Academic Journal of Environment & Earth Science, 2024, 6(6); doi: 10.25236/AJEE.2024.060604.
Xintong Dong1, Wanxin Liu2, Yu Chen3
1School of International Economics and Politics, Liaoning University, Shenyang, China
2School of Mathematics and Statistics, Liaoning University, Shenyang, China
3Faculty of Information, Liaoning University, Shenyang, China
In the context of rapid urbanisation, land use change has become a key factor affecting ecological environment and socio-economic development [1]. In this paper, the spatial and temporal evolution characteristics of precipitation and land use/land cover types and their driving factors are analysed by constructing various prediction models. For the spatio-temporal changes of precipitation, this paper adopts statistical methods to calculate the mean and standard deviation, and predicts the future monthly precipitation by the SARIMA model, which is combined with the least-squares method and one-way linear regression analyses to reveal its inter-annual change trends [2-4]. Meanwhile, the transfer probability of different land types was analysed by Markov chain model, and the dynamic characteristics of land use change were demonstrated by detecting the mutation points through CPT method [5]. In exploring the influence of topographic and climatic factors on extreme weather, this paper constructed an orographic precipitation model (OPM) and a multiple regression model [6], and used the random forest algorithm to capture the complex nonlinear relationships, and finally verified the validity of the model through assessment methods such as mean square error and spatial autocorrelation [7]. The results of the study provide an important reference for understanding land use change and its driving mechanism, and provide a scientific basis for related policy formulation.
Topographic precipitation modelling (OPM), random forest regression, Markov chain model, SARIMA model
Xintong Dong, Wanxin Liu, Yu Chen. Spatial and Temporal Evolution of Land Use Change and Precipitation Based on SARIMA and Random Forest Models. Academic Journal of Environment & Earth Science (2024), Vol. 6, Issue 6: 26-32. https://doi.org/10.25236/AJEE.2024.060604.
[1] JI Xiongjuan, CHENG Wenshi. Simulation prediction of land use change based on CLUE-S and Markov model[J]. Land and Natural Resources Research, 2024, (06):17-21. DOI: 10. 16202/ j. cnki. tnrs.2024.06.004.
[2] J.T. Huang, Z.X. Wang, Y.Q. Teng, et al. Prediction of unit output volume based on DE-SARIMA method [J]. Modern Information Technology, 2024, 8(12):116-120.DOI:10.19850/j.cnki.2096-4706. 2024. 12. 025.
[3] Bras A L , Gomes D , Filipe P A ,et al.Trends, seasonality and forecasts of pulmonary tuberculosis in Portugal[J].The International Journal of Tuberculosis and Lung Disease, 2014, 18(10):1202-1210.DOI:10.5588/ijtld.14.0158.
[4] LI Zhiqi, JI Yingjun, CHEN Xiangmin, et al. Research on ultra-wideband roadheader positioning algorithm based on constrained overall least squares[J]. Electronic Design Engineering, 2024, 32(20):144-148.DOI:10.14022/j.issn1674-6236.2024.20.030.
[5] M. Zhang, H.L. Liu, N.Z. Du, et al. Forecasting merchandise sales based on GM(1, 1)-Markov model[J]. Inner Mongolia Science and Economy, 2024, (05):77-82.
[6] Zhang Shengtang, Kang Shaozhong. Study on distributed precipitation model based on Lagrangian interpolation method to correct the influence of terrain[J]. Hydrology, 2004, (06):6-9.
[7] Jia Haoxin, Zou Menghua, Tang Xin, et al. Evaluation of construction suitability of geothermal power station based on random forest method[J/OL]. China Geology, 1-28[2024-10-02].http://kns.cnki. net/ kcms/ detail/11.1167.P.20240924.1714.003.html.