Skip to main content

Advertisement

SpringerLink
Log in

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

  • Research Article
  • Published:
Environmental Science and Pollution Research Aims and scope Submit manuscript

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.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13
Fig. 14

Similar content being viewed by others

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author (Hamid Gholami) on reasonable request.

References

  • Abdelkader D, Walter C (2006) Detecting salinity hazards within a semiarid context by means of combining soil and remote-sensing data. Geoderma 134:217–230

    Article  Google Scholar 

  • Ali M, Deo RC, Maraseni T, Downs NJ (2019) Improving SPI-derived drought forecasts incorporating synoptic-scale climate indices in multi-phase multivariate empirical mode decomposition model hybridized with simulated annealing and kernel ridge regression algorithms. J Hydrol 576:164–184

    Article  Google Scholar 

  • Alom MZ, Taha TM, Yakopcic C, Westberg S, Sidike P, Nasrin MS, Hasan M, van Essen BC, Awwal AAS, Asari VK (2019) A state-of-the-art survey on deep learning theory and architectures. Electronics 8(3):292

    Article  Google Scholar 

  • Azareh A, Rahmati O, Rafiei-Sardooi E, Sankey JB, Lee S, Shahabi H, Ahmad BB (2019) Modelling gully-erosion susceptibility in a semi-arid region, Iran: Investigation of applicability of certainty factor and maximum entropy models. Sci Total Environ 655:684–696

    Article  CAS  Google Scholar 

  • Azizi A, Gilandeh YA, Mesri-Gundoshmian T, Saleh-Bigdeli AA, Moghaddam HA (2020) Classification of soil aggregates: A novel approach based on deep learning. Soil Tillage Res 199:104586

    Article  Google Scholar 

  • Boettinger JL, Ramsey RD, Bodily JM, Cole NJ, Kienast-Brown S, Nield SJ, ..., Stum AK (2008) Landsat spectral data for digital soil mapping. In Digital soil mapping with limited data (pp. 193-202). Springer, Dordrecht

  • Breiman L (1996) Bagging predictors. Mach Learn 24(2):123–140

    Article  Google Scholar 

  • Breiman L, Friedman JH, Olsen RA, Stone CJ (1984) Classification and Regression Trees. Belmont, Wadsworth

    Google Scholar 

  • Bui DT, Tsangaratos P, Nguyen VT, Van Liem N, Trinh PT (2020) Comparing the prediction performance of a Deep Learning Neural Network model with conventional machine learning models in landslide susceptibility assessment. CATENA 188:104426

    Article  Google Scholar 

  • Cerdà A, Rodrigo-Comino J, Giménez-Morera A, Novara A, Pulido M, Kapović-Solomun M, Keesstra SD (2018a) Policies can help to apply successful strategies to control soil and water losses. The case of chipped pruned branches (CPB) in Mediterranean citrus plantations. Land Use Policy 75:734–745

    Article  Google Scholar 

  • Cerdà A, Rodrigo-Comino J, Novara A, Brevik EC, Vaezi AR, Pulido M, Giménez-Morera A, Keesstra SD (2018b) Long-term impact of rainfed agricultural land abandonment on soil erosion in the Western Mediterranean basin. Progr Phys Geogr: Earth Environ 42(2):202–219

    Article  Google Scholar 

  • Cho K, Van Merriënboer B, Gulcehre C, Bahdanau D, Bougares F, Schwenk H, Bengio Y (2014) Learning phrase representations using RNN encoder-decoder for statistical machine translation. arXiv preprint arXiv:1406.1078

  • Connor JT, Martin RD, Atlas LE (1994) Recurrent neural networks and robust time series prediction. IEEE Trans Neural Netw 5(2):240–254

    Article  CAS  Google Scholar 

  • Eishoeei E, Nazarnejad H, Miryaghoubzadeh M (2019) Temporal soil salinity modeling using SaltMod model in the west side of Urmia hyper saline Lake, Iran. Catena 176:306–314

    Article  Google Scholar 

  • Felicísimo ÁM, Cuartero A, Remondo J, Quirós E (2013) Mapping landslide susceptibility with logistic regression, multiple adaptive regression splines, classification and regression trees, and maximum entropy methods: a comparative study. Landslides 10(2):175–189

