Effects of Radial Variability in Sap Flow on up-Scaling of in Situ Evapotranspiration Measurements from the Scale of Individual Trees to the Grove Scale

Friday, 19 December 2014: 2:55 PM
Peter T Quinlan, Dudek Inc., Encinitas, CA, United States, Jill L Weinberger, Dudek and Associates, Inc., Hydrogeology, Encinitas, CA, United States and Daniel M Tartakovsky, University of California San Diego, La Jolla, CA, United States
Sap flow sensors are routinely used to obtain in situ measurements of evapotranspiration (ET) by individual trees. Flow rates within a single tree of some species, e.g., Quercus englemanii trees, exhibit high spatial variability. The latter impedes one’s ability to extrapolate single-tree ET estimates to their grove- and/or watershed-scale counterparts. A simplistic approach of using the highest observed sap flow rate might introduce a systemic bias into up-scaled estimates. We propose an approach to obtain a representative sap flow rate from all observed data; this rate is then applied to multiple trees in order to extrapolate ET rates from individual trees to the grove-scale. Eight thermal dissipation probe (TDP) pairs were deployed in individual Quercus englemanii trees, which had previously been monitored with only two TDP pairs. Analysis of the variability observed among the probes in the eight-probe arrays yields both a representative composite ET rate for the tree and quantification of the uncertainty associated with that rate. The probability distribution from the eight probe arrays can be applied to trees with one or two probes to estimate the probable range of representative ET rates for those trees and the composite rates for all measured trees is then applied to the grove as a whole.