# Improved Soil Moisture and Electrical Conductivity Prediction of Citrus Orchards Based on IoT Using Deep Bidirectional LSTM

^{1}

^{2}

^{3}

^{4}

^{*}

## Abstract

**:**

^{2}) were calculated to evaluate the performance of the models. The performance of the deep Bid-LSTM model was compared with a multi-layer neural network (MLNN). The results for the performance criteria reveal that the proposed deep Bid-LSTM networks perform better than the MLNN model, according to many of the evaluation indicators of this study.

## 1. Introduction

## 2. Materials and Methods

#### 2.1. Study Area

^{2}, and the planting variety was “Shatangju” (Citrus reticulata Blanco). There were five sections in the orchard, and drip irrigation pipelines were uniformly arranged. Since the water pressure of the irrigation pipelines in the orchard was limited, in order to ensure a good effect on irrigation and fertilizer tasks, each section was carried out separately. It typically took about 3–7 days to irrigate all of the sections, which also required consideration of water storage. If the water supply was not guaranteed in time, the process might take longer. Therefore, it is necessary to provide predicted SM and SEC data to the managers as references in a period of time. The row spacing of the fruit trees is 4 m, the plant spacing is 3 m, and the average plant height is 2 m. The soil in this area is mainly sandy loam, with high temperatures and rain in the summer. The annual rainfall of this area is 1700 mm, while there is less rainfall in winter, meaning that it is prone to drought in this period. The citrus orchard is in hilly terrain [17] with complex environmental conditions, and interference with wireless data transmission occurs regularly, which may cause packet loss.

#### 2.2. IoT System

#### 2.2.1. Structure of System

#### 2.2.2. Hardware of IoT System

^{2}catchment, meaning that $3.74\times {10}^{-8}$ sensor·m

^{−2}was enough to monitor the environmental information. Considering the high cost of sensors, five nodes deployed in the five sections of the 66,000 m

^{2}study area were enough to measure the SM and SEC.

#### 2.2.3. Software of IoT System

#### 2.3. Data Preprocessing and Correlation Analysis

_{i}is the difference between the two ranks of each observation, and n is the number of observations.

#### 2.4. Deep Bidirectional LSTM Networks

_{2.5}[27]. The sequence data, i.e., time-series data, are not only related to the current conditions but also have a high correlation with the previous data; they are difficult to model with a traditional neural network. In recent years, the recurrent neural network (RNN) [28] has been successfully applied in natural language processing, machine translation, and text prediction, showing a better result than a neural network. The RNN differs from general feedforward networks because it contains a hidden state extracting feature and then transforms to output data. On the other hand, RNN shares the weights, biases, etc., in all calculation steps, which makes it dependent on the connected nodes [29]. However, the standard RNN also has its disadvantages; for example, the input and output data are the same size. In addition, the hidden layer state of each layer of the RNN is obtained by the transformation and activation function of the hidden state of the previous layer, making it easy for a gradient vanishing and gradient exploding problem to occur when the RNNs take the derivative of backpropagation at certain steps [30]. Therefore, the standard RNN does not perform well enough to deal with long-term dependence problems [31].

_{t}, cell state S

_{t}, temporary cell state c

_{t}, input control i

_{t}, hidden layer state h

_{t}, forget gate f

_{t}, and output gate o

_{t}. The equations used in this method are as follows:

_{f}and b

_{f}are the weight and bias parameters, respectively. ${h}_{t-1}$ and ${x}_{t}$ are the previous hidden states and input values, respectively. The total length of $\left[{h}_{t-1},{x}_{t}\right]$ is the sum of the length of ${h}_{t-1}\mathrm{a}\mathrm{n}\mathrm{d}{x}_{t}$, respectively. The second part is the input gate with the sigmoid and tanh functions to control how much information should be received in the current cell calculated by Equations (4) and (5).

