Marc_Anderson_NEAFWA_Aquatic_Classification_Overview

Northeast Aquatic
Habitat Classification
and Mapping
Arlene Olivero and Mark Anderson
The Nature Conservancy Eastern Regional Office
Objective
Process
Results
Next Steps
Project Objective
Create a standardized, 13-state, aquatic habitat classification and mapping system
to provide a foundation for state and
regional conservation Northeast and Mid-
Atlantic.
Products
1. Northeastern Aquatic Habitat Classification System (NAHCS), a standardized aquatic habitat classification system
2. GIS dataset of aquatic habitat using the standardized aquatic habitat classification system
Financial support was provided through the Northeast Association of Fish and
Wildlife Agencies Northeast Regional Conservation Needs (RCN) Grant Program.
• Workgroup of representatives from all states and some federal partners (>30 participants)
• Compiled the existing aquatic classification systems
used by each state
• Held monthly topic-focused workgroup calls to:
– review potential classification variables,
– Evidence to support thresholds
– Reach consensus on an agreed upon regional taxonomy
• Created a stream reach GIS habitat dataset linked to regional taxonomy
Process
State Name Email Agency
ME Dave Halliwell [e-mail] ME Dept. of Environmental Protection
ME Dave Coutemach [e-mail] ME Dept. of Environmental Protection
ME Katherine Webster [e-mail] Dept. of Biological Sciences, UMO
ME Merry Gallagher [e-mail] ME Dept. of Inland Fisheries and Wildlife
ME Peter Vaux [e-mail] Mitchell Center for Env. & Watershed Research, UMO
NH Ben Nugent [e-mail] NH Fish and Game Commission, Dept. of Inland Fisheries
NH Mat Carpenter [e-mail] NH Fish and Game Commission, Dept. of Inland Fisheries
NH Brian Frappier [e-mail] Department of Natural Resources, UNH
VT Rich Langdon [e-mail] VT Fish and Wildlife Dept.
VT Steve Fiske [e-mail] VT Dept. of Env. Conservation, Biomonitoring Program
MA Todd Richards [e-mail] MA Division of Fisheries and Wildlife, Field Headquarters
MA Alicia Norris [e-mail] MA Division of Fisheries & Wildlife
MA Margaret Kearns [e-mail] MA Dept.of Fish and Game, Riverways Program
MA/NE Jeffrey Legros [e-mail] Northeast Instream Habitat Program, UMASS Amherst
MA Robert Brooks [e-mail] USDA Forest Service, Northern Research Unit NE-4251
CT Neal Hagstrom [e-mail] CT Dept. of Environmental Protection
NY Marcia Meixler [e-mail] Dept. of Natural Resources, Cornell University
NY Tracey Tomajer [e-mail] Division of Fish, Wildlife, & Marine Resources - NYSDEC
NJ Lisa Barno [e-mail] NJ Department of Environmental Protection
PA Mary Walsh [e-mail] PA Natural Heritage Program
PA Jeremy Deeds [e-mail] PA Natural Heritage Program
PA Mike Pruss [e-mail] PA Game Commission - State Wildlife Management Agency
PA Brian Chalfant [e-mail] PA Dept. of Environmental Protection
PA David Day [e-mail] PA Fish and Boat Commission
PA Michael Bialousz [e-mail] PA Fish and Boat Commission
DE Kevin Kalasz [e-mail] DE Division of Fish and Wildlife
MD Scott Stranko [e-mail] MD Dept. of Natural Resources
VA Dave Morton [e-mail] VA Dept. of Game and Inland Fisheries
VA Brian Roosa [e-mail] VA Dept. of Game and Inland Fisheries
WV Walter Kordek [e-mail] WV Division of Natural Resources
WV David Thorne [e-mail] WV Division of Natural Resources
WV Dan Cincotta [e-mail] WV Division of Natural Resources
MI Paul Seelbach [e-mail] MI Dept.of Natural Resources, University of Michigan
PA/DE Cara Campbell [e-mail] USGS Northern Appalachian Research Branch
NY James McKenna [e-mail] Unites States Geological Survey
MA/NE Ken Sprankle [e-mail] USFWS - Wildlife & Sport Fish Restoration Program, Region 5
MA/NE Willa Nehlsen [e-mail] U.S. Fish & Wildlife Service - Regional Fisheries Program
Workgroup
Participants
35+
State Fish and
Wildlife Agency,
DEP, Natural
Heritage Program,
Federal Agency,
University, NGO
Partners….
