Comparing Apples Oranges and Pineapples Brook trout data

  • Slides: 71
Download presentation
Comparing Apples, Oranges and Pineapples: Brook trout data; Lessons learned from the Eastern Brook

Comparing Apples, Oranges and Pineapples: Brook trout data; Lessons learned from the Eastern Brook Trout Joint Venture Mark Hudy Senior Science Advisor – Fisheries USGS

Objectives

Objectives

Objectives 1. Brief History 2. Transition from subwatershed “currency” 3. Occupancy databases 4. GIS

Objectives 1. Brief History 2. Transition from subwatershed “currency” 3. Occupancy databases 4. GIS databases 5. Products - Scale paper - Climate resilience data - Climate web project - Genetic patch sampling USDA Forest Service Fish and Aquatic Ecology Unit

History

History

Case History: Eastern Brook Trout Joint Venture 1. Evaluate the distribution of brook trout

Case History: Eastern Brook Trout Joint Venture 1. Evaluate the distribution of brook trout for the 2005 EBTJV assessment. 2. Context: -lots of states -inconsistent fine scale data 3. Hudy et al. 2008 NAJFM 28: 1069 -1085

Currency: Subwatersheds • A good start • Not fine enough for many questions? •

Currency: Subwatersheds • A good start • Not fine enough for many questions? • Only short term option for some states USDA Forest Service Fish and Aquatic Ecology Unit

Transition from Subwatershed Currency

Transition from Subwatershed Currency

Assessment Scales Sub-basins (4 th HUC; 8 digit) 220 (avg size= 254, 172 ha)

Assessment Scales Sub-basins (4 th HUC; 8 digit) 220 (avg size= 254, 172 ha) Watersheds (5 th HUC; 10 digit) 1, 472 (avg size = 41, 201 ha) Subwatersheds (6 th HUC; 12 digit) 6, 833 (avg size = 8, 879 ha) Catchments (14 digit ? ) 243, 874 (avg size = 237 ha)

Case Study: The scale at which results are reported can bias impressions of the

Case Study: The scale at which results are reported can bias impressions of the true distribution.

Case Study: The scale at which results are reported can bias impressions of the

Case Study: The scale at which results are reported can bias impressions of the true distribution. Corollary lesson: “The same database will be used to support opposite opinions”!

Sub-basins (4 th HUC) 100%

Sub-basins (4 th HUC) 100%

Watersheds (5 th HUC) 76%

Watersheds (5 th HUC) 76%

Subwatersheds (6 th HUC) 33%

Subwatersheds (6 th HUC) 33%

Catchments 11%

Catchments 11%

Patch Metrics

Patch Metrics

Identification of Brook Trout “Patches” • “Patch”= a group of contiguous catchments occupied by

Identification of Brook Trout “Patches” • “Patch”= a group of contiguous catchments occupied by wild brook trout. • Patches not connected physically – Dams, warm water habitat, downstream invasive species • Assumed to be genetically isolated populations

Patches USDA Forest Service Fish and Aquatic Ecology Unit

Patches USDA Forest Service Fish and Aquatic Ecology Unit

USDA Forest Service Fish and Aquatic Ecology Unit

USDA Forest Service Fish and Aquatic Ecology Unit

Patch - “Populations” • Number of patches 2, 698 • Average size 1, 541

Patch - “Populations” • Number of patches 2, 698 • Average size 1, 541 ha • Median size 855 ha USDA Forest Service Fish and Aquatic Ecology Unit

Patch Metrics (based on occupancy sampling of catchments) Spatial Metrics A. # of patches

Patch Metrics (based on occupancy sampling of catchments) Spatial Metrics A. # of patches B. # of patches with increasing size/connectivity(addition al upstream and downstream catchments with brook trout) C. # of patches decreasing in size/connectivity( loss of catchments) D. Average patch size of the entire resource E. # of patches with allopatric or sympatric(with brown or rainbow) populations

Why Patches ? Current Population Estimates • Mark-Recapture • Depletion Removal • Problems: •

Why Patches ? Current Population Estimates • Mark-Recapture • Depletion Removal • Problems: • Not viable for large scale monitoring • Expense • Inability to detect trend (i. e. large coefficient of variation % 50 adults; % 121 YOY) • Expansion to entire stream

Occupancy Database

Occupancy Database

Occupancy: Brook, Brown and Rainbow Sub-basins (4 th HUC; 8 digit) 107 (avg size=

Occupancy: Brook, Brown and Rainbow Sub-basins (4 th HUC; 8 digit) 107 (avg size= 254, 172 ha) Watersheds (5 th HUC; 10 digit) 808 (avg size = 41, 201 ha) Subwatersheds (6 th HUC; 12 digit) 3, 804 (avg size = 8, 879 ha) Catchments (14 digit ? ) 132, 321, 688 (avg size = 237 ha)

Brook Trout Distribution: Sub-basin (4 th HUC) 78% of 107 subbasins “Brook trout are

Brook Trout Distribution: Sub-basin (4 th HUC) 78% of 107 subbasins “Brook trout are well distributed throughout their native range”.

