1. Groundwater-dependent wetlands in Western Montana Forests
Jen Chutz, DCI West Biological Consulting
Linda Vance, Montana Natural Heritage Program
Montana Wetland Council
March 26, 2014

2. Background EPA-funded project designed to…
Document extent of Groundwater Dependent Ecosystems (GDEs)
Especially fens, focusing on Western Montana
Enhance MTNHP’s wetland reference network
Improve recognition of patterns in imagery to assist in USFWS National Wetland Inventory (NWI) mapping
The Clean Water Act (CWA) requires states and tribes to monitor and report on the quality of waters within their states, including wetlands.
The principal objective of the CWA is to restore and maintain the chemical, physical, and biological integrity of the Nation’s waters.
Because we can evaluate these factors by assessing wetland condition and ecological integrity, the EPA has funded this project which is designed to…
MTNHP has established a network of reference standard wetlands covering multiple wetland ecosystems.

3. So how do you find GDEs? Expert knowledge
Many of the largest and richest fens have been identified and surveyed, and are known to botanists and land managers, or described in publications
Trial and error
In theory, in the course of enough wetland surveys, including getting to and from target wetlands, you’ll encounter GDEs
National Wetland Inventory (NWI) mapping and high-resolution imagery
Our preferred approach!
Before we could describe GDEs, we had to locate them. There are three main approaches to this.
[After expert knowledge]: the problem with relying on expert knowledge is that the GDEs most people remember or describe in publications are often the biggest, coolest, most diverse wetlands. The species-poor, smaller ones –which are part of the overall distribution pattern– tend to get overlooked.
[After Trial and Error] Because GDEs tend to be few and far between, it’s possible to miss them, especially when your survey design is probabilistic rather than targeted.

4. 3.21.2014: 1,726 USGS 24K Quads Mapped 2,178,028 Acres
Wetlands: 1,655,061 Acres
Riparian: 522,967 Acres
MTNHP has been mapping wetlands for the National Wetlands Inventory since 2007 as part of their overall wetland program. Photo interpreters digitize wetlands using USGS 7 ½ minute maps using 1m resolution aerial imagery.
(USGS 7 ½ minute maps or quadrangles are also called 24K maps because they are mapped at 1:24,000 scale).
Karen Newlon, who presented earlier today, can answer questions on both the mapping and assessment components of MTNHP’s program.
EXTRA INFORMATION (from MTNHP website): Wetland and riparian acres represent the area mapped and classified as either wetland or riparian based upon photointerpretation of 1-m resolution aerial imagery. These acres do not necessarily represent actual on-the-ground conditions. Factors such as photo quality, scale, and environmental conditions at the time of photo acquisition can affect mapping accuracy. Additionally, digital wetland maps are static and do not reflect the dynamic nature of wetlands that are subject to drastic annual and seasonal fluctuations in size and distribution. Mapping with high resolution imagery, as required by the Federal Geographic Data Center Wetlands Mapping Standard, allows for the identification of smaller wetland areas compared to historic wetland mapping created from imagery at a coarser scale. The resulting increase in the wetland acreage mapped is a result of the greater detail provided by using high resolution imagery and does not necessarily reflect an increase in on-the-ground wetland acreage. Moreover, wetland acres presented as a result of mapping cannot be directly compared with historic estimates of wetlands; as such estimates of wetland acres are often based upon original land survey data or extrapolated from hydric soil maps, providing a simplified characterization of historic wetland extent. 

5. Using the NWI to find GDEs/Fens
Because NWI maps are based on aerial imagery, key indicators like the presence of peat aren’t detectable
However, GDEs generally…
Have no inlet
Often have no outlet
Have a fairly distinct “signature” visible to the human eye because they are often saturated through the growing season
We use digital wetland mapping to identify and specifically target GDEs, not any kind randomization or stratification for this project.
This “signature” is both a familiar sight picture seen on the ground at the GDE and “bruising”, certain colors, and patterns seen on NWI imagery.

6. Using the NWI to find GDEs/Fens
Although GDEs found from floodplains to alpine areas in MT…
GDEs generally are found…
At low points in the landscape OR…
Near slopes where groundwater intercepts surface
Associated with glacial till/outwash, alluvial fans/basins, floodplains
Over limestone deposits
Many of the fens we found were located on, adjacent to, or at the base of slopes.
These low points or slopes are often associated with glacial till, glacial outwash, alluvial fans or basins, or floodplain landforms and are often underlaid by limestone (Chadde et al and Rocky Mtn. Subalpine-Montana Fen Ecological System)