Initial Compilation of Existing Classifications
General Trends
• Very little consistency across states (>200 types)
• Used Different River Variables: stream size, coldwater/warmwater,
gradient, elevation/ecoregions, flow permanence, chemistry, or
indicator species/fish community/macroinvertebrate community
• Used Different Lake Variables: trophic status, depth, ph
Team agreed to develop a unambiguous
biophysically based standard taxonomy and map
for the region.
Size classes from state and regional
classifications using drainage area to
measure size
State Size Classes from classifications that used drainage area to measure size sq.mi.
NAHCS Stream Size Class 12345101520253035501001502005001000200040005000700090001000020000
WV WV: intermittent < 18 acres 0.02813 sqmi 1a
NY NY Rocky Headwater Stream: 5-10m across .005-.05 sq.mi.1aNY NY Marshy Headwater Stream: <3m across < 0.015 sq.mi 1a
PA PA Macroinvertebrate Physical Stream Type: 0-2 sq.mi.1aMA MA: <2 sq.mi. break for perennial vs. intermittent 2 sqmi.1a
VT
VT Biomonitoring: Small High Gradient Streams (SHG):10 square
kilometers 3.861 sqmi 1aPA PA Macroinvertebrate Physical Stream Type: 3-10 sq.mi.1a, 1b
WV WV: Small: ≤ 10,000 acres <15.63 sq.mi.1a, 1b
MA MA: <50 ft break for wadeable
<25 sq.mi. (from DA =
(w/14.7)^(2.632)1a, 1bME ME/TNC ERO: 0-29sq.mi.1a, 1b
NY NY GAP: 0-39 sq.mi.1a, 1b
PA PA Fish Community Type Synthesis Recommendations: 0-50 sq mi 1a, 1b, 2
VT VT Aquatic Communities: Brook Trout: mean 11 sqkm (3-30) 4.247 (1.158 - 11.58) sqmi 1b
VT
VT Aquatic Communities: Brook Trout - Slimy Sculpin mean 12 sqkm
(2-30)
4.633 (0.7722 - 11.58)
sqmi 1b
US USA B3: indicator: creek chub; homogeneous cluster:
eastern mean 4.247 sq.mi
(95% conf. int. 0-10 sq.mi.)1b
CT CT: <30ft break for wadeable
6.5 sq.mi. (from DA = (w/14.7)^(2.632)1b
VT VT Biomonitoring: Slow Winders:average 25 square kilometers 9.653 sqmi 1b
US USA B2: indicator: fathead minnow; least homogenous:
eastern mean 10.04 sq.mi. (95% conf. int. 0-45
sq.mi.)1b
US USA D2: highest elevations, steepest slopes, and coolest air; mountainous areas of the wester USA: indicator rainbow trout: eastern mean 12.74 sq.mi. (95% conf. int 0-28 sq.mi.1b
VT
VT Aquatic Communities: Brook Trout - Blacknose Dace: mean 41
sqkm (4-103)15.83 (1.544 - 39.77) sqmi 1b
US USA C3: streams of mountainous west, east, and north-central indicator: blacknose dace:
eastern mean 15.44 sqmi
(95% conf. int. 0-42 sq.mi.)1b
US USA C2: streams of mountainous west, east, and north-central: indicator brook trout, brown trout, mottled sculpin:
eastern mean 23.55 sq.mi.