Brook Trout Distribution: Watershed (5 th HUC) 52% of 808 watersheds “There have been

Brook Trout Distribution: Watershed (5 th HUC) 52% of 808 watersheds “There have been some losses of brook trout but they are still found in approximately half of their range”.

Brook Trout Distribution: Subwatershed (6 th HUC) 32 % of 3, 804 subwatersheds “Brook

Brook Trout Distribution: Subwatershed (6 th HUC) 32 % of 3, 804 subwatersheds “Brook trout have been extirpated from over 66% of their historic subwatersheds”.

Brook Trout Distribution: Catchments 14 % of 132, 321 catchments “Brook trout have been

Brook Trout Distribution: Catchments 14 % of 132, 321 catchments “Brook trout have been extirpated from 86% of their historic catchments”.

GIS Database

GIS Database

Multiple Metrics (85) • • Geology, soils NLCD 2001 NLCD 2006 Dams, road xing

Multiple Metrics (85) • • Geology, soils NLCD 2001 NLCD 2006 Dams, road xing • Pollution • Elevation, air temp. • Occupancy USDA Forest Service Fish and Aquatic Ecology Unit

Multiple Scales 1. 2. 3. 4. 5. 6. 7. Sub-basin (HUC 8) Watershed (HUC

Multiple Scales 1. 2. 3. 4. 5. 6. 7. Sub-basin (HUC 8) Watershed (HUC 10) Sub-watershed(HUC 12) Local Catchment – Corridor Network Catchment Corridor USDA Forest Service Fish and Aquatic Ecology Unit

Products

Products

Scale Paper

Scale Paper

CART Model : Extirpated (76%); Reduced (64%); Intact (79%) 1. 2. 3. 4. 5.

CART Model : Extirpated (76%); Reduced (64%); Intact (79%) 1. 2. 3. 4. 5. % Forest Deposition kg/ha % Agriculture Road Density km/km 2 % Forest Riparian

Core Metric: % Forest • Subwatershed threshold – 68% forested land 68% • Only

Core Metric: % Forest • Subwatershed threshold – 68% forested land 68% • Only 6% of Intact subwatersheds have less than 68% Total Forest. • 85% of Extirpated subwatersheds < 68% Total Forest 68%

Assessment Scales Sub-basins (4 th HUC; 8 digit) 220 (avg size= 254, 172 ha)

Assessment Scales Sub-basins (4 th HUC; 8 digit) 220 (avg size= 254, 172 ha) Watersheds (5 th HUC; 10 digit) 1, 472 (avg size = 41, 201 ha) Subwatersheds (6 th HUC; 12 digit) 6, 833 (avg size = 8, 879 ha) Catchments (14 digit ? ) 243, 874 (avg size = 237 ha)

Climate Resilience

Climate Resilience

Under various climate change scenarios, brook trout are predicted to be extirpated from parts

Under various climate change scenarios, brook trout are predicted to be extirpated from parts of the historic range: Chesapeake Bay • Flebbe et al. 2006. Spatial Modeling to Project Southern Appalachian Trout Distribution in a Warmer Climate. TAFS • Clark et al. 2001. Predicting Climate Change Effects on Appalachian Trout: Combining GIS and Individual –Based Modeling. Ecological Applications. • Meisner. 1990. Effect of Climatic Warming on the southern margins of the Native Brook Trout. CJAS

Secondary Data Used in Regional Models Air Temperature: – Mean annual maximum air temperature

Secondary Data Used in Regional Models Air Temperature: – Mean annual maximum air temperature (PRISM; 800 m grid) – Air Temperature from Elevation or Elevation/Latitude models – Models from local weather stations ( i. e. airports, NOAA) Water Temperature: – Usually no direct measurements – Assume a steady relationship between air and water temperatures (i. e. 1°C air temp rise = 0. 8°C water temp rise)

Sensitivity

Sensitivity

Sensitivity LE/HS HE/HS LE/LS HE/LS Exposure Figure 1. Conceptual vulnerability classification model for brook

Sensitivity LE/HS HE/HS LE/LS HE/LS Exposure Figure 1. Conceptual vulnerability classification model for brook trout populations; brook trout populations are classified into one of four quadrants based on direct measurements or model predictions of sensitivity and exposure. Low exposure, low sensitivity populations are most likely to persist under various climate change scenarios.