    Article  Google Scholar 

  • Gholami H, Mohammadifar A, Bui DT, Collins AL (2020a) Mapping wind erosion hazard with regression-based machine learning algorithms. Sci Rep 10(1):1–16

    Article  Google Scholar 

  • Gholami H, Mohamadifar A, Collins AL (2020b) Spatial mapping of the provenance of storm dust: Application of data mining and ensemble modelling. Atmos Res 233:104716

    Article  Google Scholar 

  • Gholami H, Mohammadifar A, Pourghasemi HR, Collins AL (2020c) A new integrated data mining model to map spatial variation in the susceptibility of land to act as a source of aeolian dust. Environ Sci Pollut Res 27:42022–42039

  • Gholami H, Mohamadifar A, Sorooshian A, Jansen JD (2020d) Machine-learning algorithms for predicting land susceptibility to dust emissions: thecase of the Jazmurian Basin, Iran. Atmos Pollut Res 11:1303–1315

    Article  CAS  Google Scholar 

  • Gholami H, Mohammadifar A, Golzari S, Kaskaoutis DG, Collins AL (2021) Using the Boruta algorithm and deep learning models for mapping land susceptibility to atmospheric dust emissions in Iran. Aeolian Res 50:100682

    Article  Google Scholar 

  • Hagenauer J, Omrani H, Helbich M (2019) Assessing the performance of 38 machine learning models: the case of land consumption rates in Bavaria, Germany. Int J Geogr Inf Sci 33(7):1399–1419

    Article  Google Scholar 

  • Hastie, T., Tibshirani, R., & Friedman, J. (2009). The elements of statistical learnin.

    Book  Google Scholar 

  • He F, Zhou J, Feng ZK, Liu G, Yang Y (2019) A hybrid short-term load forecasting model based on variational mode decomposition and long short-term memory networks considering relevant factors with Bayesian optimization algorithm. Appl Energy 237:103–116

    Article  Google Scholar 

  • Heung B, Ho HC, Zhang J, Knudby A, Bulmer CE, Schmidt MG (2016) An overview and comparison of machine-learning techniques for classification purposes in digital soil mapping. Geoderma 265:62–77

    Article  Google Scholar 

  • Hochreiter S, Schmidhuber J (1997) Long short-term memory. Neural Comput 9(8):1735–1780

    Article  CAS  Google Scholar 

  • Hoerl AE, Kennard RW (1970) Ridge regression: Biased estimation for nonorthogonal problems. Technometrics 12(1):55–67

    Article  Google Scholar 

  • Hoffman GJ, Shannon MC (2007) 4. Salinity. In Developments in agricultural engineering (Vol. 13, pp. 131-160). Elsevier

  • Hongyan C, Gengxing Z, Jingchun C, Ruiyan W, Mingxiu G (2015) Remote sensing inversion of saline soil salinity based on modified vegetation index in estuary area of Yellow River. Transact Chin Soc Agric Eng 31(5):107–114

    Google Scholar 

  • Houborg R, McCabe MF (2018) A hybrid training approach for leaf area index estimation via Cubist and random forests machine-learning. ISPRS J Photogramm Remote Sens 135:173–188

    Article  Google Scholar 

  • Hu J, Liu B, Peng S (2019) Forecasting salinity time series using RF and ELM approaches coupled with decomposition techniques. Stoch Env Res Risk A 33(4-6):1117–1135

    Article  Google Scholar 

  • Huang G, Liu Z, Van Der Maaten L, Weinberger KQ (2017) Densely connected convolutional networks. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 4700-4708)

  • Ivushkin K, Bartholomeus H, Bregt AK, Pulatov A, Kempen B, De Sousa L (2019) Global mapping of soil salinity change. Remote Sens Environ 231:111260

    Article  Google Scholar 

  • Jordan CF (1969) Derivation of leaf-area index from quality of light on the forest floor. Ecology 50(4):663–666

    Article  Google Scholar 

  • Kakeh J, Gorji M, Mohammadi MH, Asadi H, Khormali F, Sohrabi M, Cerdà A (2020) Biological soil crusts determine soil properties and salt dynamics under arid climatic condition in Qara Qir, Iran. Sci Total Environ 732:139168

