#UNSWSoilScienceCentral2018_NanLi
Nan Li
Role: 
PhD Candidate
Field of Research: 
Digital soil mapping
Contact details:
Phone: 
+61 432 010 778
Office: 

Room 570
Biological Sciences North (D26)
UNSW, Kensington 2052

#UNSWSoilScienceCentral2019: My Top 3

1. Everyone is passionate about life, eager to learn, and confident about the future.

2. A team in which we seek progress together with love and support from each other.

3. This is a family.

Evaluating a Bayesian Modeling Approach (INLA-SPDE) for Digital Soil Mapping of CEC Using Proximally Sensed Ancillary Data at the Field Scale

The soil cation exchange capacity (CEC) is one of the most important soil properties because it has an important bearing on soil fertility, acidity and structural resilience. This is particularly the case in the sugarcane growing areas of far north Queensland, because the soil there is sandy (>60 %), strongly acidic (pH < 5.5) and strongly sodic (ESP > 15 %). Unfortunately, obtaining information on CEC at the field level is time consuming and expensive. In this research, we use a digital soil mapping approach to value add to limited topsoil (0-0.30 m) and subsoil (0.6-0.9 m) CEC information. We do this by first collecting proximally sensed ancillary data from three sources, including from a digital elevation model, gamma-ray spectrometer (RS700) and electromagnetic induction instrument (DUALEM-421). To understand the uncertainty in the DSM, we evaluate the performance of a Bayesian inference approach called Integrated Nested Laplace Approximation with Stochastic Partial Differential Equation (INLA-SPDE) for predicting skewed topsoil and subsoil CEC. We also compare the accuracy (RMSE), bias (ME) and concordance (Lin’s) of DSM that can be generated from the different sources of ancillary data, either in combination or alone. We conclude, overall, that the INLA-SPDE approach was able to provide estimations of the posterior marginal distributions of the model parameters as well as the model responses. We also conclude that using all the ancillary data in combination is most accurate, least biased and had the highest concordance in both the topsoil and subsoil. The best set of ancillary data, when used alone, and for DSM mapping of CEC in the topsoil was gamma-ray, followed by DUALEM and elevation. For subsoil CEC, it was elevation.

Supervisor: Associate Professor John Triantafilis

PUBLICATIONS

Li, N., Zhao, X., Sefton, M., and Triantafilis, J. (2018). Digital soil mapping based site-specific nutrient management in a sugarcane field in Burdekin. Geoderma. (Submitted)

Li, N., Zare, E., Huang, J., and Triantafilis, J. (2018). Mapping Soil Cation-Exchange Capacity using Bayesian Modeling and Proximal Sensors at the Field Scale. Soil Science Society of America Journal.