Habitat Evaluation Procedures

1
Habitat Evaluation Procedures

• 1969-1976 an enlightened Congress passes
  conservation legislation
• Affecting management of fish wildlife resources
• NEPA (National Environmental Policy Act)
• ESA
• Forest Rangelands Renewable Resources Planning
  Act
• Federal Land Policy Management Act

2
Habitat Evaluation Procedures

• Stimulates federal state agencies to change
  management, thus
• simple, rapid, reliable methods to determine
  predict the species and habitats present on
  lands
• expand database for T/E, rare species
• Predict effects of various land use actions

3
Habitat Evaluation Procedures

• USFWS
• Habitat analysis models
• Goal Assess impacts at a community level
  (i.e., species representative of all
  habitats being studied)
• e.g., use guild of species?

4
Habitat Evaluation Procedures

• USFWS
• Habitat analysis models
• What is a model?
• Important points to consider relative to models?
• What variables should be measured and/or included
  in the model?

5
Habitat Evaluation Procedures
Three Categories of Techniques 1)
Single-species models a) simple correlation
models e.g., vegetation type-species
matrix Species habitat matrix
6
(No Transcript)
7
Habitat Evaluation Procedures
Three Categories of Techniques 1)
Single-species models b) statistical models
i.e., prediction of distribution and/or
abundance What types?
8
Carnivore Habitat Research at CMU Spatial Ecology

• Overlay hexagon grid onto landcover map
• Compare bobcat habitat attributes to population
  of hexagon core areas

9
Carnivore Habitat Research at CMU Spatial Ecology

• Landscape metrics include
• Composition
• (e.g., proportion cover type)
• Configuration
• (e.g., patch isolation, shape, adjacency)
• Connectivity
• (e.g., landscape permeability)

10
Carnivore Habitat Research at CMU Spatial Ecology

• Calculate and use Penrose distance to measure
  similarity between more bobcat non-bobcat
  hexagons
• Where
• population i represent core areas of
  radio-collared bobcats
• population j represents NLP hexagons
• p is the number of landscape variables evaluated
  
• µ is the landscape variable value
• k is each observation
• V is variance for each landscape variable
• after Manly (2005).

11
Penrose Model for Michigan Bobcats
Variable Mean Vector bobcat hexagons NLP hexagons
ag-openland 15.8 32.4
low forest 51.4 10.4
up forest 17.6 43.7
non-for wetland 8.6 2.3
stream 3.4 0.9
transportation 3.0 5.2
Low for core 27.6 3.6
Mean A per disjunct core 0.7 2.6
Dist ag 50.0 44.9
Dist up for 55.0 43.6
CV nonfor wet A 208.3 120.1
12
Carnivore Habitat Research at CMU Spatial Ecology

• Each hexagon in NLP then receives a Penrose
  Distance (PD) value
• Remap NLP using these hexagons
• Determine mean PD for bobcat-occupied hexagons

Preuss 2005
13
Habitat Evaluation Procedures
Three Categories of Techniques 1)
Single-species models b) statistical models
modern statistical modeling model
selection techniques e.g., logistic
regression Resource Selection Probability
Functions (RSF) RSPF for determining amount
dist. of favorable habitat
14
Habitat Evaluation Procedures
Logistic regression Y ß0 ß1X1 ß2X2 ß3X3
logit(p) Pr(Y 1 the explanatory variables
x) p p e logit(p) / 1 e logit(p)
15
Resource Selection Functions (RSF)

• Ciarniello et al. 2003
• Resource Selection Function Model for grizzly
  bear habitat
• landcover types, landscape greenness, dist to
  roads

16
Resource Selection Probability Functions (RSPF)

• Mladenoff et al. 1995
• Resource Selection Probability Function Model
  for gray wolf habitat
• road density