    Article  CAS  Google Scholar 

  • Khan NM, Rastoskuev VV, Sato Y, Shiozawa S (2005) Assessment of hydrosaline land degradation by using a simple approach of remote sensing indicators. Agric Water Manag 77(1-3):96–109

    Article  Google Scholar 

  • Kingma DP, Ba J (2014) Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980

  • Koenker R, Bassett G Jr (1978) Regression quantiles. Econometrica 46:33–50

    Article  Google Scholar 

  • Kubicz J, Lochyński P, Pawełczyk A, Karczewski M Effects of drought on environmental health risk posed by groundwater contamination. Chemosphere 263:128145

  • LeCun Y, Bengio Y, Hinton G (2015) Deep learning. Nature 521(7553):436–444

    Article  CAS  Google Scholar 

  • Lodhi B, Kang J (2019) Multipath-DenseNet: A Supervised ensemble architecture of densely connected convolutional networks. Inf Sci 482:63–72

    Article  Google Scholar 

  • Metternicht GI, Zinck JA (2003) Remote sensing of soil salinity: potentials and constraints. Remote Sens Environ 85(1):1–20

    Article  Google Scholar 

  • Milborrow S (2014) Notes on the earth package. Retrieved October, 31, 2017

  • Mohammadifar A, Gholami H, Comino JR, Collins AL (2021) Assessment of the interpretability of data mining for the spatial modelling of water erosion using game theory. CATENA 200:105178

    Article  Google Scholar 

  • Nhu VH, Hoang ND, Nguyen H, Ngo PTT, Bui TT, Hoa PV et al (2020) Effectiveness assessment of keras based deep learning with different robust optimization algorithms for shallow landslide susceptibility mapping at tropical area. Catena 188:104458

    Article  Google Scholar 

  • Pan E, Mei X, Wang Q, Ma Y, Ma J (2020) Spectral-spatial classification for hyperspectral image based on a single GRU. Neurocomputing. 387:150–160

    Article  Google Scholar 

  • Panahi M, Jaafari A, Shirzadi A, Shahabi H, Rahmati O, Omidvar E, Lee S, Bui DT (2021) Deep learning neural networks for spatially explicit prediction of flash flood probability. Geosci Front 12(3):101076

    Article  Google Scholar 

  • Pouladi N, Møller AB, Tabatabai S, Greve MH (2019) Mapping soil organic matter contents at field level with Cubist, Random Forest and kriging. Geoderma 342:85–92

    Article  CAS  Google Scholar 

  • Pyo J, Duan H, Baek S, Kim MS, Jeon T, Kwon YS, Lee H, Cho KH (2019) A convolutional neural network regression for quantifying cyanobacteria using hyperspectral imagery. Remote Sens Environ 233:111350

    Article  Google Scholar 

  • Quinlan JR (1992) Learning with continuous classes. In 5th Australian joint conference on artificial intelligence (Vol. 92, pp. 343-348)

  • Rahman MS (2019) Computations, optimization and tuning of deep feedforward neural networks. bioRxiv

  • Rodrigo-Comino J, Martinez-Hernandez C, Iserloh T, Cerda A (2018) Contrasted impact of land abandonment on soil erosion in Mediterranean agriculture fields. Pedosphere 28(4):617–631

    Article  CAS  Google Scholar 

  • Rodriguez-Galiano V, Sanchez-Castillo M, Chica-Olmo M, Chica-Rivas MJOGR (2015) Machine learning predictive models for mineral prospectivity: An evaluation of neural networks, random forest, regression trees and support vector machines. Ore Geol Rev 71:804–818

    Article  Google Scholar 

  • Rouse JW, Haas RH, Schell JA, Deering DW (1974) Monitoring vegetation systems in the Great Plains with ERTS. NASA Spec Publ 351:309

    Google Scholar 

  • Saggi MK, Jain S (2019) Reference evapotranspiration estimation and modeling of the Punjab northern India using deep learning. Comput Electron Agric 156:387–398

    Article  Google Scholar 

  • Saunders C, Gammerman A, Vovk V (1998) Ridge regression learning algorithm in dual variables

  • Scudiero E, Skaggs TH, Corwin DL (2015) Regional-scale soil salinity assessment using Landsat ETM+ canopy reflectance. Remote Sens Environ 169:335–343