7. Used NWI mapping in ArcGIS to identify ALL:
North of Whitefish, MT
Used NWI mapping in ArcGIS to identify ALL:
Palustrine Emergent wetlands AND…
Palustrine Shrub Scrub wetlands WITH…
Saturated water regimes
PEMB and PSSB
Selecting all wetlands with a “PEMB” or “PSSB” attribute can only be done in GIS, and requires that you have downloaded the NWI database from MTNHP or directly from USFWS.
The MTNHP Map Viewer’s NWI layer (found at MT Nat. Res. Info. System is like an image layer made for quick access/public information that you can’t manipulate, query for attributes, clip, etc... and therefore doesn’t allow you to select at this level of detail.
[Click to get red circle]. The red circle encloses a number of wetlands, some identified as PEMBs.
We were also curious about the one to the east of the large one [click to get yellow arrow] which is not coded as a PEMB or PSSB.
We were going to encounter this wetland anyways on the way from the road to the other 2 wetlands so we sent a crew here to check it out.

8. This is the large wetland in the center of the red circle
This is the large wetland in the center of the red circle. In the 2009 true color NAIP (Natl. Ag. Imagery Program is administered by USDA's Farm Service Agency), the wet areas appear as brownish-gray patches. Because the 2009 NAIP was flown in late summer, this indicates the wetland remained flooded. It has no inlet or outlet.
[Click for 2009 infrared] In the Infrared band, the wet areas show up distinctly, suggesting a persistent open water component. But you can see animal pathways through the center, indicating the water isn’t deep [click for arrow to illustrate this].
[Click to get 2011 NAIP True Color]. Here is the 2011 true color was a wetter year, and you can see this wetland was more flooded.
[Click to get 2011 World Imagery]. We also use ESRI’s World Imagery to inspect potential field sites. The NAIP is a 1m product; the World Imagery is 30cm, sometimes even 15cm.
This wetland, examined in all the imagery, looked like it might be a filled-in depression with an impermeable layer, with both a groundwater and a precipitation component. So it was selected as a target site.

9. Field surveys Rapid assessments (Level 2) ~ 2 hours
Intensive assessments (Level 3)
Up to 8 hours
We follow the U.S. EPA’s recommended Level framework for wetland assessment and monitoring.
It is a hierarchical three-tiered approach that involves increasing degrees of effort and cost with each increasing level of assessment, providing a flexible tool depending upon the availability of time, money, and staff.
Level 1 assessments are computer-based and typically require the least amount of time and money…I will not be discussing these assessments today.
Level 2 and Level 3 are both field-based assessments…MTNHP uses both rapid (Level 2) and intensive (Level 3) field protocols. Both protocols are based on identifying potential sites using digital imagery and both look at vegetation, hydrology, soils, and landscape context.
For either Level 2 or Level 3 assessments, we get pretty geared up for anywhere from a day trip to an 8 day wilderness pack trip, we deal with occasional inclement weather, stay away from the occasional spot fire, rock the head net to deal with every kind of biting insect, but all in all we see some beautiful landscapes and get to places few people get to go.

10. Field surveys Set up 0.5 hectare Assessment Area (AA)
Best represents site
Diversity & proportions
Usually 40m circle
Site info:
General wetland description
Landscape setting
Physical patch types
AA drawing
Photos & GPS points
We used printed maps, gazetteers, and GPS units to go to original GPS point created for us, then decide if that is appropriate or if AA center should be moved to capture both the diversity and its relative proportions represented in that wetland.
AA is usually a circle with 40m radius, but if need be, can be a rectangle or the entire small wetland with size restrictions.
IN THE PHOTO, we are sampling the GDE we were looking at a couple of slides ago
General wetland description & landscape setting – Written description of AA, amount of AA covered by water, estimated depth, etc…
Landscape setting – Written description of surrounding uplands, topographic position
Physical patch types (presence & %) – open pools, beaver dams, springs/seeps, plant hummocks, floating mat, etc…
AA drawing includes plant zones, inlets/outlets, plot and soil pit placement, any notable wetland features, etc…
We used 3 different classification systems:
1. Field Key to Wetland and Riparian Ecological Systems of Montana – developed by MTNHP??? – describes the habitat based on ecological setting at a landscape level
2. Hydrogeomorphic class – developed by Brinson (1993), adopted by USACoE, EPA, FHWA, NRCS, and USFWS – looks at hydrology and topographic position
3. Cowardin et al – can divide wetland into many Cowardin classes, but each code must be at least 20% of the AA (0.1 ha)
Classification of AA
Ecological Systems of MT
Hydrogeomorphic (HGM) class
Cowardin classification

