Rainfall information is a prerequisite to and plays a vital role in driving hydrological models. However, limited by the observation methods, the obtained precipitation data, at present, are still too coarse. In this study, a new downscaling method was proposed to obtain high spatial resolution (~1 km/hourly) precipitation estimates based on Integrated Multi-satellitE Retrievals for GPM (IMERG) data at hourly scale. Compared with original IMERG data, the downscaled precipitation results showed the similar spatial patterns with those of original IMERG data, but with finer spatial resolution. In addition, the downscaled precipitation estimates were further analyzed to quantify their improvements using the Coupled Routing and Excess STorage (CREST) model across Ganjiang River basin. Compared with the observed streamflow, the downscaled precipitation results showed satisfying hydrological performance, with Nash-Sutcliffe Coefficient of Efficiency (NSCE), Root Mean Square Error (RMSE), Relative Bias (BIAS), and Correlation Coefficient (CC). The improvement in terms of four statistic metrics in terms of streamflow simulation also indicated great potential of hydrological utility for the downscaled precipitation results.

Accurate estimation of evapotranspiration (ET) and its components is critical to developing a better understanding of climate, hydrology, and vegetation coverage conditions for areas of interest. A hybrid dual-source (H-D) model incorporating the strengths of the two-layer and two-patch schemes was proposed to estimate actual ET processes by considering varying vegetation coverage patterns and soil moisture conditions. The proposed model was tested in four different ecosystems, including deciduous broadleaf forest, woody savannas, grassland, and cropland. Performance of the H-D model was compared with that of the Penman-Monteith (P-M) model, the Shuttleworth-Wallace (S-W) model, as well as the Two-Patch (T-P) model, with ET and/or its components (i.e., transpiration and evaporation) being evaluated against eddy covariance measurements. Overall, ET estimates from the developed H-D model agreed reasonably well with the ground-based measurements at all sites, with mean absolute errors ranging from 16.3 W/m² to 38.6 W/m², indicating good performance of the H-D model in all ecosystems being tested. In addition, the H-D model provides a more reasonable partitioning of evaporation and transpiration than other models in the ecosystems tested.

Object detection in unmanned aerial vehicle (UAV) images is an extremely challenging task and involves problems such as multi-scale objects, a high proportion of small objects, and high overlap between objects. To address these issues, first, we design a Vectorized Intersection Over Union (VIOU) loss based on YOLOv5s. This loss uses the width and height of the bounding box as a vector to construct a cosine function that corresponds to the size of the box and the aspect ratio and directly compares the center point value of the box to improve the accuracy of the bounding box regression. Second, we propose a Progressive Feature Fusion Network (PFFN) that addresses the issue of insufficient semantic extraction of shallow features by Panet. This allows each node of the network to fuse semantic information from deep layers with features from the current layer, thus significantly improving the detection ability of small objects in multi-scale scenes. Finally, we propose an Asymmetric Decoupled (AD) head, which separates the classification network from the regression network and improves the classification and regression capabilities of the network. Our proposed method results in significant improvements on two benchmark datasets compared to YOLOv5s. On the VisDrone 2019 dataset, the performance increased by 9.7% from 34.9% to 44.6%, and on the DOTA dataset, the performance increased by 2.1%.

Aiming at the problem of small capacity and poor robustness of audio watermarking information, an adaptive digital audio blind watermarking algorithm based on wavelet transform is proposed. In this algorithm, each frame of the audio signal is converted into a discrete transform by using Haar wavelet, and the fine and approximate components are obtained, and the component masking threshold is calculated. Finally, the watermark embedding position is selected by the adaptive algorithm, quantized and modulated in the audio. The intermediate and low frequency coefficients of the signal. The simulation results show that the proposed algorithm has some resistance to common signal processing attacks and is robust to resampling and low-pass filtering attacks.