SCHEDULING TIMBER HARVEST FOR HYDROLOGIC CONCERNSandrewsforest.oregonstate.edu/pubs/pdf/pub572.pdf · SCHEDULING TIMBER HARVEST FOR HYDROLOGIC CONCERNS ... in that a particular column

' - : PROCEEDINGS OF WORKSHOP ON

SCHEDULING TIMBER HARVEST FOR

HYDROLOGIC CONCERNS

1161,101.11WM411114.' • •

. '''4;121).118

- 7DEPARTMENT OF AGRICULTURE

FOREST SERVICE

PACIFIC NORTHWEST REGION

PACIFIC NORTHWEST FOREST AND RANCE EXPERIMENT STATION'

DECEMBER 1979

WORKSHOP NOTES

TIMBER HARVEST SCHEDULING AND SEDIMENT ROUTING

Fred SwansonResearch Geologist

Forestry Sciences LaboratoryCorvallis, Oregon

Studies of complex natural systems, such as ecosystems, have taught usthat all parts of such systems are connected to all others, and that itis difficult, if not misleading, to analyze a system by looking at onlyone fragment. The purpose of this section of the workshop is to atleast raise the issue of considering the entire soil/sediment routingsystem in forest watersheds. The term "sediment routing" refers to thefull range of erosion processes and soil and sediment storage sites thatare involved in the movement of soil and sediment through watershedsfrom ridge top to the sea. We can view sediment routing as the transferof materials from one temporary storage site to another. The sedimentrouting regime thus involves both storage sites and transfer processes.Although it is convenient to distinguish hillslope and channel sites andprocesses for the sake of discussion, it is necessary to recognize thatthey are closely linked in an overall sediment routing context.

Erosion from hillslopes is accomplished by a group of processes thatinteract in complex fashion. Some processes are episodic, while othersoperate more continuously. Some processes operate in chain'reactionfashion, such that acceleration of one process may lead to increasedsoil transfer by successive processes. Processes are also superimposedin space, in that a particular column of soil may be subjected to soiltransfer by creep, root throw, surface erosion, and other processessimultaneously. Because of this complexity, it is commonly difficult topick out a single erosion process or group of processes to focus on inorder to assess the impact of management activities on hillslopeerosion. For example, .we may analyze debris avalanche erosion andassess short term (decade) increase in erosion rate followingclearcutting. However, long term (several rotations) effects of cuttingmay be keyed more to those processes that resupply temporary storagesites that repeatedly, but infrequently, fail by debris avalanche. Inthis case, an overyiew of sediment routing provides a basis fordistinguishing long term from short term impacts of managementpractices. Evaluation of management impacts may vary substantiallydepending on the time scale on which the system is viewed.

Sediment transport through channel systems also involves complexinteractions aMong,storage sites and transfer processes. For example,coarse sediment breaks down to finer particles during transit and

deposits of coarse particles form storage sites for finer material.Large organic . debris in many small forested streams dominates hydraulicsand sediment transport by forming steps which dissipate stream energyand trap sediment. Removal of logs from a channel as part of a loggingoperation may release large quantities of sediment which entered thechannel under conditions of natural vegetation. I raise this issue topoint out-that forest practices may affect storage sites as well as thetransfer processes we typically measure in assessing management impacts.

To say harvest scheduling and sediment routing in the same breathsuggests that we can practice a form of even flow, sustained yield ofsediment. This thought raises three questions: Can we do this? If so,how? Should we? The first two questions can be considered in light oftwo limitations: (1) foreclosed options in lands already extensivelyentered, and (2) lack of knowledge of sediment routing systems needed tointerpret effects of harvest scheduling. Perhaps by the time we havecut 30 to 50% of a forested landscape, most future decisions concerningcutting patterns are determined by history of past decisions. Placementof new cuts may be determined by existing patterns of cuts and roads andsubsequent mortality due to blow down, pathogens, and other factors. Ofcourse, we could never actually practice "even flow" of sediment yield,because in most steep, forested terrain sediment production and yieldare regulated predominantly by infrequent, extreme storm events,regardless of the character of management activities. However, inforested ecosystems typified by infrequent, catastrophic wildfirenatural sediment routing regimes may have been much more "flashy" thanrouting regimes in the same landscapes under 'conditions of managedforests. This leads back to the "should we?" question. One can arguethat some ecosystems have experienced wide spread, catastrophicdisturbance in the past. So, if we have some success in practicing"even flow" of sediment yield, we may significantly alter the naturalsediment routing regime of a geomorphic system by reducing some of theextreme periods of sediment movement.

What are our overall sediment management objectives: mimic nature orsmooth both natural and man—imposed variations in sedimentation? Whatis the role of roads in this scenario? If vegetation disturbanceinfluences the sedimentation effects of major storms, managed forestswill have less variation in long—term sediment yield than naturalforests subject to periodic, extensive wildfire (assuming roads are ofminor importance ih the managed forest). The managed forest of largedrainages on Federal land contains stands of a broad age distribution upto the rotation age. Natural forests in the same area may havecontained stands of only a few age classes and may have been largelyfreshly burned when some major storms occurred.

