Water is delivered to stream channels by a combination of overland flow and subsurface flow. Subsurface flow can occur near the surface just below the forest floor, through macropores or pipes, or as deep groundwater flow through the pore spaces in soil and bedrock. Overland flow occurs over saturated soil, usually adjacent to stream channels. The rate at which runoff is converted to stream flow is governed in part by stream drainage density. A watershed with a high drainage density will produce higher peak flows than one with a low drainage density, all other factors being equal. Also, the greater the rainfall and the rate of snowmelt, the higher the resulting peak flow. Low flows are sustained by deep groundwater.
Trees play a critical role in governing runoff rates. Mature tree crowns intercept about 30% of annual precipitation, which is then lost to evaporation. Trees also control snowmelt by casting shade on the snowpack and reducing wind speeds at the snow surface. Trees withdraw water from the soil through transpiration. Riparian species (such as streamside alder) rely on groundwater that supplies streams during low flow periods.
The presence of hillside roads increases peak flows. Hillside roads intersect surface and subsurface flows. Road ditches capture and concentrate small streams and overland flow between cross drains. This tends to increase the rate of delivery of surface water to the main stream. Subsurface water is intercepted by road cutslopes and, unless this water is returned to the soil quickly, it becomes surface flow. The result is that roads convert groundwater seepage into surface stream flow, significantly increasing the rate of water movement to the main stream channel and effectively increasing the stream drainage density. Under wet conditions, a high road density can significantly increase peak flows.
Clearcutting removes the forest canopy and changes the hydrologic behaviour of an area by altering interception, transpiration, and snowmelt processes. These changes usually result in increased peak flows and an increase in annual water yield. As the forest regenerates, the forest canopy develops, re-establishing the interception and transpiration processes. This re-establishment is referred to as hydrologic recovery.
The effect of canopy removal on peak flows is more pronounced at elevations above 300 m due to the regular occurrence of snow and the critical role of tree cover in snow interception and melt processes. Peak flows that occur due to rainfall in late fall and early winter from low elevations are influenced by logging to a lesser extent because there is usually little or no snow to contribute to runoff at these elevations. At higher elevations, more snow accumulates in openings than under a mature canopy due to the removal of interception losses. Thus, more snow is available to runoff by either rain-on-snow or radiation snowmelt processes. In addition, the canopy slows runoff production by shading the snowpack during radiation snowmelt, and also by intercepting rainfall and buffering the effect of wind on the snowpack during rain-on-snow events. Therefore, canopy removal has the potential of increasing peak flow generation significantly. This effect is decreased by subsequent regeneration of the forest. The potential for increases in peak stream flow is thus related to the ECA of the watershed.
Research is ongoing to determine the relationship between average tree height of a stand and hydrologic recovery in coastal British Columbia. An interim hydrologic recovery relationship is shown in Table 5-1. Note that below a height of 3 m, the trees are not effective at providing interception storage or at buffering from radiation snowmelt or rain-on-snow processes. Thus, recovery starts at a stand height of 3 m. At a height of 10 m, the regenerating stand approaches recovery. However, full recovery is unlikely in second-growth plantations because canopy structure will be different than in old growth, even at rotation ages. Furthermore, the recovery relationship given in Table 5-1 assumes full stocking. Often, regeneration is patchy, particularly at heights below 7 m.
Table 5-1. First approximation of hydrologic recovery for fully stocked stands that reach a maximum crown closure of 50%–70%
Low flows in summer or winter can harm fish populations by reducing the amount of available habitat. During the summer, this is exacerbated by the added stress of higher oxygen needs of fish and lower dissolved oxygen concentrations when the water is warmer. During the winter, low flows cause less oxygen stress, but overwintering eggs can be damaged by freezing or ice movement.
