Within the last decade, interest in vