Journal
CATENA
Volume 164, Issue -, Pages 88-95Publisher
ELSEVIER SCIENCE BV
DOI: 10.1016/j.catena.2018.01.011
Keywords
Bulk density; Deep profile; Spatial variation; State-space model
Funding
- National Natural Science Foundation of China [41571130081, 41371242, 41530854]
- National Key Research and Development Program of China [2016YFC0501706-03]
Ask authors/readers for more resources
The soil bulk density (BD) is an important physical parameter for estimating the carbon (C) and nutrient (N) reserves in soil, and for simulating hydraulic processes. However, few BD data are available for evaluating the soil carbon and nutrient reserves as well as for simulating the hydraulic processes in deep soil profiles (> 1 m). In the present study, BD data were obtained for a 204 m profile by soil core drilling, where the objectives were to investigate the spatial variation in BD by using classical statistics and geo-statistics, and to simulate the spatial distribution of BD based on a first order autoregressive state-space model and multiple linear regression. The results showed that BD exhibited an increasing trend along the profile with low variation (coefficient of variation = 6%). Best-fit semivariograms for the BD were obtained using a Gaussian model and the spatial dependence was strong. BD was significantly correlated with selected variables, i.e., sand, silt, clay, soil organic C and depth. State-space modeling and multiple linear regression both showed that clay and depth were important factors for the total variation in the BD. In addition, the state-space model with the best performance could explain 98% of the variation in the BD. Therefore, a first order autoregressive state-space model is suitable for simulating the distribution of the BD in a deep profile.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available