Transcript AMT 1-7 Aquatic Filters - University of Alaska system

YKL REA Northern
Pike Model
Photo: ADF&G
Fish Distribution Models
ADF&G AFFID
species
occurrence data
Process AFFID data
for use in models
GIS source data
Create stream
network and
landscape predictor
variables in GIS
Fish
distributions
Predict species
habitat across REA
study area
Classification tree
and random forest
models
Evaluate model
performance
Photo: USFWS
Stream Network
Used TauDEM to process DEM
1. Add in additional HUCs on boundary of study area that
flow into the study area
2. Fill pits
3. Calculate flow direction (D8 method)
4. Calculate contributing area
5. Create stream network based on curvature method and
drop analysis
Predictor Variables
Predictors of Fish Habitat
Elevation
Permafrost
Gradient
Floodplain
Slope over area ratio
Stream order
Watershed area
Average watershed annual
precipitation
Average watershed annual
temperature
Average watershed elevation
Average watershed slope over area
ratio
Average watershed slope
Percent permafrost cover
in USFWS
Photo:
watershed
Percent lake cover in watershed
Process AFFID data
-
-
-
-
-
Presences from AFFID
and ADF&G/BLM
telemetry project in
Kuskokwim
Absences from projects in
AFFID that listed fish
community sampling as
an objective
Resampled data in areas
of high intensity (Pebble
area and telemetry)
Shifted points along flow
direction grid until they
reached the stream
network
Extracted all predictor
variables to each data
point
Classification Trees
Asterospicularia laurae
Classification Tree Analysis
Steps:
– Identify the groups
– Choose the variables
– Identify the split that
maximizes the
homogeneity of the
resulting groups
– Determine a stopping
point for the tree
– Prune the tree using
cross-validation
Shelf: Inner, Mid
Absent
0.97
(263)
Shelf: Outer
Location: Back, Flank
Absent
0.78
(64)
(De'Ath and Fabricious 2000)
Location: Front
Depth < 3m
Absent
0.56
(9)
Depth ≥ 3m
Present
0.81
(37)
Misclassification rates: Null = 15%, Model = 9%
Photo: USFWS
Random Forests
Creates many classification trees and combines predictions
from all of them:
- Start with bootstrapped samples of data
- Observations not included are called out-of-bag (OOB)
- Fit a classification tree to each bootstrap sample, for each
node, use a subset of the predictor variables
- Determine the predicted class for each observation based
on majority vote of OOB predictions
- To determine variable importance, compare
misclassification rates for OOB observations using true and
randomly permuted data for each predictor
Run models in R
ct1<-mvpart(pres.f~.,data=fish.pred1[s1,],xv="1se")
rf1<-randomForest(pres.f~.,data=fish.pred1[s1,],ntree=999)
CT training
summary
CT validation
RF training
RF validation
1
0.096
0.161
0.108
0.113
2
0.108
0.194
0.092
0.161
3
0.12
0.161
0.096
0.097
4
0.12
0.145
0.116
0.129
5
0.108
0.194
0.108
0.145
6
0.072
0.097
0.112
0.048
7
0.124
0.177
0.108
0.097
8
0.112
0.097
0.104
0.081
9
0.137
0.081
0.124
0.065
10
0.12
0.145
0.141
0.1117
0.1452
0.1109
Photo: USFWS
0.097
0.1033
Model Performance
0
1
Confusion Matrix
0
1
Error
184
13
6.6%
21
93
18.4%
Photo: USFWS
Top five variables are watershed area, stream
order, stream elevation, percent of watershed
covered by lakes, and stream floodplain.
Northern Pike
Results:
~ 10,900 km of predicted
summer habitat (restricted to
stream reaches > 1 km in
length)
Predictor
Watershed area
Stream elevation
Stream floodplain
Watershed lake
cover
Stream order
Presence
Absence
13,000 km2
60 km2
60 m
200 m
Yes
No
2.8%
2.1%
4th
1st
Expanded ice-free season
Permafrost
Change Agents
Drivers
CE
General Effect
CE-Specific Effect
Increase depth of active
layer will increase lake
drainage area
Infrastructure
Harvest
In creased toxicity
Contaminants
Invasive
Macrophytes
Reduction in age at maturity and shift in
spawning season
Northern Pike
Esox lucius
Habitat
Temporary increases in nutrient inputs
Human Uses
Elodea ssp could reduce quality of spawning habitat
Direct destruction of habitat, hindrance of migration routes,
increased downstream turbidity and sedimentation
Climate Change
Subsistence harvest pressures on
overwintering populations
Increased contaminant sources
Precipitation
Bioaccumulation of
mercury in adults
Temperature
Increased winter
precipitation may increase
overwintering habitat
Change in
deposition rates
Permafrost thaw
Increased potential for establishment of invasive macrophytes and changing fire dynamics
Fire
Mining
Review
Please review and provide comments:
- Distribution models for fish and habitats
- Conceptual models and text descriptions for fish
Contact: Rebecca Shaftel
[email protected], 907-786-4965
Photo: USFWS