2D PHYSICAL HABITAT ANALYSIS LYR M&E Plan Spatial Structure CAN YOU SEE ME? • A population’s spatial structure encompasses its geographic distribution and the processes that generate or affect that distribution. • A population’s spatial structure depends fundamentally on habitat quality, spatial configuration, and dynamics as well as the dispersal characteristics of individuals in the population. p. 1
2D PHYSICAL HABITAT ANALYSIS Physical Habitat • Location with measurable, characteristic attributes where organisms perform a designated ecological function. • Attributes stem from interaction among hydrology, hydraulics, and geomorphology • Depth, velocity, substrate, temperature, cover are the most common attributes used. • Microhabitat- point-scale locations • Mesohabitat- patches at the 0.1-10 channel-width scale p. 2
2D PHYSICAL HABITAT ANALYSIS Physical Habitat Performance Indicators • [Species X lifestage #] use most, but not all, of their suitable microhabitat. • Sufficient quality, number, size and distribution of mesohabitats, and migration corridors between mesohabitats, exist for [Species X lifestage #] to achieve designated ecological functions. • Sufficient maintenance of watershed processes and regulatory management practices to create and maintain suitable physical habitat for all freshwater lifestages of relevant species. p. 3
2D PHYSICAL HABITAT ANALYSIS Steps to Characterize Physical Habitat 1. Make biological observations 2. Determine/quantify habitat suitability needs 3. Map habitat using 2D model results 4. Bioverify results 5. Analyze habitat statistics and geospatial structure p. 4
2D PHYSICAL HABITAT ANALYSIS LYR Chinook Redds Surveys • Weekly census of whole LYR by boat and wading. • Record submeter geographic locations and redd attributes • 2009-2010: 3180 in 622 cfs study area • 2010-2011: 3025 redds in 700 cfs study area Yellow dots are observations p. 5
2D PHYSICAL HABITAT ANALYSIS LYR Hydraulic Habitat Suitability Curves (HSCs) For Chinook Adult Spawning • Graphical representation of suitability of physical condition • May be data driven or theoretical ideals. p. 6
2D PHYSICAL HABITAT ANALYSIS LYR Substrate Visual Classification p. 7
2D PHYSICAL HABITAT ANALYSIS LYR 2010 Mean Grain Size Map p. 8
2D PHYSICAL HABITAT ANALYSIS LYR Substrate HSCs Developed by RMT • Tested 7 different ways of using RMT’s substrate data to obtain a substrate HSC. See report. “SHSC S5c” was selected by RMT 32-195 mm and relicensing stakeholders as best performer. p. 9
2D PHYSICAL HABITAT ANALYSIS 2D model of ~36 km of LYR Validation spans 530-5010 cfs spanned (0.1-1 Qbf) Velocity Magnitude Velocity Direction • High R 2 = 0.787 ✔ • Very high R 2 = 0.895 ✔ • Median error 16% • Median error 3.8% ✔ ✔ p. 10
2D PHYSICAL HABITAT ANALYSIS Habitat Suitability Index (HSI) The HSI is the value from 0-1 of habitat suitability that comes from applying an HSC to a single point in a river with specified physical properties. For any specific HSI, whether it is for an individual variable or an equation using several variables, it is possible to establish qualitatively meaningful thresholds that delineate the ecological functionality of the physical habitat A simple uniform scheme for binning HSI values into groups is as follows: Bin range Habitat Class Map Color • HSI = 0 non-habitat white or grey • 0 < HSI < 0.2 very poor quality habitat red • 0.2 < HSI <0.4 low quality habitat yellow • 0.4 < HSI <0.6 medium quality habitat green • 0.6 < HSI < 0.8 med-high quality habitat teal • 0.8 < HSI < 1.0 high quality habitat blue p. 11
2D PHYSICAL HABITAT ANALYSIS Combined Habitat Suitability Index (CHSI) Predicts habitat suitability using 2D model predicted depth and velocity as well as weighted mean substrate size. DHSI= depth habitat suitability index VHSI= velocity habitat suitability index Geometric Mean Approach for Hydraulic HSI (HHSI): HHSI = (DHSI * VHSI) ½ Combined HSI=0 if substrate HSI=0, otherwise CHSI=HHSI p. 12
2D PHYSICAL HABITAT ANALYSIS Bioverification Concept • “Bioverification” is a test of the combined predictions that results from coupling 2D model results with HSCs. • A bioverified model yields reasonable predictions of habitat availability, which may then be used in spatial and statistical analyses, such as assessment of habitat areas as a function of discharge. • Bioverification tests – CHSI Difference test (CHSIoccupied – CHSIavailable) – Mann-Whitney U Test of CHSIoccupied vs CHSIavailable – Transferability Test – Forage Ratio Electivity Index Tests (both HHSI and CHSI) p. 13
