M. Siobhan Fennessy

AN EVALUATION OF RAPID METHODS FOR ASSESSING THE ECOLOGICAL CONDITION OF WETLANDS

We analyzed 40 existing wetland rapid assessment methods that were developed for a variety of purposes, including informing regulatory decisions and local land use planning, and reviewed them for their potential to assess ecological integrity or condition. Four evaluation criteria were used. We determined if the method 1) can be used to measure condition, 2) is truly rapid, 3) includes a site visit, and 4) can be verified. This resulted in six methods being selected for evaluation relative to a conceptual model describing the core elements of a wetland assessment method, including universal indicators of soil, hydrology, and biotic communities, as well as regional indicators. An additional nine methods were kept for ideas on indicators, scoring, or regionalization. From this review, we identified five general areas that need to be addressed when adapting existing methods or developing new methods to assess condition: 1) definition of the assessment area, 2) treatment of wetland type, 3) approaches to scoring, 4) consideration of highly valued wetland types or features, and 5) procedures for validation with comprehensive ecological data. With scoring in particular, we present the advantages of a method that produces a single integrative score. Development of a rapid assessment method can assist those interested in incorporating condition assessment into their programs because they require less time in the field and less taxonomic expertise than more quantitative methods, which can lead to significant cost savings and increased sample sizes.

Patterns of plant decomposition and nutrient cycling in natural and created wetlands

Functional assessment is important to determine whether restored and created wetlands are similar to natural ones. We investigated ecosystem processes (decomposition, biomass production) and some aspects of biogeochemical cycles (plant uptake of nitrogen and phosphorus, litter N immobilization) in a population of natural and created (mitigation) wetlands. Our goals were to quantify ecosystem processes and compare some biological and physical characteristics, in order to assess the relative performance of created wetlands. The biological and biogeochemical characteristics of the natural and created sites were substantially different. Decomposition rates for both in-situ and control litter and tissue nutrient concentrations were higher in the natural wetlands, with final decomposition rate constant values (k (d−1)) averaging 0.009 for natural and 0.006 for restored sites over approximately a one-year incubation period. Aboveground biomass production was also significantly higher in the natural sites, averaging 347 g m−2 compared to 209 g m−2. Concentrations of soil percent organic carbon, percent nitrogen, and plant available P (μgP g soil−1) were significantly higher in the natural sites. Lower soil nutrient content in the created wetlands appears to propagate through the system resulting in low tissue nutrient levels, less biomass accumulation, and slower rates of decomposition.

No‐net‐loss not met for nutrient function in freshwater marshes: recommendations for wetland mitigation policies

Wetlands provide many important services throughout the world, with an estimated economic value that, in comparison to other ecosystems, far exceeds their relatively small global extent. In recognition of their importance, both national and international regulations exist to protect the worlds remaining wetlands. Of growing interest is the “no-net-loss” policy which permits unavoidable destruction of wetlands if compensated by restoration of degraded wetlands or creation of new wetlands. The fundamental assumption of no-net-loss is that wetlands can be created which function equivalently to natural wetlands. One integral function that wetlands perform is cycling of carbon, nitrogen and phosphorus. Here we demonstrate that loss of this nutrient-related function is not being mitigated by creation or restoration of wetlands. We compare indicators of plant- and microbial-mediated functions, as well as abiotic (e.g., soil character, hydrology) and biotic (e.g., plant community composition) structure, between 10 created or restored and 5 natural freshwater depressional wetlands in central Ohio, USA. Nutrient stocks were generally smaller and transformations slower in created wetlands than in natural wetlands, with little development over time. Of particular concern were differences in C- and N-related function. Created wetlands stored 90% less C within litter and 80% less C within soil and processed 60% less N through denitrification, on average compared to natural wetlands. Our study suggests that subversion of natural wetlands into restored or created wetlands could have large-scale environmental consequences such as reduced capacity for nitrate removal and C sequestration.

Effects of Human Activity on the Processing of Nitrogen in Riparian Wetlands: Implications for Watershed Water Quality

Wetlands are critical ecosystems that make substantial contributions to ecosystem services. In this study, we asked how the delivery of an ecosystem service of interest (N processing such as denitrification and mineralization) is impacted by anthropogenic activity (as evidenced by land cover change). We identify relevant factors (hydrology, nitrogen, and carbon variables), select headwater wetland sites in Ohio and Pennsylvania USA to represent a gradient of anthropogenic disturbance as indicated by land cover characteristics (represented by the Land Development Index, or LDI), and determine if there are differences in the selected variables as a function of this gradient by categorizing sites into two groups representing high and low disturbance. We utilized Classification and Regression Trees (CART) to determine which variables best separated high from low disturbance sites, for each spatial scale at which land cover patterns were determined (100 m, 200 m, 1 km radius circles surrounding a site), and within each category of water quality variable (hydrology, nitrogen and carbon). Thresholds of LDI were determined via the CART analyses that separated sites into two general classes of high and low disturbance wetlands, with associated differences in Total Nitrogen, NH4+, Soil Accretion, C:N, Maximum Water Level, Minimum Water Level, and %Time in Upper 30 cm. Low Disturbance Sites represented forested settings, and exhibit relatively higher TN, lower NH4+, lower Soil Accretion, higher C:N, higher Maximum Water Level, shallower Minimum Water Level, and higher %Time in Upper 30 cm than the remaining sites. LDIs at 100 m and 200 m were best separated into groups of high and low disturbance sites by factors expected to be proximal or local in nature, while LDIs at 1000 m predicted factors that could be related to larger scale land cover patterns that are more distal in nature. We would expect a water quality process such as denitrification to be relatively lower in forested settings, due to the low available nitrogen (associated with high C:N) and constant and saturated conditions; conditions for maximum denitrification may be found in agricultural settings, where high nitrate groundwater can interact with surface soils through a wetting and drying pattern. The use of land cover patterns, as expressed by LDI, provided useful proxies for nitrogen, carbon, and hydrology characteristics related to provision of water quality services, and should be taken into account when creating, restoring, or managing these systems on a watershed scale.

Model application niche analysis: assessing the transferability and generalizability of ecological models

The use of models by ecologists and environmental managers, to inform environmental management and decision-making, has grown exponentially in the past 50 years. Due to logistical, economical, and theoretical benefits, model users frequently transfer preexisting models to new sites where data are scarce. Modelers have made significant progress in understanding how to improve model generalizability during model development. However, models are always imperfect representations of systems and are constrained by the contextual frameworks used during their development. Thus, model users need better ways to evaluate the possibility of unintentional misapplication when transferring models to new sites. We propose a method of describing a models application niche for use during the model selection process. Using this method, model users synthesize information from databases, past studies, and/or past model transfers to create model performance curves and heat maps. We demonstrated this method using an empirical model developed to predict the ecological condition of plant communities in riverine wetlands of the Appalachian Highland physiographic region, U.S.A. We assessed this models transferability and generalizability across (1) riverine wetlands in the contiguous U.S.A., (2) wetland types in the Appalachian Highland physiographic region, and (3) wetland types in the contiguous U.S.A. With this methodology and a discussion of its critical steps, we set the stage for further inquiries into the development of consistent and transparent practices for model selection when transferring a model.