Author(s)

Nikou Heung

Affiliation(s)

Department of Plant, Food, and Environmental Sciences, Faculty of Agriculture, Dalhousie University, Canada.

Corresponding Author

Nikou Heung

ABSTRACT

The spatial distribution of soil properties is affected by complex geological environmental factors, and the spatial differentiation characteristics are very obvious. It is difficult to achieve high-precision simulation using a single global interpolation model to simulate soil properties. For the characteristics of spatial discontinuity, limited accuracy of global interpolation models and poor adaptability, this paper proposes an adaptive surface modeling method of soil properties (ASM-SP) supported by ensemble learning and integrating geoscientific environmental variables. Using 110 sample point data collected in 2013, regression kriging (RK), Bayesian kriging (BK), ordinary kriging interpolation (OK), inverse distance weighting (IDW), ASM- SP, the total potassium content of soil was interpolated in Qinghai Lake complex landform type area. This article uses the point-by-point cross validation (LOOCV) interpolation method to simulate accuracy. The results show that ASM-SP not only takes into account the nonlinear relationship between geological environmental variables and soil properties, but also integrates the adaptability advantages of multiple models. It is a new method to achieve high-precision simulation of total soil potassium content in complex landform areas.

KEYWORDS

Spatial interpolation; Adaptive surface modeling; Environmental variables; Linear sweep algorithm; Soil total potassium content; Point-by-point cross-validation

CITE THIS PAPER

Nikou Heung. Application of ensemble learning in adaptive surface modeling of soil total potassium content in complex landform areas. Academic Journal of Earth Sciences. 2024, 2(2): 4-9. DOI: 10.61784/ajesv2n205.

REFERENCES

[1] Chen Ge, Tang Chunchun, Li Zusheng. Effects of different fertilization measures on dryland fertility and crop yield in Dongting Lake Plain area. Chinese Journal of Ecological Agriculture, 2017, 25(5): 689-697.

[2] Dong Hongfang, Yu Junbao, Sun Zhigao. Spatial distribution characteristics of organic carbon in the plant-soil system of tidal flat wetlands along the Yellow River estuary. Environmental Science, 2010, 31(6): 1594-1599.

[3] OBALUM SE, OPPONG J, IGWE CA. Spatial variability of uncultivated soils in derived savanna. International Agrophysics, 2013, 27(1): 57-67.

[4] ROSEMARY F, VITHARANA UWA, INDRARATNE SP. Exploring the spatial variability of soil properties in an Alfisol soil catena． Catena, 2017, 150: 53-61.

[5] Zhao Mingsong, Zhang Ganlin, Wang Decai. Analysis of spatial variation characteristics and main controlling factors of soil organic matter in the Xuhuai Yellow Flood Plain. Journal of Soil Science, 2013, 50(1): 1-11.

[6] Long Jun, Zhang Liming, Shen Jinquan. Research on spatial interpolation method of soil organic matter in cultivated land in complex landform areas. Acta Soil Sinica, 2014, 51 (6): 1270-1281.

[7] Shi Wenjiao, Yue Tianxiang, Shi Xiaoli. Research progress on spatial interpolation methods and accuracy of continuous soil properties. Journal of Natural Resources, 2012, 27(1): 163-175.

[8] Ma Chengxia, Ding Jianli, Wang Lu. Research on interpolation method for spatial variation analysis of oasis soil surface salt content. Soil and Water Conservation Research, 2014, 21(4): 317-320.

[9] Zhao Qiaoli, Zheng Guoqing, Feng Xiao. Comparative analysis of three spatial interpolation methods of soil total nitrogen content in Anyang County, Henan Province. Soil Bulletin, 2012, 43(5): 1162-1166.

[10] Xie Yunfeng, Chen Tongbin, Lei Mei. The influence of spatial interpolation model on soil Cd pollution assessment results. Journal of Environmental Science, 2010, 30(4): 847-854.

[11] KURIAKOSE SL, DEVKOTA S, ROSSITER DG. Prediction of soil depth using environmental variables in an anthropogenic landscape: a case study in the Western Ghats of Kerala, India. Catena, 2009, 79(1): 27-38.

[12] Jiang Guirong. Spatial variation characteristics and uncertainty analysis of soil salinity at different scales in arid areas. Beijing: China University of Geosciences, 2012.

[13] Wu Chunsheng, Huang Chong, Liu Gaohuan. Research on spatial prediction method of soil salinity in the Yellow River Delta. Resource Science, 2016, 38 (4): 704-713.

[14] Yang Lin, Zhu Axing, Zhang Shujie. Comparative study of multi-level representative sampling and stratified random sampling in soil mapping, Acta Soil Sinica, 2015, 52 (1): 28-37.

[15] Wang Shengli, Liu Wei, Zhang Lianpeng. Adaptive surface modeling of soil total potassium content supported by geological environmental variables - taking the typical area of Qinghai Lake Basin as an example. Soil and Water Conservation Research, 2018, 25(1): 132-138.

[16] Huang Wenzhong. Study on the spatial variation characteristics and influencing factors of soil potassium in Yibin City. Ya'an: Sichuan Agricultural University, 2010.

[17] Xu Aiping, Sheng Wenshun, Shu Hong. Data interpolation and cross-validation of space-time product sum model. Journal of Wuhan University (Information Science Edition), 2012, 37 (7): 766-769.

[18] Li Jia, Duan Ping, Lu Haiyang. RBF morphological parameter optimization method based on improved point-by-point cross-validation and its spatial interpolation experiment. Geography and Geographical Information Science, 2016, 32(3): 39-42.

[19] Gu Chunlei, Yang Yang, Zhu Zhichun. Cross-validation of the accuracy of several interpolation methods for establishing DEM models. Surveying, Mapping and Spatial Geographic Information, 2011, 34 (5): 99-102.