Pilot wetland condition assessment of palustrine emergent

Preparing to load PDF file. please wait...

0 of 0
100%
Pilot wetland condition assessment of palustrine emergent

Transcript Of Pilot wetland condition assessment of palustrine emergent

Pilot wetland condition assessment of palustrine emergent marshes in the
Upper Hudson River watershed
New York Natural Heritage Program
A Partnership between The Nature Conservancy and the NYS Department of Environmental Conservation 625 Broadway, 5th Floor Albany, NY 12233-4757 (518) 402-8935, Fax (518) 402-8925 www.nynhp.org

Pilot wetland condition assessment of palustrine emergent marsh in the Upper Hudson River watershed
Feldmann, A.L., T.G. Howard, and E.A. Spencer
A report prepared by the
New York Natural Heritage Program
625 Broadway, 5th Floor Albany, NY 12233-4757
www.nynhp.org
for the New York State Department of Environmental Conservation
Division of Water
December 2012 Please cite this report as follows: Feldmann, A.L., T.G. Howard, and E.A. Spencer. 2012. Pilot
wetland condition assessment of palustrine emergent marsh in the Upper Hudson River watershed. A report prepared for the NYSDEC Division of Water by the New York Natural Heritage Program. Albany, NY.

Acknowledgements We would like to extend special thanks to our partners in NYSDEC for their support and collaboration, including Margaret A. Novak, J.R. Jacobson, Tim Post, and Tracey Tomajer. We would not have been able to access all of our field sites without permissions and logistical help from Tad Norton (NYSDEC L&F), Wayne Tripp (F&W Forestry Services, Inc.), and Todd Dunham (TNC). Thank you to Tony Olsen in the EPA’s Western Science Division office (Corvallis, Oregon) for providing statistical consultation and a spatially-balanced, stratified, random sample pool. Don Faber-Langendoen (NatureServe) was a generous consultant as we tailored his Level 1 Landscape Condition Assessment (LCA) and Level 2 Ecological Integrity Assessment (EIA) methods to the needs of the project; his willingness to join us in the field and provide feedback was exceptional. Joanna Lemly (CO NHP) From our own NYNHP staff, David J. Marston built the database tools that allowed us to calculate Level 3 metrics, Stephen M. Young managed the sampling calendar and served as field crew leader, and Gregory J. Edinger, Richard M. Ring, and Julie A. Lundgren were invaluable members of the field teams whose expertise informed our methodological refinements. Funding for this project was provided by the US EPA Region 2, as part of the Section 106 monitoring initiative grant awarded to New York State.
Cover photograph: NYW12-04, Benson Road Marsh. Photographer: Stephen M. Young

Table of Contents
I. Introduction ................................................................................................................................ 6 1. Background ............................................................................................................................ 6 2. Objectives .............................................................................................................................. 6
II. Methods ...................................................................................................................................... 8 1. Quality Assurance Plan........................................................................................................... 8 2. Methods Development ............................................................................................................ 8 3. Site Selection and Level 1 Landscape Analysis ..................................................................... 9 4. Field Sampling Methods ....................................................................................................... 13 A. Level 2 Rapid Assessment ............................................................................................... 14 B. Level 3 Vegetation Subplot Sampling ............................................................................. 15 5. Analytical Methods ............................................................................................................... 16
III. Results and Discussion ........................................................................................................... 17 IV. Methodological Refinements and Next Steps......................................................................... 25 V. Quality Assurance/Quality control (QA/QC) Results.............................................................. 28 Literature Cited ............................................................................................................................. 31 Appendix A. Sample field packet, including site map showing AA and buffer circles and
Level 2 datasheets. Appendix B. Sample EIA score calculator.

