Abstract
Surface soil moisture (MSS) is a key factor governing environmental interactions in any catchment. Energy flux between soil and atmosphere, soil temperature, and heat diffusion in soil are examples of impressible interactions. Consequently, the agriculture sector and its many dependent industries are influenced by this element. Hence, investigating new optimized preprocessing and input selection methods for processing, interpretation, modeling, and prediction of MSS is necessary to ensure sustainable agriculture. To this end, satellite products were studied for the province of Quebec, Canada. Two overall deep learning (DL) approaches are proposed in this study. The first and most efficient one is extracting meaningful modeling parameters by the time-series structural-analysis-based method, and the second one is using the combination of optimization algorithms and the DL method. The structure of the extracted time series from satellite data was assessed by several tests and an intense periodic pattern was detected. Therefore, additive Holt–Winter’s (SHW), seasonal standardization (Sstd), and spectral analysis (SA) were chosen as preprocessing methods for the structural analysis preprocessing. The long short-term memory (LSTM) model was utilized for short-term forecasting of un-preprocessed and preprocessed MSS datasets. Along with structural-analysis-based methodology, genetic and teacher–learner-based algorithms (GA and TLA) were coupled with LSTM to assess the coupled models’ performance for MSS forecasting for the first time. Based on the structural analysis of data, limited hidden states (ht) were selected for modeling {1, 2, 7, 9, 52}: network training and forecasts were undertaken according to these hidden states. Since the long-term characteristics of the time series like trend and level are not significant in short-term modeling, the LSTM (Sstd, 9), correlation coefficient (R) = 0.970, root-mean-square error (RMSE) = 1.339 outperformed other models, followed closely by LSTM (SHW, 1), R = 0.922, RMSE = 1.958. Conversely, for long-term forecast, as these attributes impact the structure, LSTM (SHW, 2), R = 0.922, RMSE = 0.1961 was more successful in the prediction of patterns and fluctuations, followed by LSTM (Sstd, 52), R = 0.920, RMSE = 2.064, which was more complicated than the model developed for short-term modeling. GA-LSTM (ht = 32, R = 0.930, RMSE = 1.852) and TLA-LSTM (ht = 37, R = 0.934, RMSE = 1.781) also enhanced the long-term forecasting results. Integration of these two optimization methods had two benefits. First, due to the stochastic nature of optimization algorithms and DL methods, the search space for the optimized parameter (ht) was greatly increased and many possibilities were investigated. Second, the LSTM could perform a long-term forecast of the MSS without preprocessing, which was not possible by structural analysis. On the other hand, these methods were much computationally expensive and the combination of their controlling parameters with other controlling parameters of LSTM created numerous possibilities. However, as TLA is parameter free and much less sophisticated than GA, it is a more computational-effective method, and subsequently a better option than GA.
Similar content being viewed by others
References
Vereecken H, Huisman JA, Pachepsky Y et al (2014) On the spatio-temporal dynamics of soil moisture at the field scale. J Hydrol 516:76–96
Petropoulos GP, Ireland G, Barrett B (2015) Surface soil moisture retrievals from remote sensing: Current status, products & future trends. Phys Chem Earth, Parts A/B/C 83–84:36–56. https://doi.org/10.1016/j.pce.2015.02.009
Ochsner TE, Horton R, Ren T (2001) A New Perspective on Soil Thermal Properties. Soil Sci Soc Am J 65:1641–1647. https://doi.org/10.2136/sssaj2001.1641