(95% conf. int. 0-101 sq.mi.)1b
DE DE: 1-3rd order is break for wadeable
mean 2-23 sq.mi. (10-
90% percentile = .91-40)1b
VT
VT Biomonitoring: Medium High Gradient Streams (MHG):average 88
square kilometers 33.98 sqmi 1b
VT VT Aquatic Communities: Bluntnose Minnow - Creek Chub: mean 88 sqkm (2-515)33.98 (0.7722 - 198.8) sqmi 1b, 2
PA PA Macroinvertebrate Physical Stream Type: 11-100 sq.mi.1b, 2WV WV: Medium: 10,001-100,000 acres 15.63-156.3 sq.mi.1b, 2
NY NY Confined River: >30 sq.mi. < 1000 sq.mi.1b, 2, 3a
NY NY Unconfined River: >30 sq.mi. < 1000 sq.mi.1b, 2, 3aME ME/TNC ERO: 30-199 sq.mi.2
VT VT Aquatic Communities: Pumpkinseed - Bluntnose Minnow: mean 336 sqkm (8-728)129.7 (3.089 - 281.1) sqmi 2
VT
VT Aquatic Communities: Blacknose Dace -Common Shiner: mean 104
sqkm (10-298)40.15 (3.861 - 115.1) sqmi 2
US USA B1: indicator: common shiner and white sucker:
eastern mean 46.72 sqmi
(95% conf. int. 0-128 sq.mi)2
PA PA Fish Community Type Synthesis : 50-199 sq mi 2
PA PA Fish Community Type Synthesis :
50-199 sq mi or 50-100
sq.mi.2
US
USA C1: streams of mountainous west, east, and north-central: indicator
longnose dace:
eastern mean 67.57 sq.mi. (95% conf. int 0-193
sq.mi.)2
US USA A4: indicator: central stoneroller and striped shiner;:
eastern mean 160.6 sq.mi
(95% conf. int. 0-650 sq.mi.2
VT
VT Biomonitoring: Warm Water Moderate Gradient Streams and
Rivers:average 480 square kilometers 185.3 sqmi 2NY NY GAP: 39-1158 sq.mi.2, 3a
PA PA Fish Community Type Synthesis : 100-499sqmi.2, 3a
PA PA Macroinvertebrate Physical Stream Type: >100 sq.mi.
2, 3a, 3b,
4, 5
WV WV: Large: > 100,000 acres >156.3 sq.mi.2, 3a, 3b, 4, 5
PA PA Fish Community Type Synthesis : 200-749 sq mi 3aME ME/TNC ERO: 200-999 sq.mi.3a
US
USA A2:indicator species: bluegill; consists of many different warm
water species:
eastern mean 432.4 sqmi (95% conf. int. 0-1773
sq.mi.)3a
PA PA Fish Community Type Synthesis : 500-1999 sq.mi.3a, 3bPA PA Fish Community Type Synthesis : 750-2999 sq mi 3a, 3b
US
USA A3: indicator species spotfin shiner, shorthead readhorse, and
common carp; all river species:
eastern mean 2319 sqmi (95% conf. int. 0-
7083sq.mi.)3bME ME/TNC ERO: 1000-6999 sq.mi.3b, 4
NY NY GAP: >1158 sq.mi.3b, 4, 5
NY NY Deepwater River: > 965.3 sqmi.3a, 3b, 4, 5
PA PA Fish Community Type Synthesis : 3000-6499 sq mi 4PA PA Fish Community Type Synthesis : 6500+ sq mi or > 6000 4, 5ME ME/TNC ERO: 7000+ sqmi 4,5
defined class boundaries, or for field
Product was not intended to override state classifications, but
was meant to complement state classifications and provide a
means for looking at patterns across the region.
++++
=
Classification Approach
Workgroup
determined
most
important
classification
variables
and
thresholds
for our
region within
this Higgins
et al 2006.
framework.
Initial Classification Variables …
• Size - upstream drainage area, order, mean annual flow …
• Channel Gradient (slope)
• Geology and Chemistry – bedrock types, surficial types, PH,
ANC, channel substrate (cobble, gravel, sand…)
• Water Temperature
• Network Position
•Elevation
• Air Temperature
• Precipitation
• Baseflow and flow stability
• Watershed average slope, surrounding landforms
• Riparian wetlands and floodplains
RESULTS:
Four Key Habitat Variables
SIZE
GRADIENT
GEOLOGY
TEMPERATURE
Each variable had
internal classes
Size Class Description Definition (sq.mi.)
