Comparing evapotranspiration products of different temporal and spatial scales in native and managed prairie pastures

Authors: BAJGAIN, RAJEN - Orise Fellow; XIANGMING, XIAO - University Of Oklahoma; Wagle, Pradeep; KIMBALL, JOHN - University Of Montana; BRUST, COLLIN - University Of Montana; BASARA, JEFFREY - University Of Oklahoma; Gowda, Prasanna; Starks, Patrick; Neel, James

Submitted to: Remote Sensing
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 12/23/2020
Publication Date: 12/29/2020
Citation: Bajgain, R., Xiangming, X., Wagle, P., Kimball, J., Brust, C., Basara, J., Gowda, P.H., Starks, P.J., Neel, J.P. 2020. Comparing evapotranspiration products of different temporal and spatial scales in native and managed prairie pastures. Remote Sensing. 82(13). https://doi.org/10.3390/rs13010082.
DOI: https://doi.org/10.3390/rs13010082

Interpretive Summary: Historic native tallgrass prairies are managed differently for enhancing productivity. Different management activities alter evapotranspiration (ET, water use) of grasslands. Little is known about the impacts of different land management activities on ET at different spatio-temporal scales under differently managed prairie pastures. This study quantified and compared ET between co- located introduced managed pasture (MP) and native prairie (NP) pasture during the 2015 and 2016 growing seasons. Additionally, the study compared the Moderate Resolution Imaging Spectroradiometer (MODIS)-derived ET at four different spatial scales: 30 m (ETMOD30), 200 m (ETMOD200), 500 m (ETMOD500), and 1000 m (ETMOD1000) with eddy covariance-measured ET (ETEC). Large differences in ETEC were observed between the two pastures on half-hourly to seasonal scales, with variations mainly controlled by the amount of rainfall and management activities. In the pluvial year of 2015, growing season total ETEC was 30% less in MP than in NP, demonstrating differential responses of NP and MP in a pluvial year. In comparison, growing season total ETEC was similar in both pastures despite intra-season differences caused by grazing and hay harvest in the relatively drier year of 2016. The ETMOD30 showed a better agreement with ETEC than did the ETMOD200, ETMOD500, and ETMOD1000, most likely due to inability of ET products at coarser spatial scales to capture spatial heterogeneity of vegetation growth impacted by various management activities. Thus, it is crucial to integrate flux measurements and high-resolution remote sensing data depending on the heterogeneity of landscape surrounding the flux tower.

Technical Abstract: Grasslands in the Southern Great Plains of the continental United States have major ecological and economic importance, with strong climate and water cycle connections in the semi-arid region. The historic native prairie grassland are managed differently for enhancing productivity, while consequently altering water vapor fluxes between the grassland ecosystem and the atmosphere. However, little is known about the impacts of different land management activities on evapotranspiration (ET) at different spatio-temporal scales under differently managed prairie pastures. In this study, we quantified and compared ET between co-located introduced managed pasture (MP) and native prairie (NP) pasture during the 2015 and 2016 growing seasons. Additionally, we compared the Moderate Resolution Imaging Spectroradiometer (MODIS)-derived ET at four different spatial scales: 30 m (ETMOD30), 200 m (ETMOD200), 500 m (ETMOD500), and 1000 m (ETMOD1000) with eddy covariance- measured ET (ETEC). Large differences in ETEC were observed between the two pastures on half-hourly to seasonal scales, with variations mainly controlled by the amount of rainfall and management activities. In the pluvial year of 2015, growing season total ETEC was 30% less in MP than in NP. However, growing season total ETEC was similar in both pastures despite intra-season differences caused by grazing and hay harvest (July) in MP in a relatively drier year of 2016. The results demonstrated differential responses of MP and NP in a pluvial year. The ETMOD30 showed a better agreement with ETEC than did the ETMOD200, ETMOD500, and ETMOD1000. The ETMOD200, ETMOD500, and ETMOD1000 largely underestimated ETEC, most likely due to their inability to capture the spatial heterogeneity of vegetation growth impacted by various management activities. Our results facilitate understanding of the difference in ET of MP and NP due to differences in vegetation resulting from different management activities and their differential responses to precipitation.