Both summer and winter low flows result from long periods during which the water being discharged from soils and bedrock is not replenished by rain or snowmelt. Trees affect low flows by intercepting rain and snow, by reducing the amount of water entering the soil and, through transpiration, by removing water from the soil. Transpiration, however, is directly related to moisture availability. Consider what happens in a clearcut under different conditions. During a wet summer, interception loss in a clearcut is low, resulting in more water entering the soil than would occur under a forest canopy. In addition, the water that would have been transpired from the soil by trees is available for groundwater recharge and stream flow. As a result, under wet conditions, the summertime low flow after clearcutting is greater than the low flow that would have occurred in the forest. In contrast, during a summer without rain, water input to the soil is zero regardless of whether the site is forested or not. Transpiration losses in the clearcut would probably be less than in the forest, but the forested site would have very low transpiration losses anyway. Consequently, stream flow from both sites would be very low and clearcutting would have little effect on the water balance.
There is a general public perception that clearcutting dries out soils. This is probably because the top layers of soil do in fact become drier upon exposure to stronger sunlight and wind. However, the deeper soil layers in the rooting zone of trees have been shown to have higher moisture content after clearcutting. The net effect is that total soil moisture tends to increase after clearcutting. This effect diminishes as a site becomes revegetated until there is no detectable difference within 10 to 15 years after logging.
Low flows may occasionally appear to decrease as a result of channel aggradation. In some cases, water continues to be discharged from a basin. However, it moves below the surface through the stream bed where channel aggradation has occurred.
Watershed studies have shown that tree removal tends to increase mean monthly flows a moderate amount in August, September, and October during the 10- to 15-year revegetation period. This is probably beneficial in cases where water can be impounded for human use or for delayed release downstream. However, in most cases, there may be no benefit to fish, since the very lowest flows are not increased by harvesting.
In summary, timber harvesting appears to have a negligible, or slightly positive, effect on summer low flows in most cases. Winter low flows are probably not affected by forestry activities.
Tally all opening information (as shown in Table 5-3) and summarize it in columns A, D and G in Form 2.
Example ECA calculation:
ECA below the 300 m contour total sub-basin area (column B): Divide the value obtained for ECA below the 300 m contour (column A) by the area of the entire sub-basin.
Weighted ECA below the 300 m contour (column C): After an area has been harvested, both winter snow accumulation and spring melt rates increase. This effect is less important at lower elevations, since the snow disappears before peak flow. Directly transfer results from column B into column C (ECA weighting is equal to 1).
ECA between the 300 and 800 m contours (column D): To estimate this value, determine the height of regeneration in each polygon between the 300 and 800 m contours on the 1:20 000 forest cover map. Heights may need to be extrapolated if reference material is not up-to-date. The area of each opening will then have to be reduced by the appropriate percent hydrological recovery (see Table 5-1 and ECA assumptions). Tally all opening information and summarize in Form 2.
ECA between the 300 m and 800 m contours total sub-basin area (column E): Divide the value obtained for ECA between the 300 m and 800 m contours (column D) by the area of the entire sub-basin.
Weighted ECA between the 300 m and 800 m contours (column F): This zone is the most likely to experience rain-on-snow events. Therefore, harvesting will have the greatest impact and is, hence, of greater concern than at lower elevations. To obtain this value, multiply column E by an ECA weighting factor of 1.5.
ECA above the 800 m contour line total sub-basin area (column H): Divide the value obtained for ECA above the 800 m contour line (column G) by the area of the entire sub-basin.
Weighted ECA above the 800 m contour line (column I): This is the typical winter snowpack zone. While enhanced runoff can be produced, the runoff peaks following logging from deep snowpacks are significantly less than from rain-on-snow events in the 300 to 800 m zone. Directly transfer results from column H into column I (ECA weighting is equal to 1).
Peak flow index (indicator #1): The peak flow index is derived from estimates of the area that are equivalent to the ECA. Add the weighted ECAs from column C, column F and column I to obtain the peak flow index in indicator #1.
Table 5-3. ECA tally sheet
Use this sheet to tally individual results for column A, column D and column G in Form 2.