Can carbon isotopes be used to predict watershed scale evapotranspiration?

Background/Question/Methods

We propose a new method for estimating evapotranspiration from entire mountainous watersheds. We predicted the average canopy stomatal conductance (g_s) of watershed populated by ~40-y-old Douglas-fir trees by taking advantage of the ¹²C:¹³C fractionation that occurs when plants fix carbon. To estimate g_s we first determined the isotopic composition of ecosystem respiration (δ¹³C_ER) as derived from the ¹²C:¹³C ratio of CO₂ entrained within nocturnal drainage flows exiting the base of the watershed. The measurements of δ¹³C_ER were combined with common biophysical models to estimate g_s for the entire watershed. By applying estimates of g_s and other measured environmental variables to the Penman-Monteith equation, we estimated the evapotranspiration from an entire watershed. The results are compared to direct measurements of sapflow from four locations within the watershed.

Results/Conclusions

The average direct sapflow measurements were surprisingly similar to the isotope derived estimates of evapotranspiration. This technique was possible in this watershed because the source area for respired CO₂ passing the tower at the base of this watershed was stable, the cold air drainage was persistent and CO₂ entrained within the drainage was decoupled from the synoptic wind patterns. However, while promising, the results may be fortuitous. The potential limitations and future research necessary to fully demonstrate the viability of this method are discussed.