Premium
Spatial Prediction of Soil Organic Matter Content Using Cokriging with Remotely Sensed Data
Author(s) -
Wu Chunfa,
Wu Jiaping,
Luo Yongming,
Zhang Limin,
DeGloria Stephen D.
Publication year - 2009
Publication title -
soil science society of america journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.836
H-Index - 168
eISSN - 1435-0661
pISSN - 0361-5995
DOI - 10.2136/sssaj2008.0045
Subject(s) - thematic mapper , environmental science , kriging , thematic map , remote sensing , standard deviation , spatial analysis , soil science , satellite imagery , mathematics , statistics , cartography , geography
Accurately measuring soil organic matter content (SOM) in paddy fields is important because SOM is one of the key soil properties controlling nutrient budgets in agricultural production systems. Estimation of this soil property at an acceptable level of accuracy is important; especially in the case when SOM exhibits strong spatial dependence and its measurement is a time‐ and labor‐consuming procedure. This study was conducted to evaluate and compare spatial estimation by kriging and cokriging with remotely sensed data to predict SOM using limited available data for a 367‐km 2 study area in Haining City, Zhejiang Province, China. Measured SOM ranged from 5.7 to 40.4 g kg −1 , with a mean of 19.5 g kg −1 Correlation analysis between the SOM content of 131 soil samples and the corresponding digital number (DN) of six bands (Band 1–5 and Band 7) of Landsat Enhanced Thematic Mapper (ETM) imagery showed that correlation between SOM and DN of Band 1 was the highest ( r = −0.587). We used the DN of Band 1 as auxiliary data for the SOM prediction, and used descriptive statistics and the kriging standard deviation (STD) to compare the reliabilities of the predictions. We also used cross‐validation to validate the SOM prediction. Results indicate that cokriging with remotely sensed data was superior to kriging in the case of limited available data and the moderately strong linear relationship between remotely sensed data and SOM content. Remotely sensed data such as Landsat ETM imagery have the potential as useful auxiliary variables for improving the precision and reliability of SOM prediction.