Improving the Estimation of Seasonal Leaf Area Index of Coniferous Forests for Better Carbon and Water Flux Estimation

The availability of global products of leaf area index (LAI) using remote sensing (RS) data makes it possible to model land surface processes at regional and global scales. Due to the lack of continuous observations of in-situ LAI, studies on the seasonal variability of RS LAI products are rare in the literature. Moreover, the application of existing global LAI products is largely restricted by the underestimation of winter LAI over high-latitude evergreen coniferous forests. Low winter LAI values, due to the influence of background, the presence of ice and snow on leaves and the seasonal variation of leaf pigments, exaggerate the LAI seasonality, possibly causing errors in land surface modeling. The aim of this thesis is to gain a better understanding of the quality of existing RS LAI products, to improve these products, and to further investigate the influence of the improvements on modeling carbon, water and energy fluxes. Frequent measurements of LAI in all seasons were conducted at two white pine sites near Turkey Point, Ontario, Canada, using direct and indirect methods. The in-situ LAI measurements were then used to validate the University of Toronto RS LAI product Version 2. Results show that RS LAI captures mostly canopy structural information from the onset of needleleaf growth to the seasonal peak, but captures both leaf chlorophyll and structural information from the seasonal peak to the completion of needleleaf fall. This finding has significant implications for the application of current RS LAI products to terrestrial productivity modeling and prompts the need for separating leaf structure and chlorophyll information. A simple scheme for correcting the distorted LAI seasonality of evergreen conifers is proposed in my research, through which an improved LAI product over conifers in Canada has been developed. It is demonstrated that the improved product greatly increases the accuracy of the simulation of land surface carbon, water and energy fluxes in coniferous forests in Canada.Ph.D