To interpret effect's of scheduling on sediment routing it is necessaryto understand (1) downstream effects of individual activities, (2)

2

cumulative effects of multiple activities, and (3) important thresholds

or feedback mechanisms which may trigger abrupt shifts in system

behavior. In general, our poor understanding of each of these matters

limits our ability to prescribe harvest scheduling to regulate sediment

routing. However, several specific examples in which scheduling may

affect sediment movement do come to mind.

Rapid mass movements down channels, termed debris torrents, are

• potential off-site consequences of timber harvest in some areas of steep

hillslopes and channels. We have some ability to identify critical

sites which will be points of origin of torrents that can move from

first- to third- or fourth-order channels (Swanson and Lienkaemper

1978). Since freshly clearcut areas have the highest probability of

torrent occurrence, we may wish to schedule cutting so that only a small

proportion of critical sites in a large watershed are in young clearcuts

at any particular time. In this way we distribute impacts over time and

space.

Scheduling may also be effective in minimizing impacts of timber removal

on slumps and earthflows--slow, deep-seated mass movement features.

Removal of vegetation reduces evapotranspiration, thereby increasing

soil water availability which may lead to increased movement of

seasonally active slumps and earthflows. This hypothesized link between

forest cutting and accelerated movement has not been convincingly

demonstrated. If cutting does lead to accelerated slump and earthflow

movement, hydrologic effects of forest removal could be minimized by

scheduling cutting in the earthflow a-rea and on any adjacent land that

drains into the earthflow terrain. Cutting could be scheduled so that

the evapotranspiration potential of the entire watershed area

influencing the earthflow never falls below a particular value.

The extremely high sediment producing basins of northwestern California

provide examples of sediment routing systems in which harvest scheduling

might have, been of some value. Naturally, very high sedimentation rates

in the Van Duzen River (Kelsey 1977) and Redwood Creek (Janda 1978)

basins have possibly been increased by forest management practices that

have included extensive impacts of roads and continuous cutting over

large areas. During the December 1964 flood and some subsequent floods

sediment from upper basin source areas was deposited in some main stem

reaches. This aggradation of sediment from upstream sources triggered

local bank cutting, channel widening, mortality of streamside

vegetation, and entrainment of sediment that had previously been stored

in floodplain deposits. Apparently, some threshold of stability was

exceeded along some main stem reaches triggering negative feedback

mechanisms. Perhaps these sequences of events could be avoided by

harvest scheduling to distribute management impacts over longer time

periods.

3

Effects of increased sediment availability are conspicuous in theseparticularly sediment-rich systems. These effects are more subtle andmore difficult to document in most other systems. Even in the northerncoastal California areas where land use practices of the 1950's and

1960's 'were much rougher than today, the degree of management impact onobserved channel changes has been hotly debated. Were the channelchanges due to the flood alone? So harvest scheduling for sedimentmanagement may be difficult to sell to hesitant customers.

CONCLUDING COMMENTS

Here are some personal feelings and tentative assumptions from a west-

'. side, steep forest land perspective. They are offered as points fordiscussion.

Increased sedimentation following logging is generally duemore to increased sediment availability than to increased peakor total flows. If so, hydrologic arguments for harvestscheduling may be weaker than some other arguments. Can we,and should we, distinguish systems where sediment yield isflow limited from those that are sediment supply limited?

If harvest scheduling were deemed useful and possible,prescriptions for scheduling to regulate sediment routing mayvary substantially between landscapes with contrastingsediment routing regimes. Sediment-rich and sediment-poorsystems may vary greatly in the problems and possibilitieseach presents for management of sediment routing by harvestscheduling. Earthflow dominated landscapes may need adifferent approach than areas where debris avalanches are veryimportant. Do these different types of systems requiredifferent, approaches to scheduling or just different degreesof the same basic approach?

Much of our detrimental impact on habitat and sedimentation insmall- and intermediate-sized streams may be due tomanagement practices which reduce streamside vegetation andlarge organic debris in channels rather than practices thatalter hydrology and sediment supply in upstream areas. Liveand dead vegetation helps stabilize stream banks, createsdiverse habitats, especially pools, provides cover, andperforms other functions. If so, harvest scheduling will haveonly a secondary effect in protecting this part of the aquatichabitat.

We do not know enough about sediment routing to say howharvest scheduling affects it. How should we be thinkingabout and'monitoring soil/sediment movement and storage in

4

watersheds so we can determine effects of harvest schedulingand other management activities?

The answer to this question is crucial to successful managementof sediment with scheduling techniques.

LITERATURE CITED

Janda, R. J. 1978. Summary of watershed conditions in the vicinity ofRedwood National Park, Calif., U.S. Geological Survey. Open filereport. 78-25, 82 p.

Kelsey, H. M. 1977. Landsliding, channel changes, sediment yield, andland use in the Van Duzen River basin, north coastal California,1941-1975. Ph.D. thesis, Univ. of Calif., Santa Cruz.

Swanson, F. J., and G. W. Lienkaemper. 1978. Physical consequences oflarge organic debris in Pacific Northwest streams. USDA For. Serv.Gen. Tech. Rep. PNW-69, 12 p. Pac. Northwest For. and Range Exp.Stn., Portland, Oreg.

5