Subsample 100 reproducing brook trout patches in WV, VA and MD Cluster Metrics –

Subsample 100 reproducing brook trout patches in WV, VA and MD Cluster Metrics – – – Elevation Max air % ground water % Forest Solar gain

Directly Measure Exposure and Sensitivity with Paired Air and Water: Thermographs • 30 minute

Directly Measure Exposure and Sensitivity with Paired Air and Water: Thermographs • 30 minute intervals with focus on the critical summer periods (July-September 30) • Focus on daily maximum water (DMAXW) and daily maximum air (DMAXA) Onset HOBO Water Temp Pro v 2 • Operation Range: -20 to 70°C • Accuracy: 0. 2°C over 0° to 50°C • Resolution: 0. 02°C at 25°C

Predictive Metrics Exposure: 81. 1 % concordance Sensitivity: 75. 2% concordance Elevation (m) %

Predictive Metrics Exposure: 81. 1 % concordance Sensitivity: 75. 2% concordance Elevation (m) % Forest Corridor Watershed Area (km 2) Maximum Air (°C) Watershed Area (km 2) Solar_mean_corridor

Climate Metrics

Climate Metrics

Predictive Metrics Exposure: 81. 1 % concordance Sensitivity: 75. 2% concordance Elevation (m) %

Predictive Metrics Exposure: 81. 1 % concordance Sensitivity: 75. 2% concordance Elevation (m) % Forest Corridor Watershed Area (km 2) Maximum Air (°C) Watershed Area (km 2) Solar_mean_corridor

Solar Gain – 30 meter pixel – – – Topography Aspect Elevation Latitude Longitude

Solar Gain – 30 meter pixel – – – Topography Aspect Elevation Latitude Longitude Riparian cover

Solar radiation (upper quartile) Canopy cover (<70%)

Solar radiation (upper quartile) Canopy cover (<70%)

Genetic Patch Metrics

Genetic Patch Metrics

Ne Nb diversity USDA Forest Service Fish and Aquatic Ecology Unit

Ne Nb diversity USDA Forest Service Fish and Aquatic Ecology Unit

Genetic Monitoring Metrics 1. Nb = # of individual brook trout(regardless of age) contributing

Genetic Monitoring Metrics 1. Nb = # of individual brook trout(regardless of age) contributing to year class; -Not the data requirements of Ne (Whiteley et al. 2012) 2. Amount of genetic diversity within a patch can evaluate changes in relative abundance

Virginia Pilot Study Stream Name Patch Size (Ha) Adult N (95% CL) YOY N

Virginia Pilot Study Stream Name Patch Size (Ha) Adult N (95% CL) YOY N (95% CL) Nb (95%CL) Nb/N Fridley's Gap* (above Karl's dam) 590 1, 004(882 -1, 145) 459(188 -1, 789) Skidmore Fork (above Todd Lake) 993 268(231 -346) 47(42 -130) 28 (14 -93) 0. 10 Dry Run (above Dry Run Dam) 1217 83(78 -156) 117(86 -367) 5 (3 -9) 0. 06 Briery Branch (above Briery Branch Lake) 2438 366(296 -576) 236(139 -457) 26 (21 -33) 0. 07 Dry River (above Switzer lake) 3807 1, 982(1, 726 -2, 202) 1, 285(843 -2, 077) 67 (58 -77) 0. 03 Little River (above Hearthstone Lake) 4121 728(637 -873) 463(347 -663) 46 (40 -53) 0. 06 131 (95 -192) 0. 13

Virginia Pilot Study

Virginia Pilot Study

Interpreting the Nb Metric • Increasing Nb trends: positive response from improved habitat or

Interpreting the Nb Metric • Increasing Nb trends: positive response from improved habitat or increasing population. • Decreasing Nb trends: suggest habitat loss and decreasing populations.

Number of Sample Starting Locations 1

Number of Sample Starting Locations 1

Number of Sample Starting Locations 2

Number of Sample Starting Locations 2

Number of Sample Starting Locations 3

Number of Sample Starting Locations 3

Number of Sample Starting Locations 4

Number of Sample Starting Locations 4

Conclusions • For genetic monitoring headwater brook trout populations, we recommend: – Sample size

Conclusions • For genetic monitoring headwater brook trout populations, we recommend: – Sample size of at least 75 – 3 starting locations – Focus on Nb of age-0 cohort • • Works across brook trout populations with varying family structure Should be general outcome for monitoring Nb of headwater trout populations

Monitoring Design

Monitoring Design

Proposed Monitoring Design and Methods: VA example Cluster analysis to subsample existing 331 patches:

Proposed Monitoring Design and Methods: VA example Cluster analysis to subsample existing 331 patches: – Sentinel samples- (yearly trends) – Panel samples every 5 years (long-term trends) Example: – 125 sites from cluster analysis- • 25 are designated “sentinel” to be sampled yearly. • Additional 20 sites sampled yearly on a rotating panel (each site visited every 5 years) • Equals 45 sites monitored statewide per year.

Lessons Learned

Lessons Learned

Lessons Learned Observed

Lessons Learned Observed

Thanks to the Partners!

Thanks to the Partners!

USDA Forest Service Fish and Aquatic Ecology Unit

USDA Forest Service Fish and Aquatic Ecology Unit

Smith Creek: NFWF

Smith Creek: NFWF