Mapping High-Resolution Incident Photosynthetically Active Radiation product for the North American Carbon Program

Dr. Shunlin Liang, University of Maryland

Dr. Steven Running, University of Montana

Dr. John Townshend, University of Maryland

Dr. Si-Chee Tsay, NASA/Goddard Space Flight Center

Abstract

Incident photosynthetically active radiation (PAR) is one of the key variables driving most terrestrial ecosystem models. Unfortunately, the current incident PAR products estimated from remotely sensed data at spatial and temporal resolutions are not sufficient for carbon cycle modeling. For example, the MODIS (Moderate-Resolution Imaging Spectroradiometer) science team has to use the NASA DAO (data assimilation office) shortwave solar radiation product of 1° by 1°spatial resolution to produce 1km GPP (gross primary productivity)/NPP (net primary productivity) products. The MODIS science team is producing the incident PAR over the ocean operationally, but no incident PAR over land is produced. Supported by NASA Terrestrial Ecology Program, we have developed a robust algorithm for estimating incident PAR from MODIS. In this study, we propose to apply this developed algorithm to generate the PAR product from MODIS, AVHRR (Advanced Very High-Resolution Radiometer) and GOES (Geostationary Operational Environmental Satellite) over North America in support of the North American Carbon Program (NACP).

Both instantaneous and daily average PAR products will be generated. The instantaneous PAR from the MODIS of both Terra and Aqua will be produced from 2000 at 1km resolution. The historical instantaneous incident PAR will be generated from the AVHRR beginning in 1981 at 8km resolution. The daily average PAR is calculated from the diurnal cycle of GOES data. Our strategy is to use the GOES to determine the relative shape of the PAR diurnal cycle and to anchor the poorly calibrated GOES PAR product using MODIS/AVHRR PAR as absolute reference.

The products will be extensively validated using ground observations from several networks. The products will be tested using the MODIS GPP/Biome-BGC model. The PAR estimation code will be also piped into the MODIS GPP/NPP (MOD17) production code for improving GPP/NPP calculations over North America and eventually for global applications. The product will be generated and distributed to other NACP investigators through the University of Maryland Global Land Cover Facility (GLCF).

The proposed effort builds on our existing NASA project on PAR algorithm development, and is a cost-effective strategy for applying the developed algorithms to generation of the urgently needed products for supporting the NACP. This proposal is in direct response of the NASA NRA and will greatly help reduce the uncertainty in calculating the carbon cycle over North America.