2014-05-15_REVISION - M1987064 - Laserfiche WebLink

Home Browse Search

D.C. Goodrich et al. /Agricultural and Forest Meteorology 105 (2000) 281 -309 coefficient, aatt, was set equal to 3 following Magnani et al. (1998). The only remaining quantity required to compute C/W transpiration using Eq. (1) is the bulk canopy resistance (r,). This is the resistance to water vapor transport from inside the leaf to the leaf surface, which is regulated by the plant's stoma in response to environmental conditions. The canopy resistance is related to individual leaf stomatal resistance (rs) by the following expression (for amphistomatal leaves): _ rs r� (4) 2LAI During 1997 measurements of rs, using a leaf poro- meter, were made of leaves on several trees during the April, June and August synoptic measurement pe- riods. Personnel and economic constraints prohibited obtaining a sufficient number of leaf level stomatal resistance measurements to obtain a representative sample to approximate the required bulk canopy re- sistance at the tree or stand level. Therefore, the bulk canopy resistance was treated as a calibration parameter. 4.2.1. Calibrating the P —M model For each synoptic measurement period, the stand level C/W transpiration [LT -1] (2238 trees) in the Lewis Springs intensive study reach was derived by dividing the transpiration in volume per unit time by the remotely sensed estimate of the stand canopy area. Using this data, the bulk canopy resistance can be computed by using a re- arranged form of Eq. (1): r r rAA +pacpD/ra — A — 1� 5 c = a LL ylE y ( ) An average bulk canopy resistance was calculated for each SNIP measurement period by taking the average of all computed resistances between 09:30 and 14:30 LST under mainly sunny conditions (when the solar radiation was equal or greater than 300 W m -2). 4.3. Riparian corridor water balance As noted in Section 1, water balance methods have been utilized in the past to estimate riparian ET indirectly by attempting to measure or estimate all components of the water balance except ET and then computing ET as the residual. In this study, the inde- 289 pendently derived estimates of riparian ET described above are combined with independent estimates of the additional water balance components over a se- lected reach of the San Pedro riparian corridor for a selected period of time. If all the components of the balance are correctly measured or estimated, the residual of the water balance equation will sum to zero. Assuming there are no significant compensating errors in the water balance components, the closure of the water balance provides a check on the postulated description of water fluxes into and out of the riparian system including the methods to estimate riparian ET described above. For our purposes, the water balance was expressed in volumetric terms (m3) for a specific time period and control volume as schematically illustrated in lower portion of Fig. 3 as On + GWnet + PPtws — Qout — TC /w — ETm — ETs — Ews — AStorage = 8 (6) The first three terms in this equation constitute water inputs into the control volume and the following five terms make up the outputs. For the inflow terms, Q;n is the volume of water flowing into the control volume as streamflow, GWnet the net volume of groundwater flowing into the control volume, and Pptws the vol- ume of water falling as precipitation on the stream water surface. For the outflow terms, Qout is the vol- ume of water flowing out of the control volume as streamflow, TC /w the volume of water transpired by the C/W forest, ETm the volume of water evaporated and transpired by the mesquite, ET, the volume of water evaporated and transpired by the sacaton, and Ews the volume of water evaporating from the stream water surface. AStorage is the change in storage of water in the floodplain aquifer and the change in soil moisture in the unsaturated zone within the control volume, and s the residual error of the water balance. The water balance was performed at both the stand level and a significantly larger reach of the river. Two primary factors were considered in selecting the river reach and periods of time over which to compute the water balance. First, an attempt was made to make optimal use of the measurements in hand. Second, we attempted to minimize the impacts of our lack of knowledge of how the mesquites and C/W utilize rainfall and surface runoff (Snyder and Williams,