Arkhangel’skaya TA, Umarova AB (2008) Thermal diffusivity and temperature regime of soils in large lysimeters of the experimental soil station of Moscow State University. Eurasian Soil Sc 41:276–285
Schindlbacher A (2004) Effects of soil moisture and temperature on NO, NO 2, and N 2 O emissions from European forest soils. J Geophys Res 109:1137. https://doi.org/10.1029/2004JD004590
Wei S, Zhang X, McLaughlin NB et al (2014) Effect of soil temperature and soil moisture on CO2 flux from eroded landscape positions on black soil in Northeast China. Soil Tillage Res 144:119–125. https://doi.org/10.1016/j.still.2014.07.012
Torres-Rua A, Ticlavilca A, Bachour R et al (2016) Estimation of surface soil moisture in irrigated lands by assimilation of landsat vegetation indices, surface energy balance products, and relevance vector machines. Water 8:167. https://doi.org/10.3390/w8040167
Panikov NS, Flanagan PW, Oechel WC et al (2006) Microbial activity in soils frozen to below−39 C. Soil Biol Biochem 38:785–794
Petropoulos GP (ed) (2014) Remote sensing of energy fluxes and soil moisture content. Taylor & Francis, Boca Raton
McNab WH (ed) (1991) Factors affecting temporal and spatial soil moisture variation in and adjacent to group selection openings, vol 148
Famiglietti JS, Rudnicki JW, Rodell M (1998) Variability in surface moisture content along a hillslope transect: Rattlesnake Hill, Texas. J Hydrol 210:259–281. https://doi.org/10.1016/S0022-1694(98)00187-5
Yoo C, Kim S (2004) EOF analysis of surface soil moisture field variability. Adv Water Resour 27:831–842. https://doi.org/10.1016/j.advwatres.2004.04.003
Hawley ME, Jackson TJ, McCuen RH (1983) Surface soil moisture variation on small agricultural watersheds. J Hydrol 62:179–200. https://doi.org/10.1016/0022-1694(83)90102-6
Entekhabi D, Rodriguez-Iturbe I (1994) Analytical framework for the characterization of the space-time variability of soil moisture. Adv Water Resour 17:35–45. https://doi.org/10.1016/0309-1708(94)90022-1
Crave A, Gascuel-Odoux C (1997) The influence of topography on time and space distribution of soil surface water content. Hydrol Process 11:203–210
Li Q, Li Z, Shangguan W et al (2022) Improving soil moisture prediction using a novel encoder-decoder model with residual learning. Comput Electron Agric 195:106816. https://doi.org/10.1016/j.compag.2022.106816
Li Q, Zhu Y, Shangguan W et al (2022) An attention-aware LSTM model for soil moisture and soil temperature prediction. Geoderma 409:115651. https://doi.org/10.1016/j.geoderma.2021.115651
Abbes AB, Magagi R, Goita K (eds) (2019) Soil Moisture Estimation From Smap Observations Using Long Short-Term Memory (LSTM). IEEE
Fang K, Pan M, Shen C (2018) The value of SMAP for long-term soil moisture estimation with the help of deep learning. IEEE Trans Geosci Remote Sensing 57:2221–2233
Mukhlisin M, El-shafie A, Taha MR (2012) Regularized versus non-regularized neural network model for prediction of saturated soil-water content on weathered granite soil formation. Neural Comput Applic 21:543–553. https://doi.org/10.1007/s00521-011-0545-2
Hamouda YEM, Msallam MM (2019) Smart heterogeneous precision agriculture using wireless sensor network based on extended Kalman filter. Neural Comput Applic 31:5653–5669. https://doi.org/10.1007/s00521-018-3386-4
Keswani B, Mohapatra AG, Mohanty A et al (2019) Adapting weather conditions based IoT enabled smart irrigation technique in precision agriculture mechanisms. Neural Comput Applic 31:277–292. https://doi.org/10.1007/s00521-018-3737-1
Zaji AH, Bonakdari H, Gharabaghi B (2018) Remote sensing satellite data preparation for simulating and forecasting river discharge. IEEE Trans Geosci Remote Sensing 56:3432–3441. https://doi.org/10.1109/TGRS.2018.2799901
Zaji AH, Bonakdari H, Gharabaghi B (2019) Applying upstream satellite signals and a 2-D error minimization algorithm to advance early warning and management of flood water levels and river discharge. IEEE Trans Geosci Remote Sensing 57:902–910. https://doi.org/10.1109/TGRS.2018.2862640