1a Headwaters 0<3.861
1b Creeks >=3.861<38.61
2 Small Rivers >= 38.61<200
3a Medium Tributary Rivers >=200<1000
3b Medium Mainstem Rivers >=1000<3861 4 Large Rivers >=3861<96535Great Rivers >=9653
Gradient Class Description
Definition (slope of stream channel (m/m) * 100)1 Very Low Gradient <0.02%
2 Low Gradient >= 0.02 < 0.1%
3 Moderate-Low Gradient >= 0.1 < 0.5%
4 Moderate-High Gradient >=0.5 < 2%
5 High Gradient >=2 < 5%
6 Very High Gradient >5%
Geology Class Description
Definition (index based on cumulative upstream geology; only applied to size 1a, 1b and 2 rivers)
1 Low Buffered; Acidic 100-174
2 Moderately Buffered; Neutral 175-324
3 Highly Buffered; Calc-Neutral 325-400
Temperature Estimated Natural Temperature Regime Definition
1Cold2Transitional Cool
3 Transitional Warm4Warm
Complex rules; see CART analysis and final rules on Temperature Metadata worksheet
Size
Results: 7 agreed upon classes
1) Headwaters (0-3.9)
2) Creeks (3.9-39)
3) Small River (39<200)
4) Medium Tributary Rivers
(200<1000)
5) Medium Mainstem Rivers
(1000<3900)
6) Large Rivers
(>=3900)
7) Great Rivers
(>=9653)
( square miles of upstream
drainage area. Rounding 3.89
to 3.9)
Size: 7 classes
Streams
1) Headwaters
(0-3.9)
2) Creeks (3.9-
39)
Rivers
3) Small
(39<200)
4) Medium
Tributary
(200<1000)
5) Medium
Mainstem
(1000<3900)
6) Large
(>=3900)
7) Great
(>=9653)
In square miles of
upstream
drainage area.
Stream Size Classes
tested with Pennsylvanian Fisheries Data
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0.14
0.16
0.18
0.20
12345
Size class Options
Cl
a
s
s
i
f
i
c
a
t
i
o
n

S
t
r
e
n
g
t
h
Atlantic Basin Fish
Ohio-Great Lakes
Basins Fish
Option 1 (sq.mi): 1=0-29, 2=30-199, 3=200-999, 4=1000-6999, 5=7000+
Option 2 (sq.mi): 1=2-9, 2=10-49, 3=50-99, 4=100-499, 5=500-1999, 6=2000-5999, 7=6000+
Option 3 (sq.mi): 1=0-50, 2=50-199, 3=200-749, 4=750-2999, 5=3000-6499, 6=6500+
Option 4 (sq.mi): 1=0-9, 2=10-24, 3=25-199, 4=200-499, 5=500-1999, 6=2000-3999, 7=4000-6999, 8=7000+
Option 5 (sq.mi): 1=1-9, 2=10-29, 3=30-199, 4=200-999, 5=1000-6999, 6=7000+
Our scheme
(Option 1)
had high
classification
strength in both
basins
Gradient
Results: 6 classes
1) Very low
0-0.02%
2) Low
0.02-0.1%
3) Moderate
0.1-0.5%
4) Moderate-High
0.5-2%
5) High
2-5%
6) Very High
5%
Measure = rise/run *
100
Rare Stream Biota and Gradient Classes.
A cluster analysis using 6672 points representing 317 Heritage tracked species
of fish, mussels, snails, amphibians, and aquatic insects (317 species, 6672
point occurrences) was used to test gradient classes.
Geology: 3
classes
1)Acidic Geology,
low buffering
capacity (100-200*)
2)Neutral Geology,
moderate buffering
capacity (200-300*)
3)Calcareous
Geology, highly
buffered (300+*)
measure = Norton
buffering capacity
index value of
upstream geology in
the watershed
Average Stream pH by underlying geology type
**
Norton: 100 200 300 400
Temperature
Natural Expected
Water Temperature
Results: 4 classes
1) Cold
2) Transitional Cool
3) Transitional Warm
4) Warm
measure = based on
CART Statistical
model predicting class
based on size,
cumulative air
temperature,
stream gradient,
baseflow index.
Rivers : Temperature was estimated from air temperature models and
stream size class.
Streams: Used a slightly different model incorporating base flow
Air Temp Class,
with USGS PRISM
AIR TEMP
MODEL: exact ranges
of cumulative area
weighted mean annual
temp in detree C * 10
Size 5: >= 9653
sq.mi.
Size 4: 3861-
9653 sq.mi.
Size 3b:
1000-3861
sq.mi.