#### 2.5. Multi-Layer Neural Network (MLNN)

#### 2.6. Performance Criteria of the Models

^{2}) between the predicted value of the models and the measured value was calculated to describe the interpretability of the model. The equations are as follows:

## 3. Results

#### 3.1. Model Training Settings

#### 3.2. Performance of Models

^{2}was used to describe the interpretability of models, i.e., fitting with linear regression between the predicted value and measured value. The closer R

^{2}was to 1, the better the performance of the model [45]. As we can see from Figure 12, in terms of fitting SM, the R

^{2}range of the MLNN model was 0.778–0.865, while the range of the Bid-LSTM model was 0.884–0.977. From each dataset, we can see that the R

^{2}performance of the Bid-LSTM model in node datasets 1, 2, 3, and 5 was much better than that of the MLNN model because the R

^{2}of the former was over 0.9, while the latter was less than 0.9. Additionally, the R

^{2}values of the two models in node four were close to each other, indicating that the reliability of the Bid-LSTM model was generally better than that of the MLNN model. As for SEC, the R

^{2}performances of the two models were similar; however, the fitting ability of the MLNN model for SEC was worse than that of Bid-LSTM compared to its prediction performance for SM because the R

^{2}of the MLNN model showed three of five results that were less than 0.8 (Figure 12). Thus, one can conclude that the Bid-LSTM model performed better than the MLNN model in most situations.

#### 3.3. Performance of Model Fitting

^{2}was higher than that of the MLNN model, meaning that the reliability of the Bid-LSTM model was higher. According to Figure 11, the RMSE of the Bid-LSTM model in the node five dataset was higher, while the R

^{2}was lower, showing that a single RMSE criterion was not completely equivalent to the performance of the model.

^{−1}. The SEC prediction of the MLNN model on the node one dataset had obvious large fluctuations, with a maximum predicted SEC of 104.3 μS·cm

^{−1}on day 64. Compared with the MLNN model, the Bid-LSTM model performed better on the node one dataset because it did not provide a predicted SEC that was far outside of the normal range. The fitting results of datasets 2, 3, and 4 showed that the predicted values of the SEC of the two models showed some fluctuations after fertilization and irrigation activity on the 26th day, meaning that human operation had a negative impact on model prediction. After the 52nd day, both models showed deviation, and the largest deviation was more than 100%. As can be seen from Figure 14 and Figure 16, the MLNN model had a lower R

^{2}when predicting SEC. The outliers in the fitted scatter diagrams of a2–e2 in Figure 14 and Figure 16 may be because of the complex environmental factors in the citrus orchard causing stochastic errors in the sensors when measuring SEC. Both models presented good results in predicting SM and SEC in dataset five, indicating that it was easier to obtain an ideal modeling result.

#### 3.4. Method of Model Selection

## 4. Discussion

^{2}values of the model were 0.92, 0.96, and 0.98 for the three sites, respectively, showing a performance similar to this paper. The difference was that our environmental data came from the real-time IoT system installed in the citrus orchard, which had better adaptability and flexibility for the citrus orchard. Fang et al. [51] applied SM Active Passive (SMAP) data to predict the SM of the US with an optimized LSTM model. Although they provided RMSE results with a minimum of 0 and a maximum of 1, the study focused on the macroscopic SM study, and their results were based on the prediction performance of the model between one and three days. Although the RMSE of that study showed higher performance than the model in our paper, we focused more on the SM prediction modeling at the micro-level of orchards. Hateffard et al. [52] used an artificial neural network to model topsoil SEC, which gave a minimum RMSE and maximum R

^{2}of 6.27 and 0.95, respectively. Their results are similar to our model, but we obtained a lower RMSE in most situations. According to the model selection criteria given in Table 2, the Bid-LSTM model showed good results for RMSE, R

^{2}, and other performance aspects, and the modeling and prediction of SM and SEC were sufficient to meet the practical requirements.

## 5. Conclusions

- The IoT system built in this paper aimed to collect environmental information, including the SM, SEC, ST, air temperature, air humidity, wind speed, and precipitation.
- Compared to the predicted values and measured values using regression fitting, the Bid-LSTM model showed better performance than the MLNN model, even though the former model showed a higher deviation in a few cases due to the negative impact of environmental factors. The R
^{2}criteria showed that the Bid-LSTM model was more reliable than the MLNN model. - The AIC values showed that the Bid-LSTM model was reliable in most situations compared with the MLNN model.