Moreira AA, Ruhoff AL, Roberti DR et al (2019) Assessment of terrestrial water balance using remote sensing data in South America. J Hydrol 575:131–147
Bonakdari H, Moeeni H, Ebtehaj I et al (2019) New insights into soil temperature time series modeling: linear or nonlinear? Theor Appl Climatol 135:1157–1177. https://doi.org/10.1007/s00704-018-2436-2
Zeltner N (2016) Using the Google earth engine for global glacier change assessment. Geographisches Institut der Universität Zürich, Zürich
Abou El-Magd IH, Ali EM (2012) Estimation of the evaporative losses from Lake Nasser, Egypt using optical satellite imagery. Int J Digital Earth 5:133–146
Pekel J-F, Cottam A, Gorelick N et al (2016) High-resolution mapping of global surface water and its long-term changes. Nature 540:418–422. https://doi.org/10.1038/nature20584
Entekhabi D, Njoku EG, O’Neill PE et al (2010) The soil moisture active passive (SMAP) mission. Proc IEEE 98:704–716
Das NN, Entekhabi D, Dunbar RS et al (2019) The SMAP and Copernicus Sentinel 1A/B microwave active-passive high resolution surface soil moisture product. Remote Sens Environ 233:111380
Colliander A, Jackson TJ, Bindlish R et al (2017) Validation of SMAP surface soil moisture products with core validation sites. Remote Sens Environ 191:215–231
Gorelick N, Hancher M, Dixon M et al (2017) Google earth engine: planetary-scale geospatial analysis for everyone. Remote Sens Environ 202:18–27. https://doi.org/10.1016/j.rse.2017.06.031
Sazib N, Mladenova I, Bolten J (2018) Leveraging the google earth engine for drought assessment using global soil moisture data. Remote Sensing 10:1265. https://doi.org/10.3390/rs10081265
Gholami A, Bonakdari H, Zaji AH et al (2019) An efficient classified radial basis neural network for prediction of flow variables in sharp open-channel bends. Appl Water Sci 9:1–17
Gholami A, Bonakdari H, Zaji AH et al (2020) A comparison of artificial intelligence-based classification techniques in predicting flow variables in sharp curved channels. Engineering with Computers 36:295–324
Ebtehaj I, Bonakdari H, Zaji AH et al (2015) Gene expression programming to predict the discharge coefficient in rectangular side weirs. Appl Soft Comput 35:618–628
Zeynoddin M, Bonakdari H, Azari A et al (2018) Novel hybrid linear stochastic with non-linear extreme learning machine methods for forecasting monthly rainfall a tropical climate. J Environ Manage 222:190–206
Azari A, Zeynoddin M, Ebtehaj I et al (2021) Integrated preprocessing techniques with linear stochastic approaches in groundwater level forecasting. Acta Geophys 6:472. https://doi.org/10.1007/s11600-021-00617-2
Kumar D, Singh A, Samui P et al (2019) Forecasting monthly precipitation using sequential modelling. Hydrol Sci J 64:690–700. https://doi.org/10.1080/02626667.2019.1595624
Kowtha V, Mitra V, Bartels C et al. (2020) Detecting Emotion Primitives from Speech and their use in discerning Categorical Emotions
Ha J-H, Lee YH, Kim Y-H (2016) Forecasting the precipitation of the next day using deep learning. J Korean Inst Intell Syst 26:93–98
Kratzert F, Klotz D, Brenner C et al (2018) Rainfall–runoff modelling using long short-term memory (LSTM) networks. Hydrol Earth Syst Sci 22:6005–6022
Shi X, Chen Z, Wang H et al. (2015) Convolutional LSTM network: A machine learning approach for precipitation nowcasting. arXiv preprint arXiv:1506.04214
Marini A, Termite LF, Garinei A et al (2020) Neural network models for soil moisture forecasting from remote sensed measurements. ACTA IMEKO 9:59–65
Nahvi B, Habibi J, Mohammadi K et al (2016) Using self-adaptive evolutionary algorithm to improve the performance of an extreme learning machine for estimating soil temperature. Comput Electron Agric 124:150–160. https://doi.org/10.1016/j.compag.2016.03.025
