Peter A Bisson Richard E Bigley Alex D Foster Shannon M Claeson Steven M Wondzell NASA Skokomish Queets Humptulips Clearwater Hoh Sol Duc Hoko Fulton Riparian Ecosystem Management Study Phase 1 ID: 186805
Download Presentation The PPT/PDF document "Washington’s Riparian Ecosystem Manage..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Washington’s Riparian Ecosystem Management Study (REMS): Approaches, Surprises, and Lessons Learned from 12 Years of Headwater Stream Research Peter A. BissonRichard E. BigleyAlex D. FosterShannon M. ClaesonSteven M. WondzellSlide2Slide3
NASASlide4
SkokomishQueetsHumptulipsClearwater
Hoh
Sol Duc
Hoko
Fulton
Riparian Ecosystem
Management Study – Phase 1
Location of study sites
1996-1999
Olympic National Park
Riparian condition:
no buffers
narrow (<20 m) buffers
thinned buffers
unloggedSlide5
We surveyed:Fishes Amphibians AssemblagesCutthroat trout Tailed frog All fish speciesTorrent sculpin Cope’s giant salamander All stream- dwellingCoastrange sculpin Torrent salamander amphibiansSlide6
Environmental variablesSite-level Landscape-level In-stream features Watershed attributesChannel gradient Watershed area% pools Drainage density% riffles & cascades Mean elevation% glides % south-facing% silt & sand % steep slopes (>60%)% gravel & pebble% cobble Riparian forest age% boulder & bedrock % early-seral riparian zone % mid-seral
riparian zone
% late-
seral
riparian zone
Upland conditions
% early-
seral
upland forest
% mid-
seral
upland forest
% late-seral upland forest Road density Landslides/debris torrentsSlide7
For headwater trout and sculpins, there was a negative association between abundance and late-seral riparian forest; however, fish tended to be most strongly influenced by in-stream habitat.Fish abundance was positively correlated with riparian and watershed features associated with increased primary production.Slide8
Amphibians responded positively to late-seral riparian forest and the amount of late-seral forest in their watersheds. Their abundance was negatively correlated with roads.Slide9
Northwest Forest Plan Aquatic Conservation StrategyReeves et al. 2006Slide10
Adaptive Management Areas – intended to provide site flexibilitySlide11
Surprise #1Despite the designation of Adaptive Management Areas (AMAs) under the Northwest Forest Plan, we were unable to convince National Forests in western Washington to implement a series of experimental riparian treatments at the small watershed scale.Slide12
LandsatSlide13
BACI-type study of alternative riparian management in clusters of small watershedsSlide14
DNRSlide15
Unlogged controlsContinuous (“fixed-width) buffersDiscontinuous (“patch cut”) buffersNo buffersSlide16Slide17
Capitol Forest
Willapa Hills
Seattle
Washington State
22 Streams in 6 Blocks in 2 Forests
See Control Continuous Clearcut
Rot Control Continuous Patch
Tags Control Continuous Patch Patch Clearcut
Ells Control Continuous Clearcut
Split Control Continuous Clearcut Clearcut
Lonely Control Continuous Patch ClearcutSlide18
Surprise #2A completely randomized block design was stymied by engineering considerations. Control watersheds were always at the end of a cluster.Slide19
Surprise #3Despite good intentions, implementing the treatments didn’t always go as planned. Buffers were not always consistent and logging took place over two years instead of one.Slide20
ControlSlide21
Continuous/fixed widthSlide22
Discontinuous/patch cutSlide23
No bufferSlide24
Surprise #4Drought conditions prevailed through much of the study, especially the early years!Slide25Slide26
Surprise #5Unanticipated post-logging site preparation treatments!Slide27Slide28
Surprise #6Large wind storms and blowdown at some of the sites!Slide29Slide30Slide31
In-stream measurementsSlide32Slide33Slide34
Unbuffered streams were significantly warmer, but… Temperature increases were small Trend did not match buffer design, and Responses within treatments were highly variable Temperature responses were correlated with streambed texture, wetlands, and length of stream Surface area exposure Amount of hyporheic exchange
Temperature
Summary:Slide35
Terrestrial Mites & Collembola abundanceAbundance increased in all logged sites, relative to the Control.