## Author Contributions

## Funding

## Institutional Review Board Statement

## Data Availability Statement

## Acknowledgments

## Conflicts of Interest

## References

- Licciardello, G.; Ferraro, R.; Scuderi, G.; Russo, M.; Catara, A.F. A Simulation of the Use of High Throughput Sequencing as Pre-Screening Assay to Enhance the Surveillance of Citrus Viruses and Viroids in the EPPO Region. Agriculture
**2021**, 11, 400. [Google Scholar] [CrossRef] - Huang, R.; Yao, T.; Zhan, C.; Zhang, G.; Zheng, Y. A Motor-Driven and Computer Vision-Based Intelligent E-Trap for Monitoring Citrus Flies. Agriculture
**2021**, 11, 460. [Google Scholar] [CrossRef] - Kourgialas, N.N.; Karatzas, G.P. A Modeling Approach for Agricultural Water Management in Citrus Orchards: Cost-Effective Irrigation Scheduling and Agrochemical Transport Simulation. Environ. Monit. Assess.
**2015**, 187, 462. [Google Scholar] [CrossRef] [PubMed] - Deng, X.; Huang, Z.; Zheng, Z.; Lan, Y.; Dai, F. Field Detection and Classification of Citrus Huanglongbing Based on Hyperspectral Reflectance. Comput. Electron. Agric.
**2019**, 167, 105006. [Google Scholar] [CrossRef] - Pereira, L.S.; Paredes, P.; Jovanovic, N. Soil Water Balance Models for Determining Crop Water and Irrigation Requirements and Irrigation Scheduling Focusing on the FAO56 Method and the Dual Kc Approach. Agric. Water Manag.
**2020**, 241, 106357. [Google Scholar] [CrossRef] - Panigrahi, P.; Srivastava, A.K. Effective Management of Irrigation Water in Citrus Orchards under a Water Scarce Hot Sub-Humid Region. Sci. Hortic.
**2016**, 210, 6–13. [Google Scholar] [CrossRef] - Jin, X.; Chen, M.; Fan, Y.; Yan, L.; Wang, F. Effects of Mulched Drip Irrigation on Soil Moisture and Groundwater Recharge in the Xiliao River Plain, China. Water
**2018**, 10, 1755. [Google Scholar] [CrossRef] [Green Version] - García-Tejero, I.; Jiménez-Bocanegra, J.A.; Martínez, G.; Romero, R.; Durán-Zuazo, V.H.; Muriel-Fernández, J.L. Positive Impact of Regulated Deficit Irrigation on Yield and Fruit Quality in a Commercial Citrus Orchard [Citrus Sinensis (L.) Osbeck, Cv. Salustiano]. Agric. Water Manag.
**2010**, 97, 614–622. [Google Scholar] [CrossRef] - Huang, J.; Scudiero, E.; Choo, H.; Corwin, D.L.; Triantafilis, J. Mapping Soil Moisture across an Irrigated Field Using Electromagnetic Conductivity Imaging. Agric. Water Manag.
**2016**, 163, 285–294. [Google Scholar] [CrossRef] - Yu, G.; Wang, W.; Xie, J.; Lu, H.; Lin, J.; Mo, H. Information Acquisition and Expert Decision System in Litchi Orchard Based on Internet of Things. Trans. Chin. Soc. Agric. Eng.
**2016**, 32, 144–152. [Google Scholar] - Zhang, X.; Zhang, J.; Li, L.; Zhang, Y.; Yang, G. Monitoring Citrus Soil Moisture and Nutrients Using an IoT Based System. Sensors
**2017**, 17, 447. [Google Scholar] [CrossRef] - Sawant, S.; Durbha, S.S.; Jagarlapudi, A. Interoperable Agro-Meteorological Observation and Analysis Platform for Precision Agriculture: A Case Study in Citrus Crop Water Requirement Estimation. Comput. Electron. Agric.