Gholami A, Bonakdari H, Zeynoddin M et al (2019) Reliable method of determining stable threshold channel shape using experimental and gene expression programming techniques. Neural Comput Appl 31:5799–5817
Zeynoddin M, Ebtehaj I, Bonakdari H (2020) Development of a linear based stochastic model for daily soil temperature prediction: One step forward to sustainable agriculture. Comput Electron Agric 176:105636. https://doi.org/10.1016/j.compag.2020.105636
Shin Y, Mohanty BP, Ines AVM (2018) Development of non-parametric evolutionary algorithm for predicting soil moisture dynamics. J Hydrol 564:208–221. https://doi.org/10.1016/j.jhydrol.2018.07.003
Zhang F, Wu S, Liu J et al (2021) Predicting soil moisture content over partially vegetation covered surfaces from hyperspectral data with deep learning. Soil Sci Soc Am J 85:989–1001. https://doi.org/10.1002/saj2.20193
Ma Z, Mei G, Piccialli F (2021) Machine learning for landslides prevention: a survey. Neural Comput Applic 33:10881–10907. https://doi.org/10.1007/s00521-020-05529-8
Wu CL, Chau KW, Fan C (2010) Prediction of rainfall time series using modular artificial neural networks coupled with data-preprocessing techniques. J Hydrol 389:146–167. https://doi.org/10.1016/j.jhydrol.2010.05.040
Ouyang Q, Lu W (2018) Monthly rainfall forecasting using echo state networks coupled with data preprocessing methods. Water Resour Manage 32:659–674. https://doi.org/10.1007/s11269-017-1832-1
Stajkowski S, Kumar D, Samui P et al (2020) Genetic-Algorithm-Optimized sequential model for water temperature prediction. Sustainability 12:5374
Bouktif S, Fiaz A, Ouni A et al (2018) Optimal deep learning LSTM model for electric load forecasting using feature selection and genetic algorithm: comparison with machine learning approaches †. Energies 11:1636. https://doi.org/10.3390/en11071636
Maroufi H, Mehdinejadiani B (2021) A comparative study on using metaheuristic algorithms for simultaneously estimating parameters of space fractional advection-dispersion equation. J Hydrol 602:126757. https://doi.org/10.1016/j.jhydrol.2021.126757
Bozorg-Haddad O, Sarzaeim P, Loáiciga HA (2021) Developing a novel parameter-free optimization framework for flood routing. Sci Rep 11:16183. https://doi.org/10.1038/s41598-021-95721-0
Ebrahimi M, Alavipanah SK, Hamzeh S et al (2018) Exploiting the synergy between SMAP and SMOS to improve brightness temperature simulations and soil moisture retrievals in arid regions. J Hydrol 557:740–752. https://doi.org/10.1016/j.jhydrol.2017.12.051
O'Neill P, Bindlish R, Chan S et al. (2018) Algorithm Theoretical Basis Document. Level 2 & 3 Soil Moisture (Passive) Data Products
Jonard F, Bircher S, Demontoux F et al (2018) Passive L-band microwave remote sensing of organic soil surface layers: a tower-based experiment. Remote Sensing 10:304. https://doi.org/10.3390/rs10020304
Dong J, Crow WT, Tobin KJ et al (2020) Comparison of microwave remote sensing and land surface modeling for surface soil moisture climatology estimation. Remote Sens Environ 242:111756. https://doi.org/10.1016/j.rse.2020.111756
Jackson TJ, Schmugge TJ (1991) Vegetation effects on the microwave emission of soils. Remote Sensing of Environment
Basharinov AY, Am Shutko (1975) Simulation studies of the SHF radiation characteristics of soils under moist conditions. sssr:1
Ulaby FT, Moore RK, Fung AK (1981) Microwave remote sensing: Active and passive. volume 1-microwave remote sensing fundamentals and radiometry. Addison-Wesley Publishing Company, Advanced Book Program/World Science Division
Mladenova IE, Bolten JD, Crow W et al (2020) Agricultural drought monitoring via the assimilation of SMAP soil moisture retrievals into a global soil water balance model. Front Big Data 3:10. https://doi.org/10.3389/fdata.2020.00010
