Orographic Precipitation Studies
of the
Cascade Mountains of Washington

Investigation of Other Factors besides Wind
by Don Burrows

The next step of my investigation was in 1982. The purpose was to focus on the Cascade Mountains and see if other factors besides wind could improve the correlation with precipitation. In particular I was interested in seeing if there was any correlation between cold-air pooling east of the Cascades and precipitation over the eastern Cascades.
For this study ten stations were selected, four western slope stations (Snoqualmie Falls, Cedar Lake, Startup 1E and Baring), two divide stations (Stampede Pass and Stevens Pass) and four east slope stations (Lake Wenatchee, Plain, Leavenworth 3S and Cle Elum). Monthly precipitation data mean monthly 850 mb upper air data for the five years from 1978 through 1982 was obtained from the National Climatic Data Center.
The first step was to obtain the mean monthly 850 mb wind values for the Cascades by taking the mean of the UIL and GEG values. The next step was then to determine the principal wind component, &theta0 , for each station. This is the component of the 850 mb which produces the best correlation with the monthly precipitation values for that station. The results are shown in Table 1 along with the slope and intercept for the regression lines for each station.

Table 1. Correlation between 850 mb wind components and precipitation.
Location &theta0 rmax Slope Intercept
Snoqualmie Falls 225 0.845 1.495 -1.095
Cedar Lake 235 0.834 2.081 -0.048
Startup iE 240 0.804 1.178   0.639
Baring 235 0.872 2.790 -2.342
Stampede Pass 245 0.817 1.830 -0.455
Stevens Pass 235 0.851 2.269 -2.275
Lake Wenatchee 240 0.752 1.282 -1.725
Plain 225 0.773 0.782 -0.913
Leavenworth 3S 210 0.755 0.776 -0.805
Cle Elum 230 0.684 0.539 -0.408
         
West Slope 234 0.839    
Divide 240 0.834    
East Slope 226 0.741    
Mean 232 0.799    


In general the principal wind components were from the southwest. The most westerly was at Stampede Pass and the most southerly was at Leavenworth 3S. The correlation with precipitation was distinctly better over the western slopes and the divide (rmax= 0.80-0.85) than over the eastern slopes (rmax= 0.68-0.77). For Cle Elum, rmax seems to be significantly lower than for the other three east slope stations. Lake Wenatchee, Plain and Leavenworth 3S are all close to each other in the Wenatchee River drainage while Cle Elum is further south in the Yakima River drainage. Also Cle Elum is located in a wider valley with lower mountains around than the other stations. The precise reason for the difference is not known, however.

Next it was desired to see what other variables might provide more information and improve the correlation. Ten variables were selected for testing as shown in Table 2.

Table 2. Variables used with 850 mb wind components for correlation with precipitation.
Variable Description
V1 Mean (UIL and GEG) 850 mb Dewpoint Depression
V2 UIL 850 mb Dewpoint Depression
V3 GEG 850 mb Temperature
V4 GEG 850 mb Dewpoint Depression
V5 850 mb Temperature Difference (T850 UIL - T850 GEG)
V6 UIL Stability (Tsfc - T850)
V7 Surface Temperature Difference (TUIL - TEAT)
V8 Surface Temperature Difference (TSEA - TEAT)
V9 EAT Stability (Tsfc - T850 Mean)
V10 SEA Stability (Tsfc - T850 Mean)


V1,V2 and V3 are indicators of how saturated the air is with moisture. V3 is merely an indicator of the inland above surface temperature. V5, V7 and V8 are different ways to estimate the temperature difference west of the Cascades and east of the Cascades. Finally V6, V9 and V10 are lower atmospheric stability estimates at UIL, EAT and SEA respectively. The results of the two variable correlations are shown in Table 3.



Table 3. Multiple correlation coefficients between 850 mb wind components and other variables (V1-V10) with precipitation. The bold red values are the maximum values for each station.
Location V1 V2 V3 V4 V5 V6 V7 V8 V9 V10
Snoqualmie Falls 0.876 0.860 0.868 0.877 0.870 0.846 0.871 0.868 0.865 0.845
Cedar Lake 0.905 0.882 0.888 0.898 0.872 0.836 0.880 0.876 0.860 0.847
Startup 1E 0.858 0.833 0.856 0.860 0.840 0.805 0.841 0.836 0.824 0.813
Baring 0.894 0.879 0.900 0.900 0.907 0.877 0.905 0.902 0.901 0.872
Stampede Pass 0.852 0.834 0.853 0.854 0.851 0.818 0.857 0.852 0.848 0.818
Stevens Pass 0.880 0.856 0.889 0.895 0.900 0.862 0.902 0.899 0.901 0.852
Lake Wenatchee 0.792 0.759 0.804 0.812 0.819 0.776 0.833 0.836 0.840 0.753
Plain 0.820 0.781 0.824 0.843 0.833 0.791 0.852 0.851 0.856 0.775
Leavenworth 3S 0.804 0.758 0.806 0.842 0.841 0.797 0.851 0.849 0.867 0.763
Cle Elum 0.739 0.692 0.744 0.767 0.757 0.703 0.789 0.787 0.794 0.687
                     
West Slopes 0.883 0.864 0.870 0.884 0.872 0.841 0.874 0.870 0.862 0.844
Divide 0.866 0.845 0.871 0.874 0.875 0.840 0.880 0.875 0.874 0.835
East Slopes 0.788 0.748 0.794 0.816 0.812 0.767 0.831 0.831 0.839 0.745
Overall Means 0.842 0.813 0.843 0.855 0.849 0.811 0.858 0.855 0.856 0.802


V2, V6 and V10 seem to add very little if any new information. All of these have to do with west side information only. V1 and V4 show an improvement over V2 with V4 being slightly better than V1. The conclusion would seem to be that the GEG 850 mb dewpoint depression is more significant that the UIL value, which seems somewhat counterintuitive. V7, V8 and V9 which are measures of west east temperature difference or east side atmospheric stablity greatly improve the correlations for the east side stations but have little effect on the west side stations. This is certainly indicative of a possible cold air pooling effect that affects the eastern slopes but not the western slopes.
Next Page

© 2011 by Don Burrows, All Rights Reserved