17
Predicted American Woodcock Abundance Map
18
Quantifying Habitat Use Resource Selection
Ratios
Need 1) Determine use (e.g., prop. Use) 2)
Determine availability (e.g., prop avail.)
Selection ratio for a given resource category
i wi prop use / prop avail. If wi
1 , lt 1, gt 1
19
Quantifying Habitat Use Resource Selection
Ratios
Selection ratio wi prop use / prop avail.
wi (Ui /U) / (Ai /A) Ui observations
in habitat type i U total observations
(n) Ai random points in habitat type i A
total of random points
20
Quantifying Habitat Use Resource Selection
Ratios
Look at Neu et al. (1974) moose data 117
observations of moose tracks within 4 different
vegetation habitat types
21
Quantifying Habitat Use Resource Selection
Ratios
Veg. Type Use Avail wi
Interior burn 25 0.340 (25/117)/0.340 0.628
Edge burn 22 0.101
Edge unburned 30 0.104
Interior unburned 40 0.455
Totals 117 1.000
22
Quantifying Habitat Use Resource Selection
Ratios
Veg. Type Use Avail wi
Interior burn 25 0.340 (25/117)/0.340 0.628
Edge burn 22 0.101 (22/117)/0.101 1.862
Edge unburned 30 0.104
Interior unburned 40 0.455
Totals 117 1.000
23
Quantifying Habitat Use Resource Selection
Ratios
Veg. Type Use Avail wi
Interior burn 25 0.340 (25/117)/0.340 0.628
Edge burn 22 0.101 (22/117)/0.101 1.862
Edge unburned 30 0.104 2.465
Interior unburned 40 0.455
Totals 117 1.000
24
Quantifying Habitat Use Resource Selection
Ratios
Veg. Type Use Avail wi
Interior burn 25 0.340 (25/117)/0.340 0.628
Edge burn 22 0.101 (22/117)/0.101 1.862
Edge unburned 30 0.104 2.465
Interior unburned 40 0.455 0.751
Totals 117 1.000
25
Quantifying Habitat Use Resource Selection
Ratios
Selection ratio Generally standardize wi to
0-1 scale for comparison among habitat types
std wi wi / S (wi)
26
Quantifying Habitat Use Resource Selection
Ratios
Veg. Type wi Std wi
Interior burn 0.628 0.628/5.706 0.110
Edge burn 1.862 1.862/5.706 0.326
Edge unburned 2.465 0.432
Interior unburned 0.751 0.132
Totals 5.706 1.000
27
Habitat Evaluation Procedures
Three Categories of Techniques 1)
Single-species models c) Habitat Suitability
Index (HSI) models
28
Habitat Suitability Index (HSI)
29
Habitat Suitability Index (HSI)

• Model (assess) habitat (physical biological
  attributes) for a wildlife species, e.g., USFWS
• Habitat Units (HU) (HSI) x (Area of available
  habitat)
• Ratio value of interest divided by std comparison
• HSI study area habitat conditions
• optimum habitat conditions

30
Habitat Suitability Index (HSI)

• Model (assess) habitat (physical biological
  attributes) for a wildlife species, e.g., USFWS
• HSI index value (units?) of how suitable
  habitat is
• 0 unsuitable 1 most suitable
• value assumed proportional to K

31
Habitat Suitability Index (HSI)

• include top environmental variables related to a
  species presence, distribution abundance

32
Habitat Suitability Index (HSI)

• List of Habitat Suitability Index (HSI) models
• http//el.erdc.usace.army.mil/emrrp/emris/emrishel
  p3/list_of_habitat_suitability_index_hsi_models_pa
  c.htm
• e.g., HSI for red-tailed hawk

33
Habitat Suitability Index (HSI)Red-tailed Hawk
34
Habitat Suitability Index (HSI)Red-tailed Hawk
35
Habitat Suitability Index (HSI)Red-tailed Hawk
36
Habitat Suitability Index (HSI)Red-tailed Hawk
37
Habitat Suitability Index (HSI)Red-tailed Hawk
38
Habitat Suitability Index (HSI)Red-tailed Hawk

