Implementing the pilot Federal Marshbird Monitoring Program in Wisconsin

Transcription

1 Implementing the pilot Federal Marshbird Monitoring Program in Wisconsin Initial results and future plans Andy Paulios and Ryan Brady: WI-DNR Mark Seamans: USFWS Patuxent

2 SECRETIVE MARSHBIRDS

3 Not well-monitored Low detection rate, localized studies Data from BBS, MMP, etc. suggest declines First statewide effort in Wisconsin

4 Survey Objectives Working closely with USFWS Patuxent 1. Estimate population trends for marshbirds 2. Inform habitat management decisions 3. Provide status data for harvested species

5 Wisconsin Wetland Inventory Define marshbird habitat Randomly select survey sites

6 Generalized Random Tessellation Stratification (GRTS( GRTS) Maintains randomization Provides spatial balance across state Clusters survey points to increase efficiency Two stages: Primary Sampling Units = PSUs Secondary Sampling Units = SSUs

7 PSUs = Hexagons (40 km 2 )

8 Selected PSUs

9 SSUs = Survey Points

10 Points 400 m apart Public vs. private lands Accessibility Interior vs. edge Canoe surveys?

11 Conway Protocol

12 May 1 Time of Year June 15

13 Protocol Time of Year Southern 2/3 of WI = May 1 - June 15 Northern 1/3 of WI = May 15 June 30 Ideally 3 surveys per point (SSU) in a season Each survey in a 10-day window separated by 7 days; replicate surveys separated by 2 weeks

14 Protocol Time of Day Morning OR evening 30 min before sunrise to ~3 hrs after ~3 hrs before sunset to 30 min after 1.5 hours surrounding sunrise/sunset is most critical

15 Protocol At each point Single observer Broadcast/playback 5 minutes passive, then 1 minute per broadcast species ~10 minute total Sequence important

16 Primary Target Species Least Bittern Sora Virginia Rail Common Moorhen American Coot Pied-billed Grebe

17 Primary Target Species American Bittern King Rail Yellow Rail

18 Secondary Target Species

19 Sample Datasheets

20 Results from 2008 Hybrid model 3 field techs + ~30 birders and biologists Training workshop at Horicon Marsh ~325 points surveyed over ~55 routes statewide Most common: Sora, Virginia Rail, Amer Bittern Several King and Yellow Rails detected Data also collected on secondary target species

21 Target Species Detected Number of individual birds detected, by species, during 2008 Wisconsin Marsh Bird Survey. Counts include multiple individuals detected at a single survey point during a single visit. Period1 Period2 Period3 Points Surveyed Primary Target Species American Bittern American Coot Common Moorhen King Rail Least Bittern Pied-billed Grebe Sora Virginia Rail Yellow Rail 2 0 0

22 Secondary Species Secondary Species Period 1 Period 2 Period 3 Black Tern Forster s Tern Marsh Wren Swamp Sparrow Wilson s Snipe Yellow-headed Blackbird 0 3 0

23 SORA Top model indicated occupancy was : Negatively related to wetlands classified as K (b=-1.451, SE=0.416). Positively related to Wetland size point was located in (b=0.523( = , 3, SE= ) and Percent Wetland surrounding survey point (b=0.023,( SE=0.008). Detection probability decreased throughout survey (slightly from 1-2) 1

24 Virginia Rail VIRA occupancy was positively related to: size of wetland the survey point was in (b=0.007,( SE=0.003), the percent wetland surrounding the survey point (b=0.018,( SE=0.012), wetlands with modifier H (b=0.843, SE=0.522).

25 American Bittern Top model indicated occupancy was positively related to: Percent Wetland surrounding survey point (b=0.026,( SE=0.011) size of the Wetland the survey point was located in (b=0.583( = , 3, SE= ) negatively related to wetland classified as K (b=-1.095,( SE=0.509). Detection probability decreased over time (Period1 p = 0.61 SE=0.13; Period2 p=0.19 SE=0.06; Period3 p= = 0 SE=0

26 Influence of Hydrology Occupancy Occupancy H SE(H) K SE(K) SORA VIRA AMBI Occupancy was significantly higher on permanently inundated wetlands (H) than in seasonally inundated wetlands (K). We didn t t measure water levels in 2008, but these results suggest the importance of hydrology to these species.

27 Additional Observations Detection rates for hemi-marsh species or deep marsh species (grebes, least bittern, terns, gallinules) was lower than expected. Why is our randomized design not surveying the wetter wetlands? Detection rates in northern PSUs were much lower than in SE Glacial Plains. We may be able to monitor population trend of Wilson s s Snipe at the state level. (harvested species) What did the flooding do to our data in 2008? Effects on 2009?

28 Climate Change? Preliminary analysis suggests that some of the target species are sensitive to the size and wetness of wetlands. How will climate change impact these factors in our open, emergent marshes and sedge meadows? Many of our target marshbird species are short-distance migrants. Has their phenology changed in the last 50 years? Will it continue to change? How will their food supplies (aquatic veg, macroinverts, amphibians, etc.) fare in a changing climate?

29 In the Future Finalize a sample design in 2009 that we can use for the long-term. Determine what information we collect on habitat at the site level (UW-Chicago). Implications for how we manage wetlands cause/effect monitoring. Can we monitor birds off-road on private lands? Incorporate specific wetlands of interest (NWRs, reference areas). Targeted studies to address management questions Other states coming on board in 2009 (KY, ID, NY) and 2010 (FL). Adjust design to account for restoration.

30 How you can get involved Property manager feedback on design and access. Volunteers needed to conduct surveys in 2009! (and beyond) Attend training workshop: April at Horicon Wildlife Area. (free beer and pizza)

31 More Information birdroutes/marshbirds.htm WBCI bird monitoring coordination website. Get protocols, datasheets, maps of points. Sign up for routes. Keep up with results and volunteer chatter!

32 Acknowledgments WBCI USFWS Bird Studies Canada Citizen-based Monitoring Grant Region 3 Non-game Grant. USFWS Webless Grant.