Forests and Water: A State of the Art Review for Colorado

Home Browse Search

are still elevated relative to the control catchment. Pub- lished data show a linear decline in water yields from 1958 to 1986 (Figure 2.6) ( Troendle and Nankervis, 2000), and this linear decline has continued through 2000 (C. Troendle, Matcom Corp., pers. comm., 2003). The data suggest annual water yields will return to their pre- treatment values in approximately 60 years. Hydrologic recovery is expected to be faster in most other vegetation types relative to the subalpine spruce -fir forest type. Hydrologic recovery has been estimated to be on the order of 15 -45 years for species that resprout or are faster growing, such as aspen ( Troendle and Nan- kervis, 2000). Hydrologic recovery also should be faster in areas with less annual precipitation, as less regrowth is needed to return summer water losses to pre - harvest lev- els. Paired- watershed studies in the ponderosa pine zone in Arizona showed that harvest- induced water yields persisted for only seven years on the watershed that was completely clearcut, and from three to seven years for three of the four watersheds that were partially cleared. Approximately 10 years were needed for hydrologic re- covery on the watershed that was 77% cleared, and the longer treatment effect on this watershed was attributed to the combination of north - facing aspects and shorter slopes (Baker, 1986). Paired- watershed studies in Or- egon also have shown that the time to hydrologic recov- ery may be reduced if fast - growing riparian species, such as alder, become more prevalent after forest harvest, or there are species that resprout. A similar effect might be expected in Colorado. In some cases the replacement of mature or over - mature trees with younger, more vigor- ous vegetation can cause water yields to drop below the pre - harvest values after 10 -20 years (e.g., Vertessey et al., 1996), and it is not known whether an analogous de- cline in annual water yields will occur in Colorado. 2.2.5. Effects of vegetation change on the size of peak flows. The effect of forest management on the size of peak flows is an important concern for resource managers and the public. An increase in the size or duration of high flows can increase the sediment transport capacity, alter chan- nel geometry by scour or bank erosion (Schumm, 1971), and raise water levels (stage) in the affected streams. Hydrologic theory suggests that any reduction in forest cover will have a progressively smaller effect on peak flows with increasing flow magnitude or recurrence in- terval, and most reviews on the effects of forest manage- ment on runoff have come to a similar conclusion (e.g., Anderson et al., 1976; Austin, 1999). The underlying logic is that during the largest rain or snowmelt events the soils and vegetative canopy will have little additional 13 storage capacity, and under these conditions much of the rainfall or snowmelt will be converted to runoff regard- less of the amount or type of vegetative cover. If the amount and delivery of runoff are not significantly altered by roads or soil compaction, the primary hydro- logic change in rain - dominated areas is a reduction in interception. Since interception losses should be pro- portionally less in the larger storms (Zinke, 1967), one would expect proportionally smaller increases in the larger peak flows. An analysis of daily flows from 28 paired - catchment experiments showed that the median increase in the 95 -99' percentiles of daily flows (i.e., the flows that are equaled or exceeded for 3 -20 days per year) was about 10 -15% (Austin, 1999). The effect of forest harvest on larger flows (i.e., >2 -year recurrence in- terval) in rain - dominated areas is much more difficult to discern because of the variability between basins and the small number of events for analysis. The effects of forest management on runoff from these extreme rain events is still a matter of considerable controversy (e.g., Jones and Grant, 1996; Thomas and Megahan, 1998; Jones, 2000). In the mixed conifer zone in northern Arizona the re- moval of approximately 28% of the basal area increased the smaller peak flows (97.51 percentile) by 16% and the larger peak flows (991 percentile) by only 6% (Austin, 1999). In northern Arizona the peak flow from a 100 - year storm event was estimated to have increased by 20- 28% in watersheds where 30 -50% of the timber had been removed (Brown et al., 1974). For this same storm, the estimated increase in peak flow was approximately 90% for a watershed where 77% of the timber had been re- moved and another watershed where the vegetation had been converted from forest to grass. The largest increase (170 %) was on the watershed that had been clearcut and subjected to 100% ground disturbance (Brown, 1974). In snowmelt - dominated areas the effects of forest har- vest on peak flows is more consistent, but one must still be careful to define the peak flow of concern and whether the increase is in relative or absolute terms. In general, forest harvest in snowmelt- dominated areas in the Rocky Mountains will increase the size and frequency of the larger flows, but there is little evidence for an increase in the highest instantaneous peak flows (i.e., flows with a recurrence interval greater than two years). The change in the size of peak flows in snowmelt- dominated areas is due to multiple factors, and these include an increase in the amount of shortwave radiation that reaches the snow- pack surface, an increase in the turbulent heat transfer as a result of higher wind speeds at the snowpack surface, an increase in latent heat transfer due to condensation