• For Grassland
• Food Value HSI (V12 x V2 x V3)1/4
• For Deciduous Forest
• Food Value HSI (V4 x 0.6)
• Reproductive value HSI V5

39
Habitat Evaluation Procedures
Three Categories of Techniques 1)
Single-species models c) Habitat Capability
(HC) models - USFS - describe habitat
conditions associated with or necessary to
maintain different population levels of a
species ( compositions)
40
Habitat Evaluation Procedures
Three Categories of Techniques 1)
Single-species models c) Habitat Capability
(HC) models - uses weighted values based
on habitat capacity rates at each
successional stage of veg. for
reproduction, resting, and feeding
41
Habitat Evaluation Procedures
Three Categories of Techniques 1)
Single-species models c) Habitat Capability
(HC) models -
42
Habitat Evaluation Procedures
Three Categories of Techniques 1)
Single-species models c) Pattern Recognition
(PATREC) models - use conditional
probabilities to assess whether habitat is
suitable for a species - must know what
is suitable unsuitable habitat
43
Habitat Evaluation Procedures
Three Categories of Techniques 1)
Single-species models c) Pattern Recognition
(PATREC) models - use series of habitat
attributes - must know relation of attributes
to population density
44
PATREC Models
Expected Habitat Suitability (EHS) P(H)
x P (I/H) / P(H) x P (I/H) P (L) x P
(I/L) P(H) prop. high density habitat P
(I/H) prop. area has high population
potential P (L) prop. low density habitat P
(I/L) prop. area has low population
potential Low high population potential
identified from surveys
45
Habitat Evaluation Procedures
Three Categories of Techniques 1)
Multiple-species models a) Integrated Habitat
Inventory and Classification System
(IHICS) - BLM - system of data gathering,
classification, storage - no capacity
for predicting use or how change affects
species
46
Habitat Evaluation Procedures
Three Categories of Techniques 1)
Multiple-species models b) Life-form Model -
USFS -
47
Habitat Evaluation Procedures
Three Categories of Techniques 1)
Multiple-species models b) Community Guild
Models - can be used to estimate responses
of species to alteration of habitat - (like
Life-form model) clusters species with
similar habitat requirements for feeding
reproduction
48
Three Scales of Diversity
A B alpha (?) diversity within habitat C
beta (?) diversity among habitat D gamma (?)
diversity geographic scale
49
Alpha Gamma Species Diversity Indices

• Shannon-Wiener Index most used
• sensitive to change in status of rare species

H diversity of species (range 0-1) s of
species pi proportion of total sample belonging
to ith species
50
Alpha Gamma Species Diversity Indices

• Shannon-Wiener Index

51
(No Transcript)
52
Alpha Gamma Species Diversity Indices

• Simpson Index sensitive to changes in most
  abundant species

D diversity of species (range 0-1) s of
species pi proportion of total sample belonging
to ith species
53
Alpha Gamma Species Diversity Indices

• Simpson Index

54
(No Transcript)
55
Alpha Gamma Species Diversity Indices

• Species Evenness

Hmax maximum value of H ln(s)
56
Beta Species Diversity Indices

• Sorensens Coefficient of Community Similarity
  weights species in common

Ss coefficient of similarity (range 0-1) a
species common to both samples b species in
sample 1 c species in sample 2
57
Beta Species Diversity Indices

• Sorensens Coefficient of Community Similarity

Dissimilarity DS b c / 2a b c Or 1.0 -
Ss
58
Species Sample 1 Sample 2
1 1 1
2 1 0
3 1 1
4 0 0
5 1 1
6 0 0
7 0 0
8 1 0
9 1 1
10 0 0
11 1 1
12 0 0
59
Sorensens Coefficient

• Sample 1
• Total occurrences b 7
• joint occurrences a 5
• Sample 2
• Total occurrences c 5
• joint occurrences a 5

• 2a/(2abc)
• Ss 2 5 / 10 7 5 0.45 (45)
• Ds 1 0.45 0.55 (55)