Publications

Sun, LY; Schulz, K (2017). Spatio-Temporal LAI Modelling by Integrating Climate and MODIS LAI Data in a Mesoscale Catchment. REMOTE SENSING, 9(2), 144.

Abstract
Vegetation is often represented by the leaf area index (LAI) in many ecological, hydrological and meteorological land surface models. However, the spatio-temporal dynamics of the vegetation are important to represent in these models. While the widely applied methods, such as the Canopy Structure Dynamic Model (CSDM) and the Double Logistic Model (DLM) are solely based on cumulative daily mean temperature data as input, a new spatio-temporal LAI prediction model referred to as the Temperature Precipitation Vegetation Model (TPVM) is developed that also considers cumulative precipitation data as input into the modelling process. TPVM as well as CDSM and DLM model performances are compared and evaluated against filtered LAI data from the Moderate Resolution Imaging Spectroradiometer (MODIS). The calibration/validation results of a cross-validation performed in the meso-scale Attert catchment in Luxembourg indicated that the DLM and TPVM generally provided more realistic and accurate LAI data. The TPVM performed superiorly for the agricultural land cover types compared to the other two models, which only used the temperature data. The Pearson's correlation coefficient (CC) between TPVM and the field measurement is 0.78, compared to 0.73 and 0.69 for the DLM and CSDM, respectively. The phenological metrics were derived from the TPVM model to investigate the interaction between the climate variables and the LAI variations. These interactions illustrated the dominant control of temperature on the LAI dynamics for deciduous forest cover, and a combined influence of temperature with precipitation for the agricultural land use areas.

DOI:
10.3390/rs9020144

ISSN:
2072-4292