Mladenova IE, Bolten JD, Crow WT et al (2019) Evaluating the operational application of SMAP for global agricultural drought monitoring. IEEE J Sel Top Appl Earth Obs Remote Sensing 12:3387–3397. https://doi.org/10.1109/JSTARS.2019.2923555
Hamilton JA, Nash DA, Pooch UW (1997) Distributed simulation. CRC Press, Boca Raton
Zeynoddin M, Bonakdari H (2019) Investigating methods in data preparation for stochastic rainfall modeling: a case study for Kermanshah synoptic station rainfall data, Iran. J Appl Res Water Wastewater 6:32–38
Tekleab SG, Am Kassew (2019) Hydrologic responses to land use/Land cover change in the Kesem Watershed, Awash basin, Ethiopia. Journal of Spatial Hydrology 15
Mann HB, Whitney DR (1947) On a test of whether one of two random variables is stochastically larger than the other. Ann Math Statist 18:50–60. https://doi.org/10.1214/aoms/1177730491
Kwiatkowski D, Phillips PC, Schmidt P et al (1992) Testing the null hypothesis of stationarity against the alternative of a unit root. J Econ 54:159–178. https://doi.org/10.1016/0304-4076(92)90104-y
Moeeni H, Bonakdari H (2017) Forecasting monthly inflow with extreme seasonal variation using the hybrid SARIMA-ANN model. Stoch Environ Res Risk Assess 31:1997–2010. https://doi.org/10.1007/s00477-016-1273-z
Yaseen ZM, Ebtehaj I, Bonakdari H et al (2017) Novel approach for streamflow forecasting using a hybrid ANFIS-FFA model. J Hydrol 554:263–276. https://doi.org/10.1016/j.jhydrol.2017.09.007
Yaseen ZM, Ebtehaj I, Kim S et al (2019) Novel hybrid data-intelligence model for forecasting monthly rainfall with uncertainty analysis. Water 11:502
Puah YJ, Huang YF, Chua KC et al (2016) River catchment rainfall series analysis using additive Holt–Winters method. J Earth Syst Sci 125:269–283
Ebtehaj I, Bonakdari H, Zeynoddin M et al (2020) Evaluation of preprocessing techniques for improving the accuracy of stochastic rainfall forecast models. Int J Environ Sci Technol 17:505–524
Sak H, Senior A, Beaufays F (2014) Long short-term memory based recurrent neural network architectures for large vocabulary speech recognition
Graves A, Liwicki M, Fernández S et al (2009) A novel connectionist system for unconstrained handwriting recognition. IEEE Trans Pattern Anal Mach Intell 31:855–868. https://doi.org/10.1109/TPAMI.2008.137
Gers FA, Schmidhuber J, Cummins F (2000) Learning to forget: continual prediction with LSTM. Neural Comput 12:2451–2471. https://doi.org/10.1162/089976600300015015
Razvan Pascanu, Tomas Mikolov, Yoshua Bengio (2013) On the difficulty of training recurrent neural networks. In: Sanjoy Dasgupta, David McAllester (eds) Proceedings of the 30th International Conference on Machine Learning. PMLR, pp 1310–1318
Bengio Y, Simard P, Frasconi P (1994) Learning long-term dependencies with gradient descent is difficult. IEEE Trans Neural Netw 5:157–166. https://doi.org/10.1109/72.279181
Gao P, Xie J, Yang M et al (2021) Improved soil moisture and electrical conductivity prediction of citrus orchards based on IoT using deep bidirectional LSTM. Agriculture 11:635. https://doi.org/10.3390/agriculture11070635
Cordeiro M, Markert C, Araújo SS et al (2022) Towards Smart Farming: Fog-enabled intelligent irrigation system using deep neural networks. Futur Gener Comput Syst 129:115–124. https://doi.org/10.1016/j.future.2021.11.013
Leng J (2016) Optimization techniques for structural design of cold-formed steel structures. In: Recent Trends in Cold-Formed Steel Construction, vol 123. Elsevier, pp 129–151
Gholizadeh S (2013) Structural Optimization for Frequency Constraints. In: Metaheuristic Applications in Structures and Infrastructures, vol 29. Elsevier, pp 389–417
Angelova M, Pencheva T (2011) Tuning genetic algorithm parameters to improve convergence time. Int J Chem Eng 2011:1–7. https://doi.org/10.1155/2011/646917
Donate JP, Li X, Sánchez GG et al (2013) Time series forecasting by evolving artificial neural networks with genetic algorithms, differential evolution and estimation of distribution algorithm. Neural Comput Applic 22:11–20. https://doi.org/10.1007/s00521-011-0741-0