-0.5
-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
0.4
0.5
Log (Trt / Control)
PRE
POST 1
POST 2
(mean + 1 SE)
Continuous Control
Patch Control
Clearcut ControlSlide36
Mayfly abundanceEphemeroptera = Paraleptophlebia, Baetis, Diphetor, Heptageniidae sp.Log (Trt / Control)
(mean + 1 SE)
Continuous Control
Patch Control
Clearcut Control
PRE
POST 1
POST 2
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
Abundance tended to
decrease
in all logged sites (esp.
clearcut
) 2
nd
year post-harvest, relative to the Control.Slide37
Stonefly abundancePlecoptera = Zapada, Soyedina, Sweltsa, IsoperlaLog (Trt / Control)
(mean + 1 SE)
Continuous Control
Patch Control
PRE
POST 1
POST 2
-0.6
-0.5
-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
0.4
Abundance
decreased
in the
unbuffered
sites 1
st
year post-harvest, relative to the Control.
Clearcut ControlSlide38
Litterfall measurementsSlide39
Litter fall total dry massLitter-fall decreased in the clearcut sites, relative to the Controls.
Clearcut
Control
Log (Trt / Control
)
(mean + 1 SE)
-0.3
-0.2
-0.1
0
0.1
PRE
POST 1
POST 2
Continuous Control
Patch Control
Clearcut ControlSlide40
Shredder abundanceZapada, Soyedina, Lara, Yoraperla, MoseliaShredder abundance decreased in the clearcut sites, relative to the Controls.
-0.6
-0.4
-0.2
0
0.2
0.4
Log (Trt / Control
)
(mean + 1 SE)
PRE
POST 1
POST 2
Continuous Control
Patch Control
Clearcut ControlSlide41
Riparian mollusksSlide42
Warty jumping slugPolydesmid millipede
P. BissonSlide43
Capitol Forest BlockPeriod
Fall 02
Spring 03
Fall 03
Fall 04
Spring 05
Fall 05
Spring 06
Mollusk Density m
2
0
2
4
6
8
10
12
Control
Fixed Width
No Buffer
Before
After
Willapa Block A
Period
Spring 04
Fall 04
Spring 05
Fall 05
Spring 06
0
2
4
6
8
10
12
Control
Fixed Width
No Buffer
Before
After
Mollusk Density m
2
Willapa Block B
Period
Fall 02
Spring 03
Fall 03
Fall 04
Spring 05
Fall 05
Spring 06
0
2
4
6
8
10
12
Control
Fixed Width
No Buffer - B
No Buffer - E
Before
After
Mollusk Density m
2Slide44
ConclusionsSetting up and implementing a BACI-type study at the small watershed scale was a pain! Patience and a willingness to adapt to surprises were needed.The initial 2 year pre-treatment and 2-3 year post-treatment monitoring period was inadequate for many response metrics, but the watersheds were changing anyway and it was difficult to interpret results.Preliminary findings suggest that biophysical changes in the streams and riparian zones were not great after logging, but differences were most apparent in sites without stream buffers.
Analyses were often confounded by unanticipated events.Slide45
Use a design strategy that employs legitimate treatment-control sites, or that randomizes treatment assignment over enough sites that differences average out.Verify, with power analysis, that the proposed design is adequate to resolve the anticipated effect, given available estimates of pertinent variability, including uncontrolled environmental variation and random site differences.Implement the design, and stick with it until you resolve the important question(s) or until the data show that the actual uncontrolled variation is so different from what was assumed during the planning phase that the design is not adequate to resolve the question.Don’t be afraid to investigate novel response metrics.
Lessons learned…Slide46Slide47Slide48Slide49Slide50Slide51
http://treesearch.fs.fed.us/pubs/36997