Pedosphere 33(3): 479--495, 2023
ISSN 1002-0160/CN 32-1315/P
©2023 Soil Science Society of China
Published by Elsevier B.V. and Science Press
Multi-step ahead soil temperature forecasting at different depths based on meteorological data: Integrating resampling algorithms and machine learning models
Khabat KHOSRAVI1, Ali GOLKARIAN2, Rahim BARZEGAR3,4, Mohammad T. AALAMI5, Salim HEDDAM6, Ebrahim OMIDVAR7, Saskia D. KEESSTRA8,9, Manuel LÓPEZ-VICENTE10
1 Department of Earth and Environment, Florida International University, Miami 33199(USA);
2 Department of Watershed Management Engineering, Ferdowsi University of Mashhad, Mashhad 91779-48974(Iran);
3 Department of Bioresource Engineering, McGill University, 21111 Lakeshore, Ste Anne de Bellevue, Montreal H9X 3V9(Canada);
4 Department of Geography & Environmental Studies, Wilfrid Laurier University, Waterloo N2L 3G1(Canada);
5 Faculty of Civil Engineering, University of Tabriz, 29 Bahman Blv, Tabriz 51666-16471(Iran);
6 Laboratory of Research in Biodiversity Interaction Ecosystem and Biotechnology, University 20 Août 1955, Route El Hadaik, BP 26, Skikda 21000(Algeria);
7 Department of Watershed Management Engineering, University of Kashan, Kashan 87317-53153(Iran);
8 Team Soil, Water and Land Use, Wageningen Environmental Research, Droevendaalsesteeg 3, Wageningen 6708 RC(Netherlands);
9 Departamento de Análisis Geográfico Regional y Geografía Física, Facultad de Filosofía y Letras, Campus Universitario de Cartuja, Universidad de Granada, Granada 18071(Spain);
10 Group Aquaterra, CICA, Universidade da Coruña, As Carballeiras s/n, Campus de Elviña, A Coruña 15071(Spain)
Corresponding Author:Khabat KHOSRAVI
ABSTRACT
      Direct soil temperature (ST) measurement is time-consuming and costly; thus, the use of simple and cost-effective machine learning (ML) tools is helpful. In this study, ML approaches, including KStar, instance-based K-nearest learning (IBK), and locally weighted learning (LWL), coupled with resampling algorithms of bagging (BA) and dagging (DA) (BA-IBK, BA-KStar, BA-LWL, DA-IBK, DA-KStar, and DA-LWL) were developed and tested for multi-step ahead (3, 6, and 9 d ahead) ST forecasting. In addition, a linear regression (LR) model was used as a benchmark to evaluate the results. A dataset was established, with daily ST time-series at 5 and 50 cm soil depths in a farmland as models’ output and meteorological data as models’ input, including mean (Tmean), minimum (Tmin), and maximum (Tmax) air temperatures, evaporation (Eva), sunshine hours (SSH), and solar radiation (SR), which were collected at Isfahan Synoptic Station (Iran) for 13 years (1992-2005). Six different input combination scenarios were selected based on Pearson’s correlation coefficients between inputs and outputs and fed into the models. We used 70% of the data to train the models, with the remaining 30% used for model evaluation via multiple visual and quantitative metrics. Our findings showed that Tmean was the most effective input variable for ST forecasting in most of the developed models, while in some cases the combinations of variables, including Tmean and Tmax and Tmean, Tmax, Tmin, Eva, and SSH proved to be the best input combinations. Among the evaluated models, BA-KStar showed greater compatibility, while in most cases, BA-IBK and -LWL provided more accurate results, depending on soil depth. For the 5 cm soil depth, BA-KStar had superior performance (i.e., Nash-Sutcliffe efficiency (NSE) = 0.90, 0.87, and 0.85 for 3, 6, and 9 d ahead forecasting, respectively); for the 50 cm soil depth, DA-KStar outperformed the other models (i.e., NSE = 0.88, 0.89, and 0.89 for 3, 6, and 9 d ahead forecasting, respectively). The results confirmed that all hybrid models had higher prediction capabilities than the LR model.
Key Words:  bootstrap aggregating algorithm,data mining,disjoint aggregating algorithm,ensemble modeling,hybrid model
Citation: Khosravi K, Golkarian A, Barzegar R, Aalami M T, Heddam S, Omidvar E, Keesstra S D, López-Vicente M. 2023. Multi-step ahead soil temperature forecasting at different depths based on meteorological data: Integrating resampling algorithms and machine learning models. Pedosphere. 33(3): 479-495.
View Full Text



版权所有 © 2024 《PEDOSPHERE》(土壤圈)编委会
地址:南京市北京东路71号 中科院南京土壤研究所 邮编:210008    E-mail:pedosphere@issas.ac.cn
技术支持:北京勤云科技发展有限公司  京ICP备09084417号