**2017**, 138, 175–187. [Google Scholar] [CrossRef] - Kolassa, J.; Reichle, R.H.; Liu, Q.; Alemohammad, S.H.; Gentine, P.; Aida, K.; Asanuma, J.; Bircher, S.; Caldwell, T.; Colliander, A.; et al. Estimating Surface Soil Moisture from SMAP Observations Using a Neural Network Technique. Remote Sens. Environ.
**2018**, 204, 43–59. [Google Scholar] [CrossRef] - Adeyemi, O.; Grove, I.; Peets, S.; Domun, Y.; Norton, T. Dynamic Neural Network Modelling of Soil Moisture Content for Predictive Irrigation Scheduling. Sensors
**2018**, 18, 3408. [Google Scholar] [CrossRef] [Green Version] - Liang, Y.; Ren, C.; Wang, H.; Huang, Y.; Zheng, Z. Research on Soil Moisture Inversion Method Based on GA-BP Neural Network Model. Int. J. Remote Sens.
**2019**, 40, 2087–2103. [Google Scholar] [CrossRef] - Martínez-Gimeno, M.A.; Jiménez-Bello, M.A.; Lidón, A.; Manzano, J.; Badal, E.; Pérez-Pérez, J.G.; Bonet, L.; Intrigliolo, D.S.; Esteban, A. Mandarin Irrigation Scheduling by Means of Frequency Domain Reflectometry Soil Moisture Monitoring. Agric. Water Manag.
**2020**, 235, 106151. [Google Scholar] [CrossRef] - Ahmed, N.; De, D.; Hussain, I. Internet of Things (IoT) for Smart Precision Agriculture and Farming in Rural Areas. IEEE Internet Things J.
**2018**, 5, 4890–4899. [Google Scholar] [CrossRef] - Popli, S.; Jha, R.K.; Jain, S. A Survey on Energy Efficient Narrowband Internet of Things (NBIoT): Architecture, Application and Challenges. IEEE Access
**2019**, 7, 16739–16776. [Google Scholar] [CrossRef] - Watteyne, T.; Doherty, L.; Simon, J.; Pister, K. Technical Overview of SmartMesh IP. In Proceedings of the 2013 Seventh International Conference on Innovative Mobile and Internet Services in Ubiquitous Computing, Taichung, Taiwan, 3–5 July 2013; pp. 547–551. [Google Scholar]
- Hindle, A.; Herraiz, I.; Shihab, E.; Jiang, Z.M. Mining Challenge 2010: FreeBSD, GNOME Desktop and Debian/Ubuntu. In Proceedings of the 2010 7th IEEE Working Conference on Mining Software Repositories (MSR 2010), Cape Town, South Africa, 2–3 May 2010; pp. 82–85. [Google Scholar]
- Zhuo, L.; Dai, Q.; Zhao, B.; Han, D. Soil Moisture Sensor Network Design for Hydrological Applications. Hydrol. Earth Syst. Sci.
**2020**, 24, 2577–2591. [Google Scholar] [CrossRef] - Dursun, M.; Özden, S. Optimization of Soil Moisture Sensor Placement for a PV-Powered Drip Irrigation System Using a Genetic Algorithm and Artificial Neural Network. Electr. Eng.
**2017**, 99, 407–419. [Google Scholar] [CrossRef] - Wang, P.; Wang, Y.; Wu, Q.S. Effects of Soil Tillage and Planting Grass on Arbuscular Mycorrhizal Fungal Propagules and Soil Properties in Citrus Orchards in Southeast China. Soil Tillage Res.
**2016**, 155, 54–61. [Google Scholar] [CrossRef] - Majhi, B.; Naidu, D.; Mishra, A.P.; Satapathy, S.C. Improved Prediction of Daily Pan Evaporation Using Deep-LSTM Model. Neural Comput. Appl.
**2019**. [Google Scholar] [CrossRef] - Xiao, C.; Ye, J.; Esteves, R.M.; Rong, C. Using Spearman’s Correlation Coefficients for Exploratory Data Analysis on Big Dataset. Concurr. Comput. Pract. Exp.