Size 3a:
200-1000
sq.mi.
size 2 200-
38 sq.mi.
baseflow
index <
40%
size 2 200-38
sq.mi.
baseflow
index >= 40%
2: 15-30 transitional cool
transitional
cool
transitional
cool cold cold cold
3: 30-45 transitional cool
transitional
cool
transitional
cool cold cold cold
4: 45-60
transitional
warm
transitional
cool
transitional
cool
transitional
cool
transitional
cool cold
5: 60-76
transitional
warm
transitional
warm
transitional
warm
transitional
cool
transitional
cool transitional cool
6: 75-90
transitional
warm
transitional
warm
transitional
warm pink
transitional
cool transitional cool
7: 90-105
transitional
warm
transitional
warm
transitional
warm
transitional
warm
transitional
w arm transitional cool
8: 105-120
transitional
warm
transitional
warm
transitional
warm
transitional
warm
transitional
warm pink
9: 120-135 very warm very warm
transitional
warm
transitional
warm
transitional
warm pink
10: 135-150 very warm very warm very warm very warm very warm very warm
RIVER MODEL: Definition of Water Temperature Classes
Regional Taxonomy
Combine 4 Key
Habitat Variables
259 types occur,
208 with more than 10km of length
From
Very high gradient, acidic, cold
Headwater stream
(1a_6_1_1)
To
Very low gradient, calcareous,
warm Great river
(5_1_3_4)
Size Class Description Definition (sq.mi.)
1a Headwaters 0<3.861
1b Creeks >=3.861<38.61
2 Small Rivers >= 38.61<200
3a Medium Tributary Rivers >=200<1000
3b Medium Mainstem Rivers >=1000<3861
4 Large Rivers >=3861<9653
5 Great Rivers >=9653
Gradient Class Description
Definition (slope of stream
channel (m/m) * 100)
1 Very Low Gradient <0.02%
2 Low Gradient >= 0.02 < 0.1%
3 Moderate-Low Gradient >= 0.1 < 0.5%
4 Moderate-High Gradient >=0.5 < 2%
5 High Gradient >=2 < 5%
6 Very High Gradient >5%
Geology Class Description
Definition (index based on
cumulative upstream
geology; only applied to size
1a, 1b and 2 rivers)
1 Low Buffered; Acidic 100-174
2 Moderately Buffered; Neutral 175-324
3 Highly Buffered; Calc-Neutral 325-400
Temperature Estimated Natural Temperature Regime Definition
1Cold
2 Transitional Cool
3 Transitional Warm
4Warm
Complex rules; see CART
analysis and final rules on
Temperature Metadata
worksheetSize, Gradient, Geo, Temp
Example: High gradient, acidic coldwater creeks (1b_5,6_1_1)
Highly Buffered Headwaters/CreeksExample: Highly buffered
calcareous headwater creeks
(1b_5,6_1_1)
Additional Habitat Descriptors
Distributed for Each Stream…
100+ other “habitat descriptors” variables
-drainage area,
-stream order
-channel elevation
-channel slope
-estimated mean-annual flow
-estimated mean-annual flow velocity
-PRISM model air temperature (local and cumulative upstream)
-PRISM model precipitation (local and cumulative upstream)
-local and cumulative areas of NLCD92 land cover types, etc.)
-area and % of local and cumulative geology types
-average baseflow index
-average catchment slope
-area of different landform types in local catchment
-upstream and downstream connectivity class (e.g. is the reach upstream a
lake, is the reach downstream a very large river etc.),
Streams on the edge of
transitional cool vs.
transitional warm
Use detailed baseflow and cumulative upstream air temp
query certain transitional stream types
Also included Large Scale
Descriptors
for each Stream
• Freshwater Ecoregion
• Ecological Drainage Unit
• HUC8
• Terrestrial Ecoregion
• State
Simplified Set
of types
this set retains the most
important variation in each
class while hiding some of
the detail of the full set.
This example simplification
used
• 4 size classes
• 4 gradient classes
• 3 geology classes
• 3 temperature classes
This simplified map groups
them into 92 types.
Three Simplification Methods
Mapping all 259 types is not always practical, nor necessary for certain
regional applications. Team developed 3 recommendations regarding
simplifying the classification.
1. Variable Prioritization
2. Within Variable Collapsing
3. Remove Biotically Insignificant Combinations
Allows for the combination of primary variables in
simpler forms to create a flexible and most desired
reporting scheme for your purpose
Download report from http://rcngrants.org/node/38 and
GIS data from…
http://rcngrants.org/spatialData
Anticipated Uses
• Provides common definitions and mapping of aquatic habitats across state lines.
• Facilitate a new understanding of aquatic biota on a regional scale
• Create a new opportunity to assess condition and prioritize habitats
• Facilitate more effective and efficient habitat conservation
Thank You
• Work with states to use the classification. Link types to biota. Stratifying biological and temperature sampling
Contact: [e-mail]