A hierarchical downscaling scheme for generating fine-resolution leaf area index with multisource and multiscale observations via deep learning

• A hierarchical downscaling scheme is proposed for high-resolution LAI estimations. • Multisource satellite observations are helpful for LAI downscaling via deep learning. • ISRGAN effectively increases the spatial scales of coarse resolution LAI images. • Transfer learning obtains reliable LAI estimations at the 250-m resolution. • A two-stage downscaling method enhances the performance of 30-m resolution LAI maps. Leaf area index (LAI) is one of key variables for depicting vegetation structures in land ecosystems. Land surface models necessitate uniform LAI inputs at varying spatial scales to ensure accurate outputs at multiscale levels, however, operational satellite LAI products are acquired only at low spatial resolutions, inhibiting their application at finer spatial scales. Spatial downscaling methods are beneficial for the spatial enhancement of LAI products, and the emergence of deep learning methods has provided promising options for land surface parameter downscaling. However, the potential of deep learning has not been well explored in LAI downscaling. To address this research gap, this study designed an original hierarchical downscaling approach facilitated by generative adversarial network (GAN), transfer learning (TL), and data augmentation techniques to retrieve LAI at fine spatial resolutions, leveraging multiscale satellite images, and cascading from 500-m to 250-m and then to 30-m scales. First, an improved super-resolution GAN (ISRGAN) model was pre-trained using the GLASS LAI and MOD09Q1 products to bridge the general non-linear relationships of LAI between the 500-m and 250-m resolutions. Subsequently, limited reference LAI images were applied to fine-tune this pre-trained ISRGAN model to address the domain shift in the 250-m resolution LAI estimations. Then, the fine-tuned LAI values and the 30-m resolution LAI reference images were utilized as the ISRGAN inputs to produce fine-resolution LAI maps. Finally, the downscaled LAI values derived from the proposed approach were separately validated against reference LAI maps and field measurements across the 250-m and 30-m resolutions. Results show that the fine-tuned transfer learning technique outperforms the pre-trained ISRGAN model and GLASS LAI, with a lower RMSE (0.78) and higher R 2 (0.83) at the 250-m resolution. Moreover, the proposed hierarchical downscaling framework achieves better performances for 30-m resolution LAI estimations, regardless of the validation accuracy (R 2 = 0.76; RMSE=0.95) and spatiotemporal distributions, than the ISRGAN model which was directly trained by the 500-m and 30-m resolution images. This study highlights that a hierarchical downscaling is valuable for fine-resolution LAI estimations, which leverages multiscale and multisource satellite observations via deep learning.