**2016**, 28, 3866–3878. [Google Scholar] [CrossRef] - Tufaner, F.; Demirci, Y. Prediction of Biogas Production Rate from Anaerobic Hybrid Reactor by Artificial Neural Network and Nonlinear Regressions Models. Clean Technol. Environ. Policy
**2020**, 22, 713–724. [Google Scholar] [CrossRef] - Jin, X.; Yang, N.; Wang, X.; Bai, Y.; Su, T.; Kong, J. Integrated Predictor Based on Decomposition Mechanism for PM2.5 Long-Term Prediction. Appl. Sci.
**2019**, 9, 4533. [Google Scholar] [CrossRef] [Green Version] - Yi, D.; Bu, S.; Kim, I. An Enhanced Algorithm of RNN Using Trend in Time-Series. Symmetry
**2019**, 11, 912. [Google Scholar] [CrossRef] [Green Version] - Madan, R.; Mangipudi, P.S. Predicting Computer Network Traffic: A Time Series Forecasting Approach Using DWT, ARIMA and RNN. In Proceedings of the 2018 Eleventh International Conference on Contemporary Computing (IC3), Noida, India, 2–4 August 2018; pp. 1–5. [Google Scholar]
- Canizo, M.; Triguero, I.; Conde, A.; Onieva, E. Multi-Head CNN–RNN for Multi-Time Series Anomaly Detection: An Industrial Case Study. Neurocomputing
**2019**, 363, 246–260. [Google Scholar] [CrossRef] - Sahoo, B.B.; Jha, R.; Singh, A.; Kumar, D. Long Short-Term Memory (LSTM) Recurrent Neural Network for Low-Flow Hydrological Time Series Forecasting. Acta Geophys.
**2019**, 67, 1471–1481. [Google Scholar] [CrossRef] - Sherstinsky, A. Fundamentals of Recurrent Neural Network (RNN) and Long Short-Term Memory (LSTM) Network. Phys. Nonlinear Phenom.
**2020**, 404, 132306. [Google Scholar] [CrossRef] [Green Version] - Yu, Y.; Si, X.; Hu, C.; Zhang, J. A Review of Recurrent Neural Networks: LSTM Cells and Network Architectures. Neural Comput.
**2019**, 31, 1235–1270. [Google Scholar] [CrossRef] - Liu, G.; Guo, J. Bidirectional LSTM with Attention Mechanism and Convolutional Layer for Text Classification. Neurocomputing
**2019**, 337, 325–338. [Google Scholar] [CrossRef] - Kiperwasser, E.; Goldberg, Y. Simple and Accurate Dependency Parsing Using Bidirectional LSTM Feature Representations. Trans. Assoc. Comput. Linguist.
**2016**, 4, 313–327. [Google Scholar] [CrossRef] - Yildirim, Ö. A Novel Wavelet Sequence Based on Deep Bidirectional LSTM Network Model for ECG Signal Classification. Comput. Biol. Med.
**2018**, 96, 189–202. [Google Scholar] [CrossRef] - Tao, Y.; Wang, X.; Zhang, Y. A Multitask Learning Neural Network for Short-Term Traffic Speed Prediction and Confidence Estimation. In Proceedings of the Artificial Neural Networks and Machine Learning—ICANN 2019: Deep Learning, Munich, Germany, 17–19 September 2019; Tetko, I.V., Kůrková, V., Karpov, P., Theis, F., Eds.; Springer International Publishing: Cham, Switzerland, 2019; pp. 434–449. [Google Scholar]
- Dai, H.; Ying, W.; Xu, J. Multi-Layer Neural Network for Received Signal Strength-Based Indoor Localisation. IET Commun.
