Enhancing Soil Moisture Estimation: Exploring the Synergy of Optical Trapezoid and Deep Learning models

Authors
G. Zuravand 1
V. Moosavi 1
S. R. Fallah Shamsi 2
1 Department of Watershed Management Engineering, Faculty of Natural resources, Tarbiat Modares University, Tehran, Iran.
2 Department of Natural Resource and Environment, Shiraz University, Iran And Visiting Scientist, Institute of Geographical Science and Natural Resources Research-Chinese Academy of Sciences, Beijing, P.R.China
10.22034/ecopersia.11.3.255
Abstract
Aims: This study aimed to propose an effective model for estimating soil moisture by integrating the optical trapezoid method with a deep learning Long Short-Term Memory (LSTM) model. The performance of the proposed model was compared with two other methods, i.e., Partial Least Squares (PLS) regression and Group Method of Data Handling (GMDH) multivariate neural network.

Materials & Methods: This study combined the optical trapezoid method with deep learning models to propose an effective model for soil moisture estimation in the Maragheh watershed. A total of 499 in-situ soil moisture data were collected. Relative moisture content was calculated using the optical trapezoid method and imported into the LSTM model, along with other inputs such as spectral indices and DEM-based derived variables. The performance of the mentioned models was assessed both with and without the optical trapezoid method to evaluate its efficacy on the performance of AI models.

Findings: The results demonstrate that the combined model of deep learning LSTM and the optical trapezoid method achieves satisfactory performance, with an R2 of 0.95 and a RMSE of 1.7%. The PLS and GMDH methods performed moderately, both without the involvement of the optical trapezoid method and in the combined mode.

Conclusion: This study shows that the optical trapezoid method can improve the performance of deep-learning models in estimating soil moisture. However, considering the significant difference in computational costs among these models, choosing the appropriate model depends on the user's objectives and desired level of accuracy and precision.
Keywords
deep learning
Optical trapezoid method
remote sensing
Sentinel-2
soil moisture

