  Prediction of droughts over Pakistan using machine learning algorithms , 2020 .

[95]  R. Yao,et al.  Spatiotemporal variations in monthly relative humidity in China based on observations and CMIP5 models , 2020, International Journal of Climatology.

[96]  J. Adamowski,et al.  Short-term water quality variable prediction using a hybrid CNN–LSTM deep learning model , 2020, Stochastic Environmental Research and Risk Assessment.

[97]  B. Pradhan,et al.  Short-Term Spatio-Temporal Drought Forecasting Using Random Forests Model at New South Wales, Australia , 2020 .

[98]  Q. Liu,et al.  Effects of short-term N addition on soil C fluxes in alpine Sibiraea angustata scrub on the eastern margin of the Qinghai-Tibetan Plateau , 2017 .

[99]  Mariusz Ptak,et al.  Forecasting of water level in multiple temperate lakes using machine learning models , 2020 .

[100]  Jeffrey G. Arnold,et al.  Model Evaluation Guidelines for Systematic Quantification of Accuracy in Watershed Simulations , 2007 .

[101]  P. Barbosa,et al.  World drought frequency, duration, and severity for 1951–2010 , 2014 .

[102]  Henry F. Kaiser,et al.  Coefficient Alpha for a Principal Component and the Kaiser-Guttman Rule , 1991 .

[103]  Y. Radhika,et al.  Atmospheric Temperature Prediction using Support Vector Machines , 2009 .

[104]  Jane Qiu,et al.  China drought highlights future climate threats , 2010, Nature.

[105]  Keith Beven,et al.  A manifesto for the equifinality thesis , 2006 .

[106]  T. Al-Awadhi,et al.  Space and time variability of meteorological drought in Syria , 2020, Acta Geophysica.

[107]  Sushma Jain,et al.  Reference evapotranspiration estimation and modeling of the Punjab Northern India using deep learning , 2019, Comput. Electron. Agric..

[108]  Jinsong Deng,et al.  The impact of climate changes on the water footprint of wheat and maize production in the Nile Delta, Egypt. , 2020, The Science of the total environment.

[109]  Laurel Saito,et al.  ANFIS, SVM and ANN soft-computing techniques to estimate daily global solar radiation in a warm sub-humid environment , 2017 .

[110]  Lifeng Wu,et al.  Evaluating the effect of air pollution on global and diffuse solar radiation prediction using support vector machine modeling based on sunshine duration and air temperature , 2018, Renewable and Sustainable Energy Reviews.

[111]  Jan Adamowski,et al.  Long-term SPI drought forecasting in the Awash River Basin in Ethiopia using wavelet neural network and wavelet support vector regression models , 2014 .