Liang R, Ding Y, Zhang X et al. A real-time prediction system of soil moisture content using genetic neural network based on annealing algorithm. In: 2008 IEEE International Conference on Automation and Logistics. IEEE, pp 2781–2785
Azamathulla HMd, Wu F-C, Ghani AAb et al (2008) Comparison between genetic algorithm and linear programming approach for real time operation. J Hydro-Environ Res 2:172–181. https://doi.org/10.1016/j.jher.2008.10.001
Rao RV, Savsani VJ, Vakharia DP (2011) Teaching–learning-based optimization: a novel method for constrained mechanical design optimization problems. Comput Aided Des 43:303–315. https://doi.org/10.1016/j.cad.2010.12.015
Zeinolabedini Rezaabad M, Ghazanfari S, Salajegheh M (2020) ANFIS modeling with ICA, BBO, TLBO, and IWO optimization algorithms and sensitivity analysis for predicting daily reference evapotranspiration. J Hydrol Eng 25:4020038
Nash JE, Sutcliffe JV (1970) River flow forecasting through conceptual models part I—a discussion of principles. J Hydrol 10:282–290. https://doi.org/10.1016/0022-1694(70)90255-6
Government of Canada (2019) Historical Climate Data. https://climate.weather.gc.ca/
Zheng D, Wang X, van der Velde R et al (2018) Impact of surface roughness, vegetation opacity and soil permittivity on L-band microwave emission and soil moisture retrieval in the third pole environment. Remote Sens Environ 209:633–647. https://doi.org/10.1016/j.rse.2018.03.011
Bechtold B (2019) Violin plots for MATLAB. GitHub. https://doi.org/10.5281/zenodo.4559847
Singh VP, Frevert DK (2002) Mathematical models of small watershed hydrology and applications. Water Resources Publication
Kingma DP, Ba J (2014) Adam: a method for stochastic optimization
Fan Y, van den Dool H (2004) Climate prediction center global monthly soil moisture data set at 0.5 resolution for 1948 to present. J Geophys Res 109:549. https://doi.org/10.1029/2003JD004345
McNally A, Arsenault K, Kumar S et al (2017) A land data assimilation system for sub-Saharan Africa food and water security applications. Sci Data 4:170012. https://doi.org/10.1038/sdata.2017.12
Svoboda M, LeComte D, Hayes M et al (2002) THE DROUGHT MONITOR. Bull Am Meteor Soc 83:1181–1190. https://doi.org/10.1175/1520-0477-83.8.1181
Dong J, Steele-Dunne SC, Ochsner TE et al (2015) Determining soil moisture by assimilating soil temperature measurements using the Ensemble Kalman Filter. Adv Water Resour 86:340–353. https://doi.org/10.1016/j.advwatres.2015.08.011
Lu S, Ju Z, Ren T et al (2009) A general approach to estimate soil water content from thermal inertia. Agric For Meteorol 149:1693–1698. https://doi.org/10.1016/j.agrformet.2009.05.011
Steele-Dunne SC, Rutten MM, Krzeminska DM et al (2010) Feasibility of soil moisture estimation using passive distributed temperature sensing. Water Resour Res 46:234. https://doi.org/10.1029/2009WR008272
Acknowledgements
The authors acknowledge the financial support provided by the Natural Science and Engineering Research Council of Canada (NSERC) Discovery Grant (#RGPIN-2020-04583)
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that there is no conflict of interest regarding publishing this paper.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix
Appendix
The SMAP soil moisture data sets are available at https://explorer.earthengine.google.com/#detail/NASA_USDA%2FHSL%2FSMAP_soil_moisture.
The networks’ files are provided along with the manuscript.
Rights and permissions
About this article
Cite this article
Zeynoddin, M., Bonakdari, H. Structural-optimized sequential deep learning methods for surface soil moisture forecasting, case study Quebec, Canada. Neural Comput & Applic 34, 19895–19921 (2022). https://doi.org/10.1007/s00521-022-07529-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-022-07529-2