There are an estimated 800,000 small reservoirs globally with a range of uses. Given the collective importance of these reservoirs to water resource management and wider society, it is essential that we can monitor and understand the hydrological dynamics of ungauged reservoirs, particularly in a changing climate. However, unlike large reservoirs, continuous and systematic hydrological observations of small reservoirs are often unavailable. In response, this study has developed a retrieval framework for water levels of small reservoirs using a deep learning algorithm and remotely sensed satellite data. Demonstrated at four reservoirs in California, satellite imagery from both Sentinel-1 and Sentinel-2 along with corresponding water level field measurements was collected. Post-processed images were fed into a water level inversion convolutional neural network model for water level inversion, while different combinations of these satellite images, sampling approaches for training/testing data, and attention modules were used to train the model and evaluated for accuracy. The results show that random sampling of training data coupled with Sentinel-2 satellite imagery was generally the most accurate initially. Performance is improved by incorporating a channel attention mechanism, with the average R² increasing by 8.6% and the average RMSE and MAE decreasing by 15.5% and 36.4%, respectively. The proposed framework was further validated on three additional reservoirs in different regions. In conclusion, the retrieval framework proposed in this study provides a stable and accurate methodology for water level estimation of small reservoirs and can be a powerful tool for small reservoir monitoring over large spatial scales.

Most rice growth stage predictions are currently based on a few rice varieties for prediction method studies, primarily using linear regression, machine learning, and other methods to build growth stage prediction models that tend to have poor generalization ability, low accuracy, and face various challenges. In this study, multispectral images of rice at various growth stages were captured using an unmanned aerial vehicle, and single-plant rice silhouettes were identified for 327 rice varieties by establishing a deep-learning algorithm. A growth stage prediction method was established for the 327 rice varieties based on the normalized vegetation index combined with cubic polynomial regression equations to simulate their growth changes, and it was first proposed that the growth stages of different rice varieties were inferred by analyzing the normalized difference vegetation index growth rate. Overall, the single-plant rice contour recognition model showed good contour recognition ability for different rice varieties, with most of the prediction accuracies in the range of 0.75–0.93. The accuracy of the rice growth stage prediction model in recognizing different rice varieties also showed some variation, with the root mean square error between 0.506 and 3.373 days, the relative root mean square error between 2.555% and 14.660%, the Bias between1.126 and 2.358 days, and the relative Bias between 0.787% and 9.397%; therefore, the growth stage prediction model of rice varieties can be used to effectively improve the prediction accuracy of the growth stage periods of rice.