**2016**, 10, 717–723. [Google Scholar] [CrossRef] - Hosamani, B.R.; Abbas Ali, S.; Katti, V. Assessment of Performance and Exhaust Emission Quality of Different Compression Ratio Engine Using Two Biodiesel Mixture: Artificial Neural Network Approach. Alex. Eng. J.
**2021**, 60, 837–844. [Google Scholar] [CrossRef] - Ćalasan, M.; Aleem, S.H.A.; Zobaa, A.F. On the Root Mean Square Error (RMSE) Calculation for Parameter Estimation of Photovoltaic Models: A Novel Exact Analytical Solution Based on Lambert W Function. Energy Convers. Manag.
**2020**, 210, 112716. [Google Scholar] [CrossRef] - Wang, W.; Lu, Y. Analysis of the Mean Absolute Error (MAE) and the Root Mean Square Error (RMSE) in Assessing Rounding Model. IOP Conf. Ser. Mater. Sci. Eng.
**2018**, 324, 012049. [Google Scholar] [CrossRef] - Eryılmaz, E.E.; Şahin, D.Ö.; Kılıç, E. Filtering Turkish Spam Using LSTM from Deep Learning Techniques. In Proceedings of the 2020 8th International Symposium on Digital Forensics and Security (ISDFS), Beirut, Lebanon, 1–2 June 2020; pp. 1–6. [Google Scholar]
- Zarei, A.; Asadi, E.; Ebrahimi, A.; Jafari, M.; Malekian, A.; Mohammadi Nasrabadi, H.; Chemura, A.; Maskell, G. Prediction of Future Grassland Vegetation Cover Fluctuation under Climate Change Scenarios. Ecol. Indic.
**2020**, 119, 106858. [Google Scholar] [CrossRef] - Zhou, L.; Zhao, P.; Wu, D.; Cheng, C.; Huang, H. Time Series Model for Forecasting the Number of New Admission Inpatients. BMC Med. Inform. Decis. Mak.
**2018**, 18, 39. [Google Scholar] [CrossRef] [Green Version] - Zhang, D. A Coefficient of Determination for Generalized Linear Models. Am. Stat.
**2017**, 71, 310–316. [Google Scholar] [CrossRef] - Serrano, J.; Shahidian, S.; Marques da Silva, J.; Paixão, L.; Calado, J.; Carvalho, M.D. Integration of Soil Electrical Conductivity and Indices Obtained through Satellite Imagery for Differential Management of Pasture Fertilization. AgriEngineering
**2019**, 1, 41. [Google Scholar] [CrossRef] [Green Version] - Ozlu, E.; Kumar, S. Response of Soil Organic Carbon, PH, Electrical Conductivity, and Water Stable Aggregates to Long-Term Annual Manure and Inorganic Fertilizer. Soil Sci. Soc. Am. J.
**2018**, 82, 1243–1251. [Google Scholar] [CrossRef] - Cavanaugh, J.E.; Neath, A.A. The Akaike Information Criterion: Background, Derivation, Properties, Application, Interpretation, and Refinements. WIREs Comput. Stat.
**2019**, 11, e1460. [Google Scholar] [CrossRef] - Velasco, J.A.; González-Salazar, C. Akaike Information Criterion Should Not Be a “Test” of Geographical Prediction Accuracy in Ecological Niche Modelling. Ecol. Inform.
**2019**, 51, 25–32. [Google Scholar] [CrossRef] - Cheng, H.; Xie, Z.; Wu, L.; Yu, Z.; Li, R. Data Prediction Model in Wireless Sensor Networks Based on Bidirectional LSTM. EURASIP J. Wirel. Commun. Netw.
**2019**, 2019, 203. [Google Scholar] [CrossRef] [Green Version] - Fang, K.; Pan, M.; Shen, C. The Value of SMAP for Long-Term Soil Moisture Estimation with the Help of Deep Learning. IEEE Trans. Geosci. Remote Sens.
**2019**, 57, 2221–2233. [Google Scholar] [CrossRef] - Hateffard, F.; Dolati, P.; Heidari, A.; Zolfaghari, A.A. Assessing the Performance of Decision Tree and Neural Network Models in Mapping Soil Properties. J. Mt. Sci.
**2019**, 16, 1833–1847. [Google Scholar] [CrossRef]

