Estimation of Meteorological Drought Based on Machine Learning Models in Zhejiang Province, China
DOI:
https://doi.org/10.13052/spee1048-5236.4517Keywords:
Meteorological drought, BP model, SPEI, remote sensing, Zhejiang provinceAbstract
Drought represents a critically hazardous natural disaster, and precise drought forecasting is important for agriculture, environment, and human activities. Machine learning models are effective tools for drought prediction because they can analyze complex hierarchical and nonlinear relationships. The potential of a backpropagation (BP) neural network to evaluate the standardized precipitation evapotranspiration index (SPEI) by utilizing remote sensing datasets from Zhejiang Province, China is explored in this research. Three-variable input were selected: precipitation, soil moisture, and the difference between precipitation and potential evapotranspiration. Three input variable combinations were evaluated: single-variable input (Scheme I), two-variable input (Scheme II), and three-variable input (Scheme III). Results show that, in Zhejiang Province, the BP model exhibits good performance, with Nash–Sutcliffe efficiency values ranging from 0.84 to 0.99, correlation coefficient values from 0.92 to 0.99, and root mean square error values from 0.12 to 0.42. Notably, model performance improves significantly from Scheme I to Scheme II. However, the transition from Scheme II to Scheme III yields only slight improvements at six stations, and the performance of the BP model under Scheme II remains superior to that under Scheme III. Furthermore, the findings suggest that adding more input variables is unnecessary to enhance the prediction accuracy of SPEI3 (SPEI at three months) using the BP model.
Downloads
References
S. Lee, D. N. Moriasi, A. Danandeh Mehr, and A. Mirchi, “Sensitivity of standardized precipitation and evapotranspiration index (SPEI) to the choice of SPEI probability distribution and evapotranspiration method,” J. Hydrol. Reg. Stud., vol. 53, p. 101761, June 2024.
A. Abu Arra and E. Şişman, “A comprehensive analysis and comparison of SPI and SPEI for spatiotemporal drought evaluation,” Environ. Monit. Assess., vol. 196, no. 10, p. 980, Oct. 2024.
V. Gumus, “Evaluating the effect of the SPI and SPEI methods on drought monitoring over turkey,” J. Hydrol., vol. 626, p. 130386, Nov. 2023.
J. Uwimbabazi, Y. Jing, V. Iyakaremye, I. Ullah, and B. Ayugi, “Observed changes in meteorological drought events during 1981–2020 over rwanda, east africa,” Sustainability, vol. 14, no. 3, p. 1519, Jan. 2022.
W. Zhang et al., “Temporal and spatial evolution of meteorological drought in inner mongolia inland river basin and its driving factors,” Sustainability, vol. 16, no. 5, p. 2212, Mar. 2024.
B. Ayugi et al., “Review of meteorological drought in africa: Historical trends, impacts, mitigation measures, and prospects,” Pure Appl. Geophys., vol. 179, no. 4, pp. 1365–1386, Apr. 2022.
A. A. Pathak and B. M. Dodamani, “Connection between meteorological and groundwater drought with copula-based bivariate frequency analysis,” J. Hydrol. Eng., vol. 26, no. 7, p. 05021015, July 2021.
R. Xue et al., “Future projections of meteorological, agricultural and hydrological droughts in China using the emergent constraint,” J. Hydrol. Reg. Stud., vol. 53, p. 101767, June 2024.
T. Lan and X. Yan, “Analysis of drought characteristics and causes in yunnan province in the last 60 years (1961–2020),” J. Hydrometeorol., vol. 25, no. 1, pp. 177–190, Jan. 2024.
L. Zhang et al., “Dynamic multi-dimensional identification of yunnan droughts and its seasonal scale linkages to the el niño-southern oscillation,” J. Hydrol. Reg. Stud., vol. 42, p. 101128, Aug. 2022.
Z. Hao, D. Xiong, and J. Zheng, “How ancient China dealt with summer droughts – a case study of the whole process of the 1751 drought in the qing dynasty,” Clim. Change, vol. 165, no. 1–2, p. 16, Mar. 2021.
I. J. Nwayor and S. M. Robeson, “Exploring the relationship between SPI and SPEI in a warming world,” Theor. Appl. Climatol., vol. 155, no. 4, pp. 2559–2569, Apr. 2024.
A. Merabti, H. Darouich, P. Paredes, M. Meddi, and L. S. Pereira, “Assessing spatial variability and trends of droughts in eastern algeria using SPI, RDI, PDSI, and MedPDSI – a novel drought index using the FAO56 evapotranspiration method,” Water, vol. 15, no. 4, p. 626, Feb. 2023.
M. Kamruzzaman et al., “Spatiotemporal drought analysis in bangladesh using the standardized precipitation index (SPI) and standardized precipitation evapotranspiration index (SPEI),” Sci. Rep., vol. 12, no. 1, p. 20694, Nov. 2022.
D. C. D. Melo et al., “Are remote sensing evapotranspiration models reliable across south American ecoregions?,” Water Resour. Res., vol. 57, no. 11, p. e2020WR028752, Nov. 2021.
S. Wanniarachchi and R. Sarukkalige, “A review on evapotranspiration estimation in agricultural water management: Past, present, and future,” Hydrology, vol. 9, no. 7, p. 123, July 2022.
Y. Zhao et al., “Drought monitoring and performance evaluation based on machine learning fusion of multi-source remote sensing drought factors,” Remote Sens., vol. 14, no. 24, p. 6398, Dec. 2022.