Table of Figures
Figure 1. NYSDEC DOW watersheds used to guide freshwater monitoring and assessment. ...... 7 Figure 2. Pool of all wetlands on conservation lands and connected to surface waters in the
Upper Hudson River watershed. ....................................................................................... 10 Figure 3. Our sample frame, palustrine emergent wetland polygons on conservation lands and
connected to surface waters in the Upper Hudson River basin. ....................................... 10 Figure 4. Results from the statewide Landscape Condition Model, depicting areas of high stress
in increasingly dark shades of gray. The Adirondack Park boundary is shown in blue and the Upper Hudson River watershed’s subbasins are shown in purple. ............................. 12 Figure 5. Palustrine emergent wetlands (red dots) selected for sampling in the Upper Hudson River watershed (outlined in blue) displayed over a view of land use-land cover. .......... 13 Figure 6. Results of Non-metric Multidimensional Scaling ordination displayed with site names and codes. SM=Sedge meadow, SEM= Shallow emergent marsh. .................................. 20 Figure 7. Linear regression solution for a plot of Level 1 LCA scores against L2 EIA scores (n=18)................................................................................................................................ 21 Figure 8. Linear regression solution for a plot of Level 1 LCA scores against Level 3 Mean C scores (n=14)..................................................................................................................... 22 Figure 9. Linear regression solution for a plot of Level 2 EIA scores against Level 3 Mean C scores (n=14)..................................................................................................................... 23 Figure 10. Linear regression solution for a plot of Level 1 LCA scores against Level 3 Simpson diversity index (D) scores (n=14). .................................................................................... 23 Figure 11. Linear regression solution for a plot of Level LCA scores against Level 3 ShannonWeiner diversity index (H’) scores (n=14). ...................................................................... 24 Figure 12. Linear regression solution for a plot of Level 2 EIA scores against Level 3 ShannonWeiner diversity index (H’) scores (n=14). ...................................................................... 24 Figure 13. Box plots for each site sampled at Level 3 showing mean and variance of dominant species seen in subplots. ................................................................................................... 30

I. INTRODUCTION
New York State does not currently conduct statewide comprehensive monitoring for wetland quality and condition, although the importance of wetlands and their health is well understood. Wetlands are critical in flood and storm water control; surface and groundwater exchange; erosion control; nutrient cycling and food chain support; recreation; and as habitat for fish, shellfish, other wildlife, and wetland plants. The catastrophic flood events over much of eastern New York in 2011 make clear the importance of our wetlands, especially in connection to surface waters. Rigorous, ongoing monitoring of the condition of these wetlands is therefore critical to the conservation of many of the state’s environmental resources, and especially for understanding baseline condition and change in wetland condition over time.
This wetland condition assessment effort establishes wetland surveys within the basins that are being sampled in the Division of Water’s Rotating Integrated Basin Studies (RIBS) network. One target wetland type (palustrine emergent marsh) was chosen to maximize our ability to calibrate metrics. In 2012, we targeted wetlands in the Upper Hudson River basin. Condition was assessed following the EPA’s three-tiered approach: Level 1, a remote landscape assessment; Level 2, a rapid field survey evaluating ecological integrity and stressors; and Level 3, a quantitative description of vegetation.
1. Background
The New York State Department of Environmental Conservation (NYSDEC) recognizes the vital importance of wetlands to the health of the New York’s environment through various Wetland Program projects, including ongoing mapping efforts and attempts to measure net gains or losses of wetlands. Despite this, “New York State has not yet integrated wetlands into existing surface water monitoring programs, nor undertaken efforts to monitor the biological, physical, and chemical integrity of wetlands” (New York State Department of Environmental Conservation 2010). This project complements the ongoing regulatory wetland mapping carried out by the Division of Fish, Wildlife, and Marine Resources by developing and enacting a protocol for evaluating the health and quality of the state’s wetlands. It also complements an effort currently underway for wetlands along the Lake Ontario shoreline by The New York Natural Heritage Program (EPA GLRI grant # GL00E00842-0), through the collection of consistent metrics across sites (e.g., Floristic Quality Index and presence, cover, and diversity of invasive species). This pilot study sampled different wetland types than those being sampled for the Lake Ontario project, and thus expanded the overall pool of wetlands being evaluated. Similarly, in the Adirondacks, the Adirondack Park Agency has complete wetland maps and is beginning a pilot citizen-science wetland monitoring program (Agreement: CD-972080-00).
2. Objectives
The objective of this pilot monitoring project is to assess how to apply the approach taken by the statewide ambient water quality program into wetlands, while building on the work in New York conducted as part of the EPA National Wetland Condition Assessment (NWCA). To do this, the New York Natural Heritage Program (NYNHP) piloted wetland condition assessment protocols that follow the EPA’s three-tiered approach. We explored whether the NWCA protocol was
6
New York Natural Heritage Program

