Joseph G. Alfieri 1, William P. Kustas 1, John H. Prueger 2, Lynn G. McKee 1, Feng Gao 1 Lawrence E. Hipps 3, Sebastian Los 3 1 USDA, ARS, Hydrology & Remote Sensing Lab, Beltsville MD 2 USDA,ARS, National Lab for Agriculture & Environment, Ames IA 3 Utah State University, Department of Plants Soil and Climate, Logan UT May 15, 2018
Motivations Wine grapes are one of the largest specialty crops in California. There are 325,000 ha of wine grape vineyards in California. California wine grape production is valued at nearly $6B annually. The wine industry employs more than 300,000 people and contributes in excess of $60B to the state s economy each year. Like many parts of the world, fresh water is an increasingly scarce resource in California. It is critical that limited water resources are managed efficiently to meet the competing demands of urban, industrial, and agricultural consumers. Remote sensing is the most viable approach for monitoring evapotranspiration (ET) across spatial and temporal scales. This data is needed to manage irrigation and maximize the efficient use of water resources. The Grape Remote Sensing Atmospheric Profiling and Evapotranspiration Experiment (GRAPEX) is a multi-institutional field campaign ongoing in the Central Valley of California. The overarching goal of GRAPEX is to develop a multi-scale remote sensing-based ET toolkit to improve irrigation scheduling and water management.
Study Objectives 1.45 m 2.25 m The canopy structure of the GRAPEX sites is characterized by: Wide row spacing (~3.35 m). Tall canopy (~2.25 m). 3.35 m Vegetation concentrated in the upper half of the canopy. The structure of the vines differs significantly from the tightly-spaced closed canopy typical of cereal grains and other row crops. The aim of this study is to understand how the unique characteristics of the vineyards impact the turbulent fluxes from vineyards, specifically ET.
Site Description The data was collected at a pair of adjacent vineyards near Lodi in California s Central Valley from 2013 to 2017. Both vineyards are planted with the pinot noir varietal. The vines in the northern vineyard (Site 1) were planted in 2009 while those planted in the southern vineyard (Site 2) were planted in 2011. Both vineyards are drip irrigated and share the same management practices. Site 1 Site 2 The relevant measurements collected at each site include: Meteorological data, e.g. wind speed and air temperature. Surface energy fluxes. Wind speed profiles measured at 2.5, 3.75, 5, and 8 m, agl. Daily leaf area index (LAI) derived from satellite imagery.
Meteorology Not unexpectedly, the meteorological conditions did not vary between the two vineyards. Similarly, the annual patterns are also quite similar year-to-year. During the growing season, the wind speed averaged ~ 2.1 m s -1. Wind speeds were significantly greater (~0.3 m s-1) during 2013 and 2014 compared to subsequent years. No otherquantity showed statistically significant variability. The air temperature peaked near 26 C during the growing season. Similarly, humidity peaks at ~1.5 kpa during the growing season. Pressure is more variable, but there is a clear annual pattern with a minima during the growing season.
Surface Fluxes There was very little inter-annual or inter-site variability in net radiation (R n ). Particularly during the growing season, the soil heat flux (G) varied greatly both between sites and year-to-year. Although it was initially higher in Site 2, the greater G has been at Site 1 since 2015. The sensible heat flux (H) tended to be higher in the Site 2. Depending on the year, H differed by between 10 and 50 W m -2. The latent heat flux (λe) tended to be greater in Vineyard 1. With a maximum discrepancy of nearly 60 W m -2, the largest differences occurred during the 2013 growing season.
Surface Fluxes & LAI Much of the difference in the fluxes measurements in the two vineyards can be attributed to differences in the vegetation density. The coefficient of determination (r 2 ) range between 0.817 and 0.885 excluding H. The single clearly anomalous data point is in the relationship between H and LAI. The difference in the measurements of H during 2014 was much smaller than expected given the difference in LAI. Indirect evidence suggest that this anomaly may be due to differences in the irrigation regime in the two vineyards. Flux r 2 Net Radiation 0.839 Soil Heat Flux 0.817 Sensible Heat Flux 0.121 Latent Heat Flux 0.885 Flux MAE Net Radiation 2.2 Soil Heat Flux 4.1 Sensible Heat Flux 12.1 Latent Heat Flux 10.4
Surface Fluxes & the Impact of Soil Moisture Using data from near-surface (90 cm) soil moisture (θ) profiles in each vineyard, the mean difference in soil moisture during the growing season was determined. By including θ, more than 90% of the variability in the surface fluxes is explained. The coefficient of determination (r 2 ) range between 0.903 and 0.959. The relationship with H is especially improved. The coefficient of determination increased from 0.121 to 0.926. The MAE decreased by over 70% from 12.1 W m -2 to 3.5 W m -2. Flux r 2 Net Radiation 0.959 Soil Heat Flux 0.903 Sensible Heat Flux 0.926 Latent Heat Flux 0.946 Flux MAE Net Radiation 1.2 Soil Heat Flux 3.7 Sensible Heat Flux 3.5 Latent Heat Flux 5.9
Irrigation Effects Although the amount of irrigation water applied to each vineyard is unavailable, there is indirect evidence suggesting Site 2 was irrigated more heavily in 2014 than during the other years. 2016 (interpolated) The crop yield from Site 2 was significantly higher than other years. The cumulative ET map for Site 2 shows much higher ET during 2014 compared to 2013 and 2015. The advective enhancement of ET in Site 2 was both greater and began earlier in the day during the 2014 growing season compared to other years. 2014 2015
Concluding Remarks Although the meteorological conditions and management practices are the same in the two adjacent vineyards, they exhibit significantly different surface fluxes. The results of this analysis suggest that the discrepancy in the surface fluxes can be largely explained by differences in the vegetation density as expressed by LAI and differences water availability as expressed by θ. This study demonstrates the importance of vineyard management practices on ET. It also reinforce the importance of understanding the effects of the highly structured canopy of vineyards on turbulent production, exchange, and transport.
The authors would like to thank the many researchers within the USDA and other governmental agencies, university collaborators, and industry partners who have contributed to the GRAPEX. Specifically, the authors would like to thank E.&J. Gallo Winery for financial and logistical support. The authors would also like to thank Mr. Ernie Dosio of Pacific Agri Lands Management and the staff at the Borden/ McMannis Vineyard for their support of this research. Finally, the authors would like to acknowledge financial support for this research from NASA [NNH16ZDA001N-WATER.