Effects of in-situ and reanalysis climate data on estimation of cropland gross primary production using the Vegetation Photosynthesis Model

Cui Jin; Xiangming Xiao; Pradeep Wagle; Timothy Griffis; Jinwei Dong; Chaoyang Wu; Yuanwei Qin; David R. Cook

doi:10.1016/j.agrformet.2015.07.003

Effects of in-situ and reanalysis climate data on estimation of cropland gross primary production using the Vegetation Photosynthesis Model

Cui Jin, Xiangming Xiao, Pradeep Wagle, Timothy Griffis, Jinwei Dong, Chaoyang Wu, Yuanwei Qin, David R. Cook

Soil, Water, and Climate

Research output: Contribution to journal › Article › peer-review

32 Scopus citations

Abstract

Satellite-based Production Efficiency Models (PEMs) often require meteorological reanalysis data such as the North America Regional Reanalysis (NARR) by the National Centers for Environmental Prediction (NCEP) as model inputs to simulate Gross Primary Production (GPP) at regional and global scales. This study first evaluated the accuracies of air temperature (T_NARR) and downward shortwave radiation (R_NARR) of the NARR by comparing with in-situ meteorological measurements at 37 AmeriFlux non-crop eddy flux sites, then used one PEM - the Vegetation Photosynthesis Model (VPM) to simulate 8-day mean GPP (GPP_VPM) at seven AmeriFlux crop sites, and investigated the uncertainties in GPP_VPM from climate inputs as compared with eddy covariance-based GPP (GPP_EC). Results showed that T_NARR agreed well with in-situ measurements; R_NARR, however, was positively biased. An empirical linear correction was applied to R_NARR, and significantly reduced the relative error of R_NARR by ~25% for crop site-years. Overall, GPP_VPM calculated from the in-situ (GPP_VPM(EC)), original (GPP_VPM(NARR)) and adjusted NARR (GPP_VPM(adjNARR)) climate data tracked the seasonality of GPP_EC well, albeit with different degrees of biases. GPP_VPM(EC) showed a good match with GPP_EC for maize (Zea mays L.), but was slightly underestimated for soybean (Glycine max L.). Replacing the in-situ climate data with the NARR resulted in a significant overestimation of GPP_VPM(NARR) (18.4/29.6% for irrigated/rainfed maize and 12.7/12.5% for irrigated/rainfed soybean)_. GPP_VPM(adjNARR) showed a good agreement with GPP_VPM(EC) for both crops due to the reduction in the bias of R_NARR. The results imply that the bias of R_NARR introduced significant uncertainties into the PEM-based GPP estimates, suggesting that more accurate surface radiation datasets are needed to estimate primary production of terrestrial ecosystems at regional and global scales.

Original language	English (US)
Pages (from-to)	240-250
Number of pages	11
Journal	Agricultural and Forest Meteorology
Volume	213
DOIs	https://doi.org/10.1016/j.agrformet.2015.07.003
State	Published - Nov 1 2015

Bibliographical note

Funding Information:
This study was supported by the National Institute of Food and Agriculture, U.S. Department of Agriculture , under award number 2013-69002-23146 , and a research grant from the National Science Foundation EPSCoR program (Project No. IIA-1301789 ). We would like to thank Drs. Andrew E. Suyker, John Baker, Tilden Meyers, Roser Matamala for providing eddy flux data.

Publisher Copyright:
© 2015 Elsevier B.V.

Keywords

AmeriFlux
Downward shortwave radiation
MODIS
NARR
Vegetation Photosynthesis Model (VPM)
Vegetation indices