11. Field surveys Hydrologic inputs/outlets If any Major vegetation zones
Structure and species composition
Anthropogenic disturbance metrics:
Distance to, degree of, cause of disturbance/changes to…
Landscape
Vegetation
GDEs sustained by inputs from local landscape
GDEs are stable, but not resilient
Soil
Hydrology
Walk entire wetland edge if possible to determine hydrologic inputs/outlets and landscape position.
This walk will help you as you determine veg zones. Each veg zone must represent no less than 5% of the AA.
Center zone: Potamogeton or Polygonum???; Next ring out: Carex atherodes; Next ring out: Carex utriculata (probably);
Outer ring/zone: Juncus balticus, Dasiphora fruticosa (PHOTO: From N. Fork of the Sun River on the Rocky Mtn Front west of Choteau)
Chadde et al. 1998: “…peatlands [and GDEs] are sustained by water and nutrient resources derived from much larger portions of the surrounding landscapes…
…Under naturally occurring hydrologic regimes, peatlands are relatively stable ecosystems. However, resiliency, or the ability to recover from disturbance, is low in peatlands…
…Recovery from major disruptions to water or nutrient flows or the removal of vegetation may require centuries….
…Rates of peat accumulation [are extremely slow, anywhere from <1 to]… 2 cm per century (added data from MT Field Guide to Rocky Mtn. Subalpine-Montane Fen).
So…Land-use activities that directly impact peatlands (such as peat mining, draining) and indirect impacts such as upslope timber harvest or road construction can cause changes to peatland biodiversity because many species are sensitive to minor changes in water chemistry and hydrology”.

12. Soil surveys
Min. 2 soil pits, each in different vegetation & hydrology, if possible
Min. 60 cm or until mineral soil
Soil Texture - Organic vs. Mineral
Peat, mucky peat, muck
Ribbon test…Sand  Loam  Clay Loam  Clay
Other hydric soil indicators
Redox concentrations and depletions (Fe & Mn)
Hydrogen Sulfide
Soil Color – Munsell Soil Color Chart
Peat ≥75% fibers (fibrist), mucky peat 16-74% fibers (hemist), muck ≤15% fibers (saprist) (NRCS Field Indicators of Hydric Soils 2010)
>40cm peat/MP/M = histosol (organic soil), 20-40cm peat/MP/M = histic epipedon
No ribbon: sand; <2.5cm: loam; 2.5-5cm: clay loam; >5cm: clay
“The hydraulic conductivity of highly decomposed peat is lower than finely textured clay soils” (Chadde et al. 1998).
Redox concentrations are the result of iron oxidation as groundwater levels fluctuate throughout the growing season. These concentrations are orange/reddish-brown (because of iron) and dark reddish-brown/black (because of manganese).
Redox depletions occur when soils are flooded and iron and manganese are reduced to their soluble forms. These soluble forms of iron and manganese can be leached out of the soil, leaving the natural color (gray or black) of the parent sand, silt, or clay (i.e., the matrix) behind.
Gleying is extreme redox depletion when iron is reduced and leaches from soil, leaving blue-gray or green-gray color

13. Level 3: Intensive assessments
Within AA, we lay out ten 10x10m modules
Within 4 selected modules, record…
Vegetation species…
Presence
Stratum
% canopy cover
Ground Cover…
Type
Deep water, gravel, litter, woody debris, etc…
% cover
Depth
If the vegetation is homogenous, intensive modules will be modules 3, 7, 9, and either 1 or 5, giving the broadest spread of modules across the vegetation plot.
If the vegetation is heterogeneous, the crew can decide which modules should be sampled to best capture the variability of the vegetation and associated microtopography and hydrology.
Stratum (height) and % cover are both measured in classes, i.e., trace, <1%, 1-2%, 2-5%, etc… and <0.5m, 0.5-1m, etc…

14. AA & Plot Placement
Image from Teresa Magee, US EPA Office of Research and Development, Corvallis, Oregon found in MT Ecological Integrity Assessment Field Manual 2013.

