Abstract
This study aimed to investigate the potential of spatially distributed data products in estimating soil respiration (Rs), including land surface temperature (LST) and spectral vegetation index from the Moderate Resolution Imaging Spectroradiometer (MODIS) and root zone soil moisture derived from the assimilation of the NASA Advanced Microwave Scanning Radiometer-EOS and a land surface model, at a deciduous broadleaf forest site in the Midwest USA. Several statistical models were used to examine the dependencies of Rs on these spatial data products, and accuracy of these models was compared to the models based on in situ measurements. The models based on mean LST (i.e., averaging nighttime and daytime LST from MODIS) and root zone soil moisture explained 82% and 72% of seasonal variations in Rs for spring and winter dormant periods, respectively. In the growing season, the models depending on mean LST, root zone soil moisture, and photosynthesis-related enhanced vegetation index showed comparable accuracy with the models entirely based on in situ measured data, except for the midgrowing period. Drought stress led to a relatively low explanation capacity for the Rs model based on spatial data products during the midgrowing period. However, this model still explained 76% of temporal dynamics of Rs over the midgrowing period. Our results suggested that simple models based entirely on spatial data products have the potential to estimate Rs at the temperate deciduous forest site. The conclusions drawn from the present study provided valuable information for large-scale estimates of Rs in temperate deciduous forest ecosystems.
| Original language | English |
|---|---|
| Pages (from-to) | 6393-6408 |
| Number of pages | 16 |
| Journal | Journal of Geophysical Research: Biogeosciences |
| Volume | 119 |
| Issue number | 11 |
| DOIs | |
| State | Published - Jun 16 2014 |
Funding
We sincerely thank the anonymous reviewers for their important and constructive revision advices on the manuscript. We like to offer thanks to the America Fluxnet (http://public.ornl.gov/ ameriflux/) which provides the data for our analysis. We also thank B. Yang from Environmental Sciences Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee, USA, for useful suggestions about the use of soil CO2 flux data, and J.D. Bolten from Hydrological Sciences Branch, NASA Goddard Space Flight Center, Beltsville, MD, for helpful explanation about the root zone soil moisture product. This work was supported by the National Natural Science Foundation of China (41301498), the Special Foundation for Young Scientists of the State Laboratory of Remote Sensing Science (13RC-07), and the Major State Basic Research Development Program of China (2013CB733405).