Publisher link

10.1016/j.agrformet.2015.07.003

Find It

Find It at U of M Libraries

Cite this

@article{f8fa2c789b844ba79795d8e46e691124,

title = "Effects of in-situ and reanalysis climate data on estimation of cropland gross primary production using the Vegetation Photosynthesis Model",

abstract = "Satellite-based Production Efficiency Models (PEMs) often require meteorological reanalysis data such as the North America Regional Reanalysis (NARR) by the National Centers for Environmental Prediction (NCEP) as model inputs to simulate Gross Primary Production (GPP) at regional and global scales. This study first evaluated the accuracies of air temperature (TNARR) and downward shortwave radiation (RNARR) of the NARR by comparing with in-situ meteorological measurements at 37 AmeriFlux non-crop eddy flux sites, then used one PEM - the Vegetation Photosynthesis Model (VPM) to simulate 8-day mean GPP (GPPVPM) at seven AmeriFlux crop sites, and investigated the uncertainties in GPPVPM from climate inputs as compared with eddy covariance-based GPP (GPPEC). Results showed that TNARR agreed well with in-situ measurements; RNARR, however, was positively biased. An empirical linear correction was applied to RNARR, and significantly reduced the relative error of RNARR by ~25% for crop site-years. Overall, GPPVPM calculated from the in-situ (GPPVPM(EC)), original (GPPVPM(NARR)) and adjusted NARR (GPPVPM(adjNARR)) climate data tracked the seasonality of GPPEC well, albeit with different degrees of biases. GPPVPM(EC) showed a good match with GPPEC for maize (Zea mays L.), but was slightly underestimated for soybean (Glycine max L.). Replacing the in-situ climate data with the NARR resulted in a significant overestimation of GPPVPM(NARR) (18.4/29.6% for irrigated/rainfed maize and 12.7/12.5% for irrigated/rainfed soybean). GPPVPM(adjNARR) showed a good agreement with GPPVPM(EC) for both crops due to the reduction in the bias of RNARR. The results imply that the bias of RNARR introduced significant uncertainties into the PEM-based GPP estimates, suggesting that more accurate surface radiation datasets are needed to estimate primary production of terrestrial ecosystems at regional and global scales.",

keywords = "AmeriFlux, Downward shortwave radiation, MODIS, NARR, Vegetation Photosynthesis Model (VPM), Vegetation indices",

author = "Cui Jin and Xiangming Xiao and Pradeep Wagle and Timothy Griffis and Jinwei Dong and Chaoyang Wu and Yuanwei Qin and Cook, {David R.}",

note = "Funding Information: This study was supported by the National Institute of Food and Agriculture, U.S. Department of Agriculture , under award number 2013-69002-23146 , and a research grant from the National Science Foundation EPSCoR program (Project No. IIA-1301789 ). We would like to thank Drs. Andrew E. Suyker, John Baker, Tilden Meyers, Roser Matamala for providing eddy flux data. Publisher Copyright: {\textcopyright} 2015 Elsevier B.V.",

year = "2015",

month = nov,

day = "1",

doi = "10.1016/j.agrformet.2015.07.003",

language = "English (US)",

volume = "213",

pages = "240--250",

journal = "Agricultural and Forest Meteorology",

issn = "0168-1923",

publisher = "Elsevier",

}

TY - JOUR

T1 - Effects of in-situ and reanalysis climate data on estimation of cropland gross primary production using the Vegetation Photosynthesis Model

AU - Jin, Cui

AU - Xiao, Xiangming

AU - Wagle, Pradeep

AU - Griffis, Timothy

AU - Dong, Jinwei

AU - Wu, Chaoyang

AU - Qin, Yuanwei

AU - Cook, David R.

N1 - Funding Information: This study was supported by the National Institute of Food and Agriculture, U.S. Department of Agriculture , under award number 2013-69002-23146 , and a research grant from the National Science Foundation EPSCoR program (Project No. IIA-1301789 ). We would like to thank Drs. Andrew E. Suyker, John Baker, Tilden Meyers, Roser Matamala for providing eddy flux data. Publisher Copyright: © 2015 Elsevier B.V.

