Spatial modelling of soil salinity: deep or shallow learning models?

A - Papers appearing in refereed journals

Mohammadifar, A., Gholami, H., Golzari, S. and Collins, A. L. 2021. Spatial modelling of soil salinity: deep or shallow learning models? Environmental Science and Pollution Research. 28, pp. 39432-39450. https://doi.org/10.1007/s11356-021-13503-7

AuthorsMohammadifar, A., Gholami, H., Golzari, S. and Collins, A. L.
Abstract

Understanding the spatial distribution of soil salinity is required to conserve land against degradation and desertification. Against this background, this study is the first attempt to predict soil salinity in the Jaghin basin, in southern Iran, by applying and comparing the performance of four deep learning (DL) models (deep convolutional neural networks—DCNNs, dense connected deep neural networks—DenseDNNs, recurrent neural networks-long short-term memory—RNN-LSTM and recurrent neural networks-gated recurrent unit—RNN-GRU) and six shallow machine learning (ML) models (bagged classification and regression tree—BCART, cforest, cubist, quantile regression with LASSO penalty—QR-LASSO, ridge regression—RR and support vectore machine—SVM). To do this, 49 environmental landsat8-derived variables including digital elevation model (DEM)-extracted covariates, soil-salinity indices, and other variables (e.g., soil order, lithology, land use) were mapped spatially. For assessing the relationships between soil salinity (EC) and factors controlling EC, we collected 319 surficial (0–5 cm depth) soil samples for measuring soil salinity on the basis of electrical conductivity (EC). We then selected the most important features (covariates) controlling soil salinity by applying a MARS model. The performance of the DL and shallow ML models for generating soil salinity spatial maps (SSSMs) was assessed using a Taylor diagram and the Nash Sutcliff coefficient (NSE). Among all 10 predictive models, DL models with NSE ≥ 0.9 (DCNNs was the most accurate model with NSE = 0.96) were selected as the four best models, and performed better than the six shallow ML models with NSE ≤ 0.83 (QR-LASSO was the weakest predictive model with NSE = 0.50). Based on DCNNs-, the values of the EC ranged between 0.67 and 14.73 dS/m, whereas for QR-LASSO the corresponding EC values were 0.37 to 19.6 dS/m. Overall, DL models performed better than shallow ML models for production of the SSSMs and therefore we recommend applying DL models for prediction purposes in environmental sciences.

KeywordsDeep learning models; Shallow machine learning models; Soil salinity spatial maps; Deep convolutional neural networks
Year of Publication2021
JournalEnvironmental Science and Pollution Research
Journal citation28, pp. 39432-39450
Digital Object Identifier (DOI)https://doi.org/10.1007/s11356-021-13503-7
Web address (URL)https://doi.org/10.1007/s11356-021-13503-7
Open accessPublished as non-open access
FunderBiotechnology and Biological Sciences Research Council
Funder project or codeS2N - Soil to Nutrition - Work package 3 (WP3) - Sustainable intensification - optimisation at multiple scales
Output statusPublished
Publication dates
Online23 Mar 2021
Publication process dates
Accepted15 Mar 2021
PublisherSpringer Heidelberg
ISSN0944-1344

Permalink - https://repository.rothamsted.ac.uk/item/98464/spatial-modelling-of-soil-salinity-deep-or-shallow-learning-models

Restricted files

Publisher's version

Under embargo indefinitely

151 total views
0 total downloads
0 views this month
0 downloads this month