Abstract: Soil moisture content in the root zone is one of the most crucial factors in determining the healthy growth and yield of kiwifruit trees, particularly in dynamic monitoring of the soil moisture during fruit expansion. However, traditional monitoring cannot capture the continuous variation in soil moisture. In this study, a time-series data inversion was carried out on the soil moisture content in the root zone of kiwifruit. The fruits were taken from the Meixian kiwifruit experimental station in Baoji City, Shaanxi Province, China. Both unmanned aerial vehicles (UAV) equipped with multispectral sensors and ground-based moisture sensors were utilized to collect the spectral reflectance and soil moisture data over a period of 60 days, respectively, resulting in a total of 1440 datasets. The spectral data was processed to initially extract 20 vegetation indices, such as the Normalized Difference Vegetation Index (NDVI), Soil-Adjusted Vegetation Index (SAVI), and Green Normalized Difference Vegetation Index (GNDVI). The dataset was then refined to identify the most critical features for the soil moisture inversion. Pearson and Spearman correlation coefficients were employed to determine the nine most relevant vegetation indices. The indices were also optimized to reduce the model complexity for the high predictive power. Subsequently, the optimal indices were used as the inputs for three machine learning models: the Feedforward Neural Network (FFNN), the Long Short-Term Memory (LSTM) network, and the Bidirectional Long Short-Term Memory (BiLSTM) network. Among them, the FFNN served as a baseline to compare with the temporal models, due to the temporal independencies in the data. The experimental setup involved training and testing the three models using the dataset. The FFNN model was trained with the input features representing the selected vegetation indices without any temporal information. In contrast, the LSTM and BiLSTM models were designed to utilize the multi-day historical data, thus capturing the temporal dependencies in the soil moisture dynamics. The LSTM model was particularly effective in retaining the information over longer sequences, due to its memory cell architecture. While the BiLSTM extended the capability to consider both forward and backward temporal dependencies, thus providing a more comprehensive understanding of temporal relationships. The results indicated that the LSTM and BiLSTM models with the multi-day historical data significantly improved the prediction accuracy. Specifically, the LSTM model was achieved in the test dataset coefficient of determination (R²) value of 0.548 and a root mean square error (RMSE) of 2.68%, indicating the more effective temporal dependencies, compared with the FFNN. The FFNN model exhibited a relatively low performance on the test dataset, with an R² value of 0.269 and an RMSE of 3.56%, due to its inability to capture time-series information. Among them, the BiLSTM performed the best, thus achieving a test dataset R² value of 0.624 and an RMSE of 2.45%. This superior performance of the BiLSTM was attributed to both past and future contexts within the sequence, which enhanced its predictive capability for soil moisture dynamics. The better performance of temporal models was achieved in the dynamic prediction of soil moisture, compared with the non-temporal models. The LSTM and BiLSTM were incorporated with the temporal data for more accurate modeling of the complex interactions between vegetation indices and soil moisture content. Particularly, the precise prediction of soil moisture during fruit expansion was crucial to optimize the yield and quality. In conclusion, the time-series LSTM and BiLSTM models were proved to effectively monitor the soil moisture in the period of the kiwifruit fruit expansion. This approach can also offer valuable theoretical support to the water practices in orchards. The promising potential can integrate the multispectral UAV remote sensing data with deep learning in smart agriculture. Advanced temporal modeling can be expected to enhance the monitoring precision of soil moisture in sustainable and efficient crop production.