PY - 2015/11/1

Y1 - 2015/11/1

N2 - Satellite-based Production Efficiency Models (PEMs) often require meteorological reanalysis data such as the North America Regional Reanalysis (NARR) by the National Centers for Environmental Prediction (NCEP) as model inputs to simulate Gross Primary Production (GPP) at regional and global scales. This study first evaluated the accuracies of air temperature (TNARR) and downward shortwave radiation (RNARR) of the NARR by comparing with in-situ meteorological measurements at 37 AmeriFlux non-crop eddy flux sites, then used one PEM - the Vegetation Photosynthesis Model (VPM) to simulate 8-day mean GPP (GPPVPM) at seven AmeriFlux crop sites, and investigated the uncertainties in GPPVPM from climate inputs as compared with eddy covariance-based GPP (GPPEC). Results showed that TNARR agreed well with in-situ measurements; RNARR, however, was positively biased. An empirical linear correction was applied to RNARR, and significantly reduced the relative error of RNARR by ~25% for crop site-years. Overall, GPPVPM calculated from the in-situ (GPPVPM(EC)), original (GPPVPM(NARR)) and adjusted NARR (GPPVPM(adjNARR)) climate data tracked the seasonality of GPPEC well, albeit with different degrees of biases. GPPVPM(EC) showed a good match with GPPEC for maize (Zea mays L.), but was slightly underestimated for soybean (Glycine max L.). Replacing the in-situ climate data with the NARR resulted in a significant overestimation of GPPVPM(NARR) (18.4/29.6% for irrigated/rainfed maize and 12.7/12.5% for irrigated/rainfed soybean). GPPVPM(adjNARR) showed a good agreement with GPPVPM(EC) for both crops due to the reduction in the bias of RNARR. The results imply that the bias of RNARR introduced significant uncertainties into the PEM-based GPP estimates, suggesting that more accurate surface radiation datasets are needed to estimate primary production of terrestrial ecosystems at regional and global scales.

AB - Satellite-based Production Efficiency Models (PEMs) often require meteorological reanalysis data such as the North America Regional Reanalysis (NARR) by the National Centers for Environmental Prediction (NCEP) as model inputs to simulate Gross Primary Production (GPP) at regional and global scales. This study first evaluated the accuracies of air temperature (TNARR) and downward shortwave radiation (RNARR) of the NARR by comparing with in-situ meteorological measurements at 37 AmeriFlux non-crop eddy flux sites, then used one PEM - the Vegetation Photosynthesis Model (VPM) to simulate 8-day mean GPP (GPPVPM) at seven AmeriFlux crop sites, and investigated the uncertainties in GPPVPM from climate inputs as compared with eddy covariance-based GPP (GPPEC). Results showed that TNARR agreed well with in-situ measurements; RNARR, however, was positively biased. An empirical linear correction was applied to RNARR, and significantly reduced the relative error of RNARR by ~25% for crop site-years. Overall, GPPVPM calculated from the in-situ (GPPVPM(EC)), original (GPPVPM(NARR)) and adjusted NARR (GPPVPM(adjNARR)) climate data tracked the seasonality of GPPEC well, albeit with different degrees of biases. GPPVPM(EC) showed a good match with GPPEC for maize (Zea mays L.), but was slightly underestimated for soybean (Glycine max L.). Replacing the in-situ climate data with the NARR resulted in a significant overestimation of GPPVPM(NARR) (18.4/29.6% for irrigated/rainfed maize and 12.7/12.5% for irrigated/rainfed soybean). GPPVPM(adjNARR) showed a good agreement with GPPVPM(EC) for both crops due to the reduction in the bias of RNARR. The results imply that the bias of RNARR introduced significant uncertainties into the PEM-based GPP estimates, suggesting that more accurate surface radiation datasets are needed to estimate primary production of terrestrial ecosystems at regional and global scales.

KW - AmeriFlux

KW - Downward shortwave radiation

KW - MODIS

KW - NARR

KW - Vegetation Photosynthesis Model (VPM)

KW - Vegetation indices

UR - http://www.scopus.com/inward/record.url?scp=84938800554&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84938800554&partnerID=8YFLogxK

U2 - 10.1016/j.agrformet.2015.07.003

DO - 10.1016/j.agrformet.2015.07.003

M3 - Article

AN - SCOPUS:84938800554

SN - 0168-1923

VL - 213

SP - 240

EP - 250

JO - Agricultural and Forest Meteorology

JF - Agricultural and Forest Meteorology

ER -

Effects of in-situ and reanalysis climate data on estimation of cropland gross primary production using the Vegetation Photosynthesis Model

Abstract

Bibliographical note

Keywords

Publisher link

Find It

Other files and links

Fingerprint

Cite this