    Article  Google Scholar 

  • Shamsolmoali P, Li X, Wang R (2019) Single image resolution enhancement by efficient dilated densely connected residual network. Signal Process Image Commun 79:13–23

    Article  Google Scholar 

  • Shao Z, Cai J, Fu P, Hu L, Liu T (2019) Deep learning-based fusion of Landsat-8 and Sentinel-2 images for a harmonized surface reflectance product. Remote Sens Environ 235:111425

    Article  Google Scholar 

  • Shen R, Huang A, Li B, Guo J (2019) Construction of a drought monitoring model using deep learning based on multi-source remote sensing data. Int J Appl Earth Obs Geoinf 79:48–57

    Google Scholar 

  • Sidike P, Sagan V, Maimaitijiang M, Maimaitiyiming M, Shakoor N, Burken J, Mockler T, Fritschi FB (2019) dPEN: deep Progressively Expanded Network for mapping heterogeneous agricultural landscape using WorldView-3 satellite imagery. Remote Sens Environ 221:756–772

    Article  Google Scholar 

  • Sze V, Chen Y, Yang T, Emer J (2017) Efficient processing of deep neual networks: a tutoria and survey, 2017. arXiv preprint arXiv:1703.09039

  • Taghizadeh-Mehrjardi R, Minasny B, Sarmadian F, Malone BP (2014) Digital mapping of soil salinity in Ardakan region, central Iran. Geoderma 213:15–28

    Article  CAS  Google Scholar 

  • Tang R, Zhou G, Wang J, Zhao G, Lai Z, Jiu F (2020) A new method for estimating salt expansion in saturated saline soils during cooling based on electrical conductivity. Cold Reg Sci Technol 170:102943

    Article  Google Scholar 

  • Taylor KE (2001) Summarizing multiple aspects of model performance in a single diagram. J Geophys Res-Atmos 106(D7):7183–7192

    Article  Google Scholar 

  • Tibshirani R (1996) Regression shrinkage and selection via the lasso. J R Stat Soc Ser B Methodol 58(1):267–288

    Google Scholar 

  • Tieleman T, Hinton G (2012) Lecture 6.5-rmsprop, coursera: Neural networks for machine learning. University of Toronto, Technical Report

  • Tipping ME (2000) The relevance vector machine. In Advances in neural information processing systems (pp. 652-658)

  • Vapnik VN (1995) The nature of statistical learning. Theory

  • Vazquez JG, Grande JA, Barragán FJ, Ocaña JA, De La Torre ML (2005) Nitrate accumulation and other components of the groundwater in relation to cropping system in an aquifer in Southwestern Spain. Water Resour Manag 19(1):1–22

    Article  Google Scholar 

  • Wang Y, Wang H, Srinivasan D, Hu Q (2019a) Robust functional regression for wind speed forecasting based on Sparse Bayesian learning. Renew Energy 132:43–60

    Article  Google Scholar 

  • Wang F, Yang S, Yang W, Yang X, Jianli D (2019b) Comparison of machine learning algorithms for soil salinity predictions in three dryland oases located in Xinjiang Uyghur Autonomous Region (XJUAR) of China. Eur J Remote Sens 52(1):256–276

    Article  Google Scholar 

  • Wang J, Ding J, Yu D, Teng D, He B, Chen X et al (2020a) Machine learning-based detection of soil salinity in an arid desert region, Northwest China: A comparison between Landsat-8 OLI and Sentinel-2 MSI. Sci Total Environ 707:136092

    Article  CAS  Google Scholar 

  • Wang Z, Hong T, Piette MA (2020b) Building thermal load prediction through shallow machine learning and deep learning. Appl Energy 263:114683

    Article  Google Scholar 

  • Xu Y, Ho HC, Wong MS, Deng C, Shi Y, Chan TC, Knudby A (2018) Evaluation of machine learning techniques with multiple remote sensing datasets in estimating monthly concentrations of ground-level PM2. 5. Environ Pollut 242:1417–1426

    Article  CAS  Google Scholar 

  • Yao Z, Li J, Guan Z, Ye Y, Chen Y (2020) Liver disease screening based on densely connected deep neural networks. Neural Netw 123:299–304