appropriate as the basis for a state monitoring program, and we were informed by Level 1, 2, and 3 methods that have been developed nationally and locally (Mack 2001, U.S. Environmental Protection Agency 2006, 2011, Faber-Langendoen et al. 2009, Lemly et al. 2011, Walz and Domber 2011, Lemly and Gilligan 2012) to create a strategy for an ongoing wetland condition monitoring effort here in New York. We aimed to apply the broader approach of the statewide ambient water quality program into wetlands, and monitoring sites were selected within the framework of the rotating basins approach followed by the NYSDEC DOW’s flowing waters, lakes, and groundwater monitoring programs (Figure 1). Of the basins that the Division of Water sampled in 2012, in consultation with DOW and DEC’s Freshwater Wetlands Program in the Division of Fish, Wildlife and Marine Resources (DFWMR), we selected the Upper Hudson. We sampled during the middle of the growing season, in July and August.
Figure 1. NYSDEC DOW watersheds used to guide freshwater monitoring and assessment. Because our primary goal is to pilot assessment methods at Levels 1, 2, and 3, and because we had a limited field sample size (n=18 for Level 2, n=14 for level 3), we selected one wetland type to maximize the strength of our analyses. Both Ecological Integrity Assessment (EIA; Level 2) metrics and Floristic Quality Assessment (FQA; Level 3) indices are typically calibrated to natural community type (Andreas et al. 2004, Faber-Langendoen et al. 2012), so our approach of selecting sites of a common community type across a landscape condition gradient allows us to explore the range of variation within a type, maximize our sample size, and calibrate metrics for this type.
7 New York Natural Heritage Program

II. METHODS
1. Quality Assurance Plan
Following the EPA’s guidelines (U.S. Environmental Protection Agency 2012), we began by developing a detailed project plan to describe our approach to data collection, analysis, and storage in the context of specifying measurable standards for quality control throughout the project’s lifecycle (NYNHP and NYSDEC 2012). This Quality Assurance Project Plan (QAPP) was completed by NYNHP and approved by NYSDEC, an EPA-delegated state for quality assurance approval. We conducted periodic informal audits to ensure procedural compliance with the QAPP, and we calculated metrics to assess accuracy, precision, completeness, comparability, and representativeness. These findings are presented in the quality assurance and quality control (QA/QC) results section of this report.
2. Methods Development
Arguably our most important preliminary task was to review existing wetland condition assessment protocols with an eye towards selecting methods that could form the basis of a statewide monitoring program. Our intention was that this pilot project would allow us to determine whether our chosen methods were appropriate and what refinements would be needed as we moved forward. The NWCA methods followed the “three-level framework” (Fennessy et al. 2007) and we decided to follow suit, since, in addition to being a federal standard, this approach is being adopted by state programs (e.g., U.S. Environmental Protection Agency 2006, Lemly et al. 2011, Maine Wetland Interagency Team 2011). We next explored our options for data collection at each level, outlined here and discussed in detail in subsequent sections.
Our Level 1 method followed Comer and Hak’s (2012) Landscape Condition Assessment because it seemed like an elegant integration of relevant landscape-level stressors. We wanted to move beyond NWCA’s desktop evaluation of the buffer around each assessment area towards a semi-quantitative model of landscape condition.
The NWCA study included a Level 2 Rapid Assessment Method (USA-RAM), but because scoring for the metrics was still in development, we decided not to use it. Instead, we turned to another RAM, a wetland-specific Ecological Integrity Assessment (EIA) protocol developed by NatureServe for the EPA (Faber-Langendoen et al. 2012), with some modifications (California Wetlands Monitoring Workgroup (CWMW) 2012, Lemly and Gilligan 2012). Because the metrics and scoring for the EIA have been tested nationwide and have been repeatedly refined, we felt confident that the approach would be a solid starting point for us.
Level 3 data collection for the NWCA was quite intense; it required teams of four and encompassed algae and water chemistry (collecting and processing multiple samples), soils (digging numerous soil pits and bagging samples for lab analysis), and vegetation (documenting composition within nested plots). We decided to significantly scale that approach back to something that could be accomplished by a two-person team. Because we did not have the resources to process water or soil samples and because we have extensive vegetation sampling experience, we opted to focus our intense sampling on plant communities. We felt that the nested
8
New York Natural Heritage Program