**Figure 1.**Study area. (

**a**) Geographic location of Huizhou within China; (

**b**) location of citrus orchard area in Huizhou; (

**c**) top view of the study area taken from Google Maps.

**Figure 10.**MAE performance comparison results between LSTM and MLNN for the prediction of SM and SEC.

**Figure 11.**RMSE performance comparison results between LSTM and MLNN for the prediction of SM and SEC.

**Figure 12.**R

^{2}performance comparison results between LSTM and MLNN for the prediction of SM and SEC.

**Figure 13.**Comparison of the Bid-LSTM model in different datasets for SM. (

**a**–

**e**) represent node 1–5, respectively; in (

**a1**–

**e1**), the blue line represents the measured SM, and the orange line represents the predicted SM; (

**a2**–

**e2**) represents the regression fitting between the predicted SM and measured SM.

**Figure 14.**Comparison of the Bid-LSTM model in different datasets for SEC. (

**a**–

**e**) represent node 1–5, respectively; in (

**a1**–

**e1**), the blue line represents the measured SEC, and the orange line represents the predicted SEC; (

**a2**–

**e2**) represents the regression fitting between the predicted SEC and measured SEC.

**Figure 15.**Comparison of the MLNN model in different datasets for SM. (

**a**–

**e**) represent node 1–5, respectively; in (

**a1**–

**e1**), the blue line represents the measured SEC, and the orange line represents the predicted SEC; (

**a2**–

**e2**) represents the regression fitting between the predicted SEC and measured SEC.

**Figure 16.**Comparison of the MLNN model in different datasets for SEC. (

**a**–

**e**) represents node 1–5, respectively; in (

**a1**–

**e1**), the blue line represents the measured SEC, and the orange line represents the predicted SEC; (

**a2**–

**e2**) represents the regression fitting between the predicted SEC and measured SEC.

**Table 1.**Correlation coefficient and p-value results between the environmental information and SM and SEC.

Environmental Factors | SM | SEC | ||
---|---|---|---|---|

Correlation Coefficient | p-Value | Correlation Coefficient | p-Value | |

Max temperature | 0.40 | $2.19\times {10}^{-26}$ | 0.35 | $2.91\times {10}^{-27}$ |

Min temperature | 0.50 | $1.30\times {10}^{-26}$ | 0.43 | $5.03\times {10}^{-28}$ |

Mean temperature | 0.45 | $1.05\times {10}^{-24}$ | 0.41 | $5.09\times {10}^{-26}$ |

Precipitation | 0.30 | $5.40\times {10}^{-18}$ | 0.13 | $4.90\times {10}^{-14}$ |

Air humidity | 0.53 | $1.88\times {10}^{-26}$ | 0.27 | $4.22\times {10}^{-24}$ |

Soil temperature | 0.55 | $3.34\times {10}^{-27}$ | 0.49 | $1.01\times {10}^{-19}$ |

Node | Models | AIC | |
---|---|---|---|

SM | SEC | ||

1 | MLNN | 448 | 379 |

Bid-LSTM | 335 | 76.33 | |

2 | MLNN | 611 | 525 |

Bid-LSTM | 440 | 471 | |

3 | MLNN | 660 | 506 |

Bid-LSTM | 530 | 405 | |

4 | MLNN | 657 | 528 |

Bid-LSTM | 369 | 612 | |

5 | MLNN | 117 | 65 |

Bid-LSTM | 126 | 64 |

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |

© 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Gao, P.; Xie, J.; Yang, M.; Zhou, P.; Chen, W.; Liang, G.; Chen, Y.; Han, X.; Wang, W.
Improved Soil Moisture and Electrical Conductivity Prediction of Citrus Orchards Based on IoT Using Deep Bidirectional LSTM. *Agriculture* **2021**, *11*, 635.
https://doi.org/10.3390/agriculture11070635

**AMA Style**

Gao P, Xie J, Yang M, Zhou P, Chen W, Liang G, Chen Y, Han X, Wang W.
Improved Soil Moisture and Electrical Conductivity Prediction of Citrus Orchards Based on IoT Using Deep Bidirectional LSTM. *Agriculture*. 2021; 11(7):635.
https://doi.org/10.3390/agriculture11070635

**Chicago/Turabian Style**

Gao, Peng, Jiaxing Xie, Mingxin Yang, Ping Zhou, Wenbin Chen, Gaotian Liang, Yufeng Chen, Xiongzhe Han, and Weixing Wang.
2021. "Improved Soil Moisture and Electrical Conductivity Prediction of Citrus Orchards Based on IoT Using Deep Bidirectional LSTM" *Agriculture* 11, no. 7: 635.
https://doi.org/10.3390/agriculture11070635