Summertime Evapotranspiration in Eastern WA
Evapotranspiration (ET) refers to the total flux of water vapor from the land to the atmosphere. It includes two contributions: (1) evaporation from the ground and (2) the loss of water from plants, or transpiration, primarily through their leaves. ET is a key component of the water cycle, especially when the weather is warm and hence the evaporation rate and the growth of plants are often substantial. The potential or reference ET represents the flux of water vapor from a location with a low cover of vegetation covering the ground, and enough soil moisture so that the movement of water within this vegetation is not limited. In other words, the potential ET is essentially the actual ET when there is adequate soil moisture. It is calculated on the basis of the energy available for evaporation from the ground and transpiration from the vegetation.
The focus of the present summary is to examine how to potential ET during summer has varied in the Columbia Basin of eastern WA since the late 1980s, and the causal factors for these variations. Many crops in eastern Washington State are irrigated, and the potential ET minus precipitation relates to the amount of water that must be delivered to maintain adequate soil moisture through the dry summer months. This highlight constitutes the preliminary stages of a more thorough study that is in progress, and has been undertaken to complement the research carried out by OWSC on historical heat waves on the western side of the Cascade Mountains (Bumbaco et al. 2013).
We use daily values of potential ET estimated by the Kimberly-Penman method as provided by the Pacific Northwest Cooperative Agricultural Weather Network (AgriMet) under the auspices of the US Bureau of Reclamation. These data are available from the following web-site: http://www.usbr.gov/pn/agrimet/webarcread.html. Four low elevation stations in eastern WA have records extending back to the 1980s (George, Harrah, Lind, and Odessa). Here we consider the data from these four station for the months of June, July, and August for the period of 1987-2012. The Kimberly-Penman formulation is based on the measurements of minimum and maximum daily temperatures, relative humidity, daily solar radiation, and daily wind run (i.e., the daily mean wind speed multiplied by 24 hours) from the AgriMet weather stations. It combines the effects of the estimated net radiative heating, the sensible heat flux through the soil, and the heat transferred by the wind to quantify the energy available for ET.
As context for the regional results that follow, we first consider a map of the ratio of actual (not potential) ET to precipitation (P) for the US as a whole (Figure 1; reproduced from Sanford and Selnick 2012). Note the large variations in ET/P across the Pacific Northwest. The low values (<0.4) of this ratio for western WA signify that the run-off exceeds the ET in the mean. In fact portions of WA State have the dubious honor of having among the lowest values of ET/P in the lower 48 states. Residents of these regions, for example the southwest side of the Olympic Mountains, can attest to the copious precipitation that makes the ET/P ratio so low. On the other hand, the fraction of precipitation lost to ET is greater than 1 in the Columbia basin of WA, as for some other arid regions of the western WA. For this to occur water must be supplied, and in the case of eastern Washington it is largely by the Columbia River and its tributaries with source waters in higher and wetter areas. With this in mind, has there been much variability over the last 25 years in the potential ET and hence water required in the Columbia basin during summer? We were surprised by what we found.
Time series of seasonal mean potential ET for the 4 eastern WA stations considered (Figure 2) reveal not just the expected year-to-year fluctuations, but also prominent upward trends in ET over the period of record. The overall changes are equivalent to and extra 4” of water required in recent versus early years. The magnitude of this change motivated us to delve into the cause(s). We first examined how the time series of potential ET compared with their counterparts for mean air temperature (not shown). This exercise indicated that interannual variability in seasonal mean potential ET has a positive correspondence with temperature (with correlation coefficients of ~0.6 for the individual stations). On the other hand, negligible trends in temperature have occurred over the interval considered and so temperature does not seem to be associated with the overall increase found in potential ET. We also checked the seasonal mean values of solar radiation, humidity, and wind and something interesting turned up. Specifically, the period of record has features a general increase in solar radiation at all four stations, as shown in Figure 3. The trends in humidity are negligible; the winds have increased slightly (not shown). Summers in eastern WA that are relatively sunny with high potential ET tend to be associated with positive 500 hPa geopotential height anomalies, but we have no real explanation for the trends found here.
OWSC plans to pursue the subject of summer ET in eastern WA. This research will include analysis of the day-to-day variability in potential ET. The specific direction of this work will involve determining the regional atmospheric circulation patterns associated with anomalous ET on 1-3 day time scales. These results will be used to determine whether it is feasible to infer potential ET from global climate model simulations of these circulation patterns. If so, empirical downscaling may represent a reasonable means for estimating how potential ET is liable to change over future decades.
References:
Bumbaco, K.A., K.D. Dello, and N.A. Bond, 2013: History of Pacific Northwest heat waves: Synoptic pattern and trends. J. Appl. Meteor. Climatol. [in press]
Sanford, W.E., and D.L. Selnick, 2012: Estimation of evapotranspiration across the continental United States using a regression with climate and land-cover data. J. Amer. Water Res. Assoc., 49(1): 217-230.