    Article  Google Scholar 

  • Yu X, Wang Y, Wu L, Chen G, Wang L, Qin H (2020) Comparison of support vector regression and extreme gradient boosting for decomposition-based data-driven 10-day streamflow forecasting. J Hydrol 582:124293

    Article  Google Scholar 

  • Yuan Q, Shen H, Li T, Li Z, Li S, Jiang Y et al (2020) Deep learning in environmental remote sensing: Achievements and challenges. Remote Sens Environ 241:111716

    Article  Google Scholar 

  • Zhang TT, Qi JG, Gao Y, Ouyang ZT, Zeng SL, Zhao B (2015) Detecting soil salinity with MODIS time series VI data. Ecol Indic 52:480–489

    Article  Google Scholar 

  • Zhang Q, Yang LT, Chen Z, Li P (2018) A survey on deep learning for big data. Inform Fusion 42:146–157

    Article  Google Scholar 

  • Zhang C, Sargent I, Pan X, Li H, Gardiner A, Hare J, Atkinson PM (2019) Joint Deep Learning for land cover and land use classification. Remote Sens Environ 221:173–187

    Article  Google Scholar 

  • Zheng Q, Gallagher C, Kulasekera KB (2013) Adaptive penalized quantile regression for high dimensional data. J Stat Plann Infer 143(6):1029–1038

    Article  Google Scholar 

  • Zhong L, Hu L, Zhou H (2019) Deep learning based multi-temporal crop classification. Remote Sens Environ 221:430–443

    Article  Google Scholar 

  • Zhou J, Shi XZ, Huang RD, Qiu XY, Chen C (2016) Feasibility of stochastic gradient boosting approach for predicting rockburst damage in burst-prone mines. Trans Nonferrous Metals Soc China 26(7):1938–1945

    Article  Google Scholar 

  • Zhou J, Li E, Wei H, Li C, Qiao Q, Armaghani DJ (2019) Random forests and cubist algorithms for predicting shear strengths of rockfill materials. Appl Sci 9(8):1621

    Article  Google Scholar 

  • Zuo R, Xiong Y, Wang J, Carranza EJM (2019) Deep learning and its application in geochemical mapping. Earth-science reviews

Download references

Acknowledgements

The authors would like to thank the Faculty of Agriculture and Natural Resources, University of Hormozgan, Iran, for supporting this joint research project. Rothamsted Research receives strategic funding from the UK-BBSRC (UK Research and Innovation—Biotechnology and Biological Sciences Research Council). The contribution to this paper by ALC was funded by research grant BBS/E/C/000I0330 —Soil to Nutrition work package 3—Sustainable intensification—optimization at multiple scales.

Author information

Authors and Affiliations

  1. Department of Natural Resources Engineering, University of Hormozgan, Bandar-Abbas, Hormozgan, Iran

    Aliakbar Mohammadifar & Hamid Gholami

  2. Department of Electrical and Computer Engineering, University of Hormozgan, Bandar-Abbas, Hormozgan, Iran

    Shahram Golzari

  3. Deep Learning Research Group, University of Hormozgan, Bandar Abbas, Hormozgan, Iran

    Shahram Golzari

  4. Sustainable Agriculture Sciences Department, Rothamsted Research, North Wyke, Okehampton, Devon, EX20, 2SB, UK

    Adrian L. Collins

Authors
  1. Aliakbar Mohammadifar
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Hamid Gholami
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Shahram Golzari
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Adrian L. Collins
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

Aliakbar Mohammadifar: software, formal analysis, investigation. Hamid Gholami: software, formal analysis, investigation, visualization, writing original draft, supervision, project administration, review & editing. Shahram Golzari: formal analysis, investigation, visualization, review & editing. Adrian Collins: formal analysis, investigation, visualization, writing original draft, review & editing.

Corresponding author

Correspondence to Hamid Gholami.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Conflict of interest

The authors declare that they have no conflict of interest.

Additional information

Responsible Editor: Marcus Schulz

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Mohammadifar, A., Gholami, H., Golzari, S. et al. Spatial modelling of soil salinity: deep or shallow learning models?. Environ Sci Pollut Res 28, 39432–39450 (2021). https://doi.org/10.1007/s11356-021-13503-7

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11356-021-13503-7

Keywords

Search

Navigation