H. Zhao, Y. Huang, X. Wang, X. Li, and T. Lei, “The performance of SPEI integrated remote sensing data for monitoring agricultural drought in the north China plain,” Field Crops Res., vol. 302, p. 109041, Oct. 2023.
P. S. Wable, M. K. Jha, S. Adamala, M. K. Tiwari, and S. Biswal, “Application of hybrid ANN techniques for drought forecasting in the semi-arid region of India,” Environ. Monit. Assess., vol. 195, no. 9, p. 1090, Sept. 2023.
M. S. Oyounalsoud, A. G. Yilmaz, M. Abdallah, and A. Abdeljaber, “Drought prediction using artificial intelligence models based on climate data and soil moisture,” Sci. Rep., vol. 14, no. 1, p. 19700, Aug. 2024.
T. Yildirim, D. N. Moriasi, P. J. Starks, and D. Chakraborty, “Using artificial neural network (ANN) for short-range prediction of cotton yield in data-scarce regions,” Agronomy, vol. 12, no. 4, p. 828, Mar. 2022.
A. Mokhtar et al., “Estimation of SPEI meteorological drought using machine learning algorithms,” IEEE Access, vol. 9, pp. 65503–65523, 2021.
B. Poudel, D. Dahal, M. Banjara, and A. Kalra, “Assessing meteorological drought patterns and forecasting accuracy with SPI and SPEI using machine learning models,” Forecasting, vol. 6, no. 4, pp. 1026–1044, Nov. 2024.
Y. Xu, S. Pan, G. Fu, Y. Tian, and X. Zhang, “Future potential evapotranspiration changes and contribution analysis in zhejiang province, east China,” J. Geophys. Res. Atmospheres, vol. 119, no. 5, pp. 2174–2192, Mar. 2014.
M. Ghaderi and M. Rahimzadegan, “Improving precipitable water vapor estimations of the moderate resolution imaging spectroradiometer (MODIS) using metaheuristic algorithms,” Adv. Space Res., vol. 69, no. 12, pp. 4274–4287, June 2022.
S. Kloos, Y. Yuan, M. Castelli, and A. Menzel, “Agricultural drought detection with MODIS based vegetation health indices in southeast germany,” Remote Sens., vol. 13, no. 19, p. 3907, Jan. 2021.
K. Yang et al., “China meteorological forcing dataset v1.6 (1979–2018).” National Tibetan Plateau Data Center, 2019.
K. Yang, J. He, W. Tang, J. Qin, and C. C. K. Cheng, “On downward shortwave and longwave radiations over high altitude regions: Observation and modeling in the tibetan plateau,” Agric. For. Meteorol., vol. 150, no. 1, pp. 38–46, Jan. 2010.
D. G. Miralles, R. A. M. De Jeu, J. H. Gash, T. R. H. Holmes, and A. J. Dolman, “Magnitude and variability of land evaporation and its components at the global scale,” Hydrol. Earth Syst. Sci., vol. 15, no. 3, pp. 967–981, Mar. 2011.
D. G. Miralles et al., “GLEAM4: Global land evaporation and soil moisture dataset at 0.1∘ resolution from 1980 to near present,” Sci. Data, vol. 12, no. 1, p. 416, Mar. 2025.
J. Liu, J. Zhang, D. Kong, X. Feng, S. Feng, and M. Xiao, “Contributions of anthropogenic forcings to evapotranspiration changes over 1980–2020 using GLEAM and CMIP6 simulations,” J. Geophys. Res. Atmospheres, vol. 126, no. 22, p. e2021JD035367, Nov. 2021.
X. Yang, B. Yong, L. Ren, Y. Zhang, and D. Long, “Multi-scale validation of GLEAM evapotranspiration products over China via ChinaFLUX ET measurements,” Int. J. Remote Sens., vol. 38, no. 20, pp. 5688–5709, Oct. 2017.
S. M. Vicente-Serrano, S. Beguería, and J. I. López-Moreno, “A multiscalar drought index sensitive to global warming: The standardized precipitation evapotranspiration index,” J. Clim., vol. 23, no. 7, pp. 1696–1718, Apr. 2010.
S. Pan, Y.-P. Xu, W. Xuan, H. Gu, and Z. Bai, “Appropriateness of potential evapotranspiration models for climate change impact analysis in yarlung zangbo river basin, China,” Atmosphere, vol. 10, no. 8, p. 453, Aug. 2019.
R. G. Allen, M. Smith, A. Perrier, and L. S. Pereira, “An update for the calculation of potential evapotranspiration,” ICID Bull, vol. 43, pp. 35–92, 1994.
L. Zhao et al., “Prediction model for reference crop evapotranspiration based on the back-propagation algorithm with limited factors,” Water Resour. Manag., vol. 37, no. 3, pp. 1207–1222, Feb. 2023.
Z. Zhang, Y. Gong, and Z. Wang, “Accessible remote sensing data based reference evapotranspiration estimation modelling,” Agric. Water Manag., vol. 210, pp. 59–69, Nov. 2018.
Y. Gong, Y. Zhang, S. Lan, and H. Wang, “A comparative study of artificial neural networks, support vector machines and adaptive neuro fuzzy inference system for forecasting groundwater levels near lake okeechobee, florida,” Water Resour. Manag., vol. 30, no. 1, pp. 375–391, Jan. 2016.
H. Tao, L. Diop, A. Bodian, K. Djaman, P. M. Ndiaye, and Z. M. Yaseen, “Reference evapotranspiration prediction using hybridized fuzzy model with firefly algorithm: Regional case study in burkina faso,” Agric. Water Manag., vol. 208, pp. 140–151, Sept. 2018.
https://aeeeuropeenergy.com/