2D PHYSICAL HABITAT ANALYSIS CHSI Difference Test Results • All three sets of HSCs had a difference between utilized and non- utilized CHSIs for both redds surveys. • TRTAC HSCs performed best for both discharges, because they are so strict in their velocities that much of the river has a CHSI of 0. • USFWS HSCs performed worst, because they are so generous with considering many areas to be spawning habitat, whether they are ever used or not. p. 14
2D PHYSICAL HABITAT ANALYSIS Forage Ratio As An Electivity Index (EI) • The forage ratio is the ratio of % occurrence or % utilization (%U i ) to % available area (%A i ) in each habitat quality class “i” (e.g. low, medium, or high quality habitat) ⋅ # redds bed area % U = × = 100 × = i i i U A EI % 100 % ⋅ ⋅ ⋅ i i total # redds total area % A i • The forage ratio has been widely used in ecology with these constraints: – If %A i < 1% it is excluded from EI analysis – With %A i > 1%, the maximum EI is 100 p. 15
2D PHYSICAL HABITAT ANALYSIS EI Example Habitat # Stars % stars % area EI blue 18 72 35 2.06 green 4 16 15 1.07 yellow 2 8 20 0.40 red 1 4 15 0.27 white 0 0 15 0.00 p. 16
2D PHYSICAL HABITAT ANALYSIS EI Interpretation • EI = 1 : Random distribution—% occurrence is exactly proportional to % available area • EI = 0.5 : % occurrence is 50% less than what is expected with random occurrence • EI = 1.5 : % occurrence is 50% greater than what is expected with random occurrence Statistically Significant EI values for limited sized datasets • EI > 1+2*SD indicates preference of habitat class i • 1-2*SD < EI< 1+2*SD indicates tolerance of habitat class i • EI < 1-2*SD indicates avoidance of habitat class i • All areas whose HSI value is in a preferred habitat class may be quantified as “available habitat” in spatial and statistical habitat analysis, such as assessment of amount of suitable habitat area as a function of discharge. p. 17
2D PHYSICAL HABITAT ANALYSIS “Bootstrap” Test for Statistical Significance GOAL: Given limited data, determine the 95% confidence threshold for preference and avoidance. Test 10 sets of randomly distributed points with these steps: • Use ET Geowizards to create random pts with = # real observed pts • Calculate HHSI at all pts in random and real sets • EI analysis on each sets (both random and real) • Calculate mean and SD of EI among 10 random sets in each habitat quality class bin • Calculate preference (1+2SD) and avoidance (1-2SD) thresholds for each bin • Calculate the area weighted average of preference and avoidance thresholds across all bins to yield uniform thresholds for all bins p. 18
2D PHYSICAL HABITAT ANALYSIS Bioverification Performance Indicator 1 • A pairing of a 2D model with HSCs must yield one or more habitat classes with EI>1+2SD and one or more with EI<1- 2SD. This indicates that it is predicting both preference and tolerance. • Must take a risk to have specificity! Trivial Prediction! EI>1.5 Risky Prediction! EI<0.5 p. 19
2D PHYSICAL HABITAT ANALYSIS Bioverification Performance Indicator 2 • Among preferred habitat quality classes, EI values must not increase as assumed quality decreases. If 0.6-0.8 is preferred but Violates HSC 0.8-1 is not, this is not bioverified. 1.5 Consistent with HSC 0.5 0.2-0.4 0.4-0.6 0.8-1.0 0-0.0001 0.0001-0.2 0.6-0.8 p. 20
2D PHYSICAL HABITAT ANALYSIS LYR Chinook Bioverification FR Result CDGF p. 21
2D PHYSICAL HABITAT ANALYSIS LYR Chinook Bioverification FR Result CDGF p. 22
2D PHYSICAL HABITAT ANALYSIS Example CHSI Map • Report has map of entire river for your use. • Observed redds are Black extremely concentrated dots are where CHSI >0.6 observed redds • Observed redds strongly avoid areas where CHSI <0.2 p. 23
2D PHYSICAL HABITAT ANALYSIS LYR Chinook Bioverification Summary CDFG HSC + 2D Model + S5c substrate HSC was best • correctly identified 76.5% of redds in 2009-2010 survey • correctly identified 69.3% of redds in 2010-2011 survey • Of the ones it got wrong, 1/3 were within 3’ and 1/2 were within 5’ of preferred habitat. • ~11-15% of individuals whose microhabitat behavior was indistinguishable from random chance • If you made a bet where you won $100 per redd in predicted preferred habitat and lost $100 per redd in predicted avoided or tolerated habitat, you would net $284,600 over two years! p. 24
2D PHYSICAL HABITAT ANALYSIS Weighted Usable Area (WUA) Curve WUA = ∑ (HSI i x Pixel Area) Calculate WUA for each discharge (Q) p. 25
Recommend
More recommend