Subjects

1. Rouse J.W., Haas R.H., Schell J.A., Deering D.W. Monitoring vegetation systems in the great plains with ERTS third earth resources technology satellite-1 symposium. NASA Spec 1974; 351: 309-317.
2. Fathalolomi S., Vaezi E.R., Alavi Panah K., Ghorbani A. Modeling the effect of biophysical properties and surface topography on the spatial distribution of soil moisture in summer (case study: Balkhali Chai watershed). ECOHYDROLOGY 2020; 7(3): 563-581. https://doi.org/10.22059/ije.2020.299783.1307
3. Koohbanani H., Yazdani R. Mapping the moisture of surface soil using landsat 8 imagery, (case study: suburb of semnan city). Geogr. Environ. Sustain 2019; 8(3): 65-77. https://dorl.net/dor/20.1001.1.23223197.1397.8.3.5.1
4. Tabatabaeenejad A., Burgin M., Duan X., Moghaddam M. P-band radar retrieval of subsurface soil moisture profile as a second-order polynomial: First AirMOSS results. IEEE Transactions on Geoscience and Remote Sensing 2014; 53(2): 645-658. https://doi.org/10.1109/TGRS.2014.2326839
5. Sanli F.B., Kurucu Y., Esetlili M.T., Abdikana S. Soil moisture estimation from RADARSAT-1, ASAR and PALSAR datain agricultural fiels of menemen plain of western turkey. Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci. Beijing 2008; 75-81.
6. Srivastava H.S., Patel P., Sharma Y., Navalgund R.R. Large-area soil moisture estimation using multi-incidence-angle RADARSAT-1 SAR data. IEEE. T. GEOSCI. REMOTE 2009; 47(8): 2528-2535. https://doi.org/10.1109/TGRS.2009.2018448
7. Silva B.M., Silva S.H.G., Oliveira G.C.d., Peters P.H.C.R., Santos W.J.R.d., Curi N. Soil moisture assessed by digital mapping techniques and its field validation. Cienc. Agrotec 2014; 38(2): 8-140. https://doi.org/10.1590/S1413-70542014000200005
8. Javadi P., Asadi H., Vazife M. Estimation of spatial changes of soil moisture using random forest method and environmental characteristics obtained from satellite images in Marghab basin of Khuzestan. Iran J. Soil Water Res 2022; 52(11): 2859-2874. https://dx.doi.org/10.22059/ijswr.2021.331962.669094
9. Foroughi H., Naseri A.A., Boroomandnasab S., Hamzeh S., Jones S.B. Presenting a new method for soil-moisture estimation using optical remotely-sensed imagery. Iran J. Soil Water Res 2019; 50(3): 641-652. https://doi.org/10.22059/ijswr.2019.266967.668024
10. Shokri Sh., Farrokhian Firoozi A., Babaeian E. Estimation of soil moisture by combining physical and hydraulic characteristics of soil with remote sensing optical data using machine learning method. Iran J. Soil Water Res 2022; 53(7): 1575-1591. http//doi.org/10.22059/ijswr.2022.341492.669243
11. Sadeghi M., Babaeian E., Tuller M., Jones S.B. The optical trapezoid model A novel approach to remote sensing of soil moisture applied to Sentinel-2 and Landsat-8 observations. Remote Sens. Environ 2017; 198: 52-68. https://doi.org/10.1016/j.rse.2017.05.041
12. Sedaghat A., Shabanpoor M., Norozi E.A., Fallah E., Bayat H. Using spectral indices in estimating soil surface moisture based on machine learning algorithm. Iran J. Soil Water Res 2021; 52(12): 3001-3018. https://dx.doi.org/10.22059/ijswr.2022.333856.669130
13. Sedaghat A., Shabanpoor M., Norozi E.A., Fallah E., Bayat H. Modeling soil surface moisture using machine learning models and Sentinel-2 satellite data. The 17th Iranian Congress of Soil Science and the 4th National Conference on Water Management in the Farm, Wise Soil Revival and Wise Water Governance 2021. https://civilica.com/doc/1312606
14. Norozi Aghdam A., Behbahani M.R., Rahimi Khoob E., Aghighi H. Moisture model of the soil surface layer using From the index NDVI (a case study: pastures of Razavi Khorasan province). Environ. Mag 2008; 34(48): 127-136.
15. Norozi Aghdam A., Karami V. Application of remote sensing technology in monitoring and evaluation of irrigation networks drainage in use. The twelfth conference of the National Committee Irrigation and drainage of Iran 2009;
16. Bagheri K., Bagheri M., Hoseynzade E.A., Parvin M. Estimation of soil moisture using optical, thermal and radar remote sensing (case study: lands south of Tehran). Iran. J. Water. Sci. Eng 2019; 13(47): 63-74. http://jwmsei.ir/article-1-853-fa.html
17. Zeyliger A.M., Muzalevskiy K.V., Zinchenko E.V., Ermolaeva O.S. Field test of the surface soil moisture mapping using Sentinel-1 radar data. Sci. Total Environ 2022; 807(2): 121-151. https://doi.org/10.1016/j.scitotenv.2021.151121
18. Ramezani Charmhine E.A., Zonemat Kermani M. Investigating the effectiveness of multi-layer perceptron neural network support vector regression and variable linear regression methods to predict the level of underground water (study area: Shahrekord plain). J. Watershed Manag. Res 2017; 8(15): 1-12. http://jwmr.sanru.ac.ir/article-1-837-fa.html
19. Avazpoor S., Bakhtiari B., Ghaderi K. Investigating the effectiveness of neural network and multivariate regression methods in estimating total solar radiation in several representative stations of arid and semi-arid climates. Iran J. Soil Water Res 2019; 50(8): 1855-1869. https://doi.org/10.22059/ijswr.2019.277373.668142
20. Golkar E., Ahmadi M.M., Qaderi K., Rahimpour M. Peak velocity of pollutant transport prediction in rivers using Group Method Data Handling (GMDH) and intelligent hybrid method (GMDH-HS). J. Water and Waste 2018; 30(1): 64-76. dx.doi.org/10.22093/wwj.2018.83092.2389
21. Godarzi M.R., Godarzi H. Investigating the effectiveness of data group classification method and wavelet transformation in runoff forecasting (study area: Ghareso watershed). Sci. Res. J. Irrig. Water Eng 2020; 10(4): 67-81. https://doi.org/10.22125/iwe.2020.109914
22. Moosavi V., Talebi A., Hadian M.R. Development of a hybrid wavelet packet forecasting group method of data handling (WPGMDH) model for runoff. Water Resour. Manag 2017; 31(1): 43–59. https://doi.org/10.1007/s11269-016-1507-3
23. Adalat M.H., Azmi R., Bagherinejad J. An enhanced LSTM. Method to improve the accuracy of the business process prediction. Indus. Manag. Perspec 2020; 10(39): 71-97. https://doi.org/10.52547/jimp.10.3.71
24. Jordan M. A. Parallel distributed processing approach. Adv. PSYCHOL 1997; 121: 471-495. https://doi.org/10.1016/S0166-4115(97)80111-2
25. Yu M., Xu F., Xu W., Sun J., Cervone G. Using long short-term memory (LSTM) and internet of things (IoT) for localized surface temperature forecasting in an urban environment. IEEE Access 2021; 9: 137406-137418. https://doi.org/10.1109/ACCESS.2021.3116809
26. Liu Z.H., Meng X.D., Wei H.L., Chen L, Lu B.L., Wang Z.H., Chen L. A regularized LSTM method for predicting remaining useful life of rolling bearings. International Journal of Automation and Computing 2021; 18(4): 581-593. https://doi.org/10.1007/s11633-020-1276-6
27. Haddadian H., Karimi E., Esfandiar poor E., Hagh nia Gh. RMSE and NRMSE to calculate the effective value of the error of models with different training data sets. The 14th Congress of Soil Sciences of Iran Vali Asr University of Rafsanjan 2015; 16-18 September.
28. Ahmadinezhad Baghban F., Moosavi V. Convolutional neural networks (CNN)-signal processing combination for daily runoff forecasting. ECOPERSIA 2022; 10(3): 231-243. http://dorl.net/dor/20.1001.1.23222700.2022.10.3.6.3
29. Zhou Y., Zhang Y., Wang R., Chen H., Zhao Q., Liu B., Shao Q., Cao L., Sun S. Deep learning for daily spatiotemporally continuity of satellite surface soil moisture over eastern china in summer. J. of Hydrol 2023; 619: 129308. https://doi.org/10.1016/j.jhydrol.2023.129308
30. Yinglan A., Wang G., Hu P., Lai P., Xue B., Fang Q. Root-zone soil moisture estimation based on remote sensing data and deep learning. Environ. Res 2022; 212: 113278. https://doi.org/10.1016/j.envres.2022.113278
31. Achieng K.O. Modelling of soil moisture retention curve using machine learning techniques: Artificial and deep neural networks vs support vector regression models. Comput and Geosci 2019; 133: 104320. https://doi.org/10.1016/j.cageo.2019.104320
32. Joshi R.C., Ryu D., Lane P.N.J., Sheridan G.J. Seasonal forecast of soil moisture over mediterranean-climate forest catchments using a machine learning approach. J. of Hydrol 2023;619: 12930. https://doi.org/10.1016/j.jhydrol.2023.129307
ECOPERSIA
Volume 11, Issue 3 - Serial Number 45
September 2023
Pages 255-274