plot sampling design would, again, be too time intense, so we followed a subplot approach that has been piloted in other states and that is derived from traditional relevé plot work (Peet et al. 1998, Walz and Domber 2011, Lemly and Gilligan 2012). Essentially, we collected detailed data in four subplots instead of sampling five plots with nested corners.
Analytical procedures for each level of sampling were available, which drove our selection process; those procedures are detailed in a subsequent section.
3. Site Selection and Level 1 Landscape Analysis
Prior to sampling, we determined our target population for study. We chose from naturallyoccurring vegetated wetlands classified as palustrine emergent, scrub-shrub, and forested in the Upper Hudson River watershed that are located on conservation lands and are ‘connected to’ (within 10 m of) surface waters. The sample frame consisted of Adirondack Park Agency (APA) and National Wetlands Inventory (NWI) polygons of the target type, with adjacent polygons of same Cowardin class merged. We set our minimum polygon size to 2 acres to accommodate an assessment area of 1.2 acres (a 40 m radius around a sample point).
Within the Upper Hudson watershed, wetland delineations have been developed by the Adirondack Park Agency (for the Northern Appalachian / Acadian ecoregion) and the U.S. Fish and Wildlife Service’s National Wetlands Inventory (for the Lower New England / Northern Piedmont ecoregion). In selecting one wetland type to focus on, we evaluated the distribution and abundance of all wetlands in the watershed that were connected to surface waters and located on protected lands. First, we selected those that intersected “conservation” lands (Kinal 2012), minimizing our need to gain formal access permission. To link our findings to those of the DOW’s monitoring program, we wanted to ensure that our sites had a hydrologic connection to surface water, so we discarded wetlands that were further than 10 m from the National Hydrography Dataset’s flowline and waterbody features. A total of 11,257 polygons met those initial criteria, and Figure 2 shows the distribution of this first set. Not surprisingly, the vast majority of sites were in the Northern Appalachian / Acadian ecoregion on protected lands within the boundary of the Adirondack Park.
We next merged adjacent polygons of the same Cowardin class and extracted three types for evaluation: palustrine emergent marsh (PEM), palustrine forested (PFO), and palustrine shrubscrub (PSS). We eliminated polygons that were less than 2 acres in size, deeming them too small to comfortably accommodate our 1.2 acre assessment area. We were left with 677 PEM polygons, 2102 PFOs, and 1616 PSS sites. In consultation with DOW and the freshwater wetlands program of DFWMR we mutually agreed that PEM was the most appropriate system to target for this pilot study. Our sample frame of 677 palustrine emergent wetlands is depicted in Figure 3.
9
New York Natural Heritage Program

Figure 2. Pool of all wetlands on conservation lands and connected to surface waters in the Upper Hudson River watershed.
Figure 3. Our sample frame, palustrine emergent wetland polygons on conservation lands and connected to surface waters in the Upper Hudson River basin. We wanted to use a landscape-scale (Level 1) parameter to stratify our sample frame before randomly selecting twenty point locations for field sampling. We developed a statewide, coarsescale Landscape Condition Assessment (LCA) that integrated anthropogenic stressors into overall landscape condition scores to provide an estimate of ecological stress. Our inputs included a variety of roads layers, from dirt roads to primary highways, electrical and gas transmission
10 New York Natural Heritage Program
WetlandsYorkPalustrineSurface WatersApproach