15. GDEs vs Fens Not all GDEs are fens, and not all fens are GDEs
Fens are defined as having an organic layer with ≥40cm of peat
GDEs rely on groundwater for the majority of their water input
But, we encountered many flow-through fens, some distinct channels
Must “dig” for indicators of groundwater dependency
Soil, pH, landscape, etc…
Fens take centuries to millenia to form, and are defined by a peat layer. “Peat accumulates at the rate of [<1 to 2cm per century], making peat a repository of 10,000 years of post-glacial history” (MT Field Guide to Rocky Mtn. Subalpine-Montane Fen”, Crum 1988).
As a rule, the main hydrologic input is groundwater, but we encountered a number of flow-through fens, some with distinct channels. In the field, both indicators of groundwater dependency and indicators of fen status require some serious digging. In our intensive assessments, we dug at least two, and often four, soil pits, using a shovel or auger, and measured each soil layer. We also measured the pH of soil and surface water, as species-rich fens are often alkaline.
We would look at where the wetland laid in the landscape (on a slope, in glacial till, base of alluvial fan, etc…)
Chadde et al. 1998:
Flow-through wetlands develops along streams and on gentle slopes and benches, requiring a continuous inflow of ground and surface water. Inflow provides nutrients, outflow washes away humic acids so reduces acidity.
“Fens are minerotrophic, receiving nutrients from water than has percolated through mineral soil and bedrock, or which has run off from uplands into a surface source such as a creek before entering the fen”. Bogs are not or are only slightly connected to ground water or surface water, both of which contain minerals.
pH goes from extremely acidic to alkaline, nutrient poor to rich, and sphagnum-rich to poor as go from (ombrotrophic) bog (domed peat means only nutrients come from direct precip., not the landscape), to poor fen(pH ), rich fen, and extremely rich fen (pH>7).

16. GDE 131
GDE 131, our earlier example, was classified as a wet meadow rather than a fen, although it was a close call.
Soil pits had from cm of peat, some with mucky peat underneath, some with a silty loam
Vegetation cover was primarily Carex utriculata, C. atherodes, and Calamagrostis canadensis
Aside for first point: curiously enough, the wetland to the east of this one, which was not classified as saturated in the NWI, did turn out to be a fen, with peat almost 100cm deep in one soil pit. Vegetation was similar, but with C. lasiocarpa dominant

17. One of the challenges in finding GDEs is that NAIP quality isn’t consistent. This is a 2009 image. The grey bruised-looking area is potentially a fen [click for circles around bruising].
[Click for true color] In the true color, that bruised area appears brown, which often indicates a brown moss or low shrub area, typically of peat-rich fens.
[click for 2011]. In the 2011 CIR, the whole area looks dry, even though it was a wet year. [click for true color]. But the brownish hue is still apparent.
[click for world imagery] The world imagery gives a suggestion of patterning.

18. And in fact, it was both a GDE and a fen, with a peat layer more than 70cm deep, and several fen obligate plant species like Carex livida and Menyanthes trifoliata [click for 2nd picture]

19. One of the things we learned in this project is that even peat-rich fens are not always species-rich.
[click for World Imagery] We picked this site for a survey because it appeared to have a fair amount of structural diversity.
[click for photo] While it was peat-rich with a strong non-vascular component, the dominant vascular species were carices, with a very low forb component [click for additional photo].
VEG: Betula (glandulosa?), Englemann spruce, Eriophorum (chamissonis?), moss, etc…

20. Project summary
In total from 2012 & 2013, we completed intensive surveys of…
131 GDEs
109 fens
19 wet meadows
3 emergent marshes
Identified 350 species in the Assessment Areas of these GDEs
Fens
AA vascular plant species richness varied widely
From 1 to 44 species identified per fen
Mean of 16.4 species

21. Project summary Nativity was high:
77 of the fens had no observed exotic species
Only one –in Glacier National Park– had more than 4
For fens, Mean Floristic Quality Index = 58
This was adjusted for cover weight and native species
Very near the value of 60.6 MTNHP found for reference-standard fens in an earlier study
Floristic Quality Index: assigns to plant species a rating that reflects the fundamental conservatism that the species exhibits for natural habitats. 
A native species that exhibits specific adaptations to a narrow spectrum of the environment is given a high rating (9-10). 
Conversely, an introduced, ubiquitous species that exhibits adaptation to a broad spectrum of environmental variables is given a low rating (0). 
Utilizing this method, a Floristic Quality Index (FQI) and Native Mean C are derived for a given area. 
The FQI is an indication of native vegetative quality for an area.
It is worthwhile to note that the values can be highly variable between different wetland ecological systems, for example, in Montana FQI indices were higher for the Great Plains Prairie Pothole, Rocky Mountain Alpine-Montane Wet Meadow, and Rocky Mountain Subalpine-Montane Fen ecological system, but generally low FQI indices in Western Great Plains Depressional Wetland and Western North American Emergent Marsh.
Fens had consistently higher values for all vegetation metrics except those associated with exotic species (Newlon & Vance 2011, A Reference Wetland Network for Assessment and Monitoring of Montana’s Herbaceous Wetlands).
VEG: Drosera rotundifolia

22. Next steps
Species of Concern (SOC) plant data will be entered into MTNHP databases once all IDs are confirmed
Final report on project will be available from MTNHP by late April
Database and GIS available on request from MTNHP
Contact: Linda Vance
Data forms and protocol are available on MTNHP’s website