AUTHOR=Li Shangzhi , Zhang Meng 

TITLE=Improving the MODIS leaf area index product for a cropland with the nonlinear autoregressive neural network with eXogenous input model

JOURNAL=Frontiers in Earth Science

VOLUME=Volume 10 - 2022

YEAR=2023

URL=https://www.frontiersin.org/journals/earth-science/articles/10.3389/feart.2022.962498

DOI=10.3389/feart.2022.962498

ISSN=2296-6463

ABSTRACT=Leaf area index (LAI) is a crucial descriptive parameter of the dynamic change of ground vegetation. The latest version of MODIS LAI product (MCD15A2H), namely Collection 6 (C6) is becoming more and more indispensable to climate change and energy balance investigations. However, researches showed that pixels produced by the backup NDVI-LAI relationship algorithm were not as reliable as the 3-Dimensional Radioactive Transfer algorithm. Aiming to alleviate the backup algorithm's adverse effects, this paper implemented a Nonlinear Auto Regressive neural network with eXogenous input (NARXNN) model, which was constructed by spectral reflectance, view angles and historical LAI, to replace unreliable pixels in MCD15A2H product. Case studies were implemented at two seasons a year croplands in Wuzhi, Xinzheng, and Xiangcheng, Henan province. This research inversed 46 periods of NARXNN model improved LAI, which went through rigid in-situ LAI validation. In-situ measured LAI by LAI-2000 were used to validate the accuracy of NARXNN enhanced LAI data. Direct validation using in-situ measured LAI demonstrates that the NARXNN model enhanced LAI data was more accurate and had a lower bias than MCD15A2H. A comparison of the time series change indicates that the NARXNN enhanced LAI shows a smoother bimodal change trend, and closer to up-scaled validation data than original MODIS product. The results indicated that the NARXNN neural network further increased the accuracy of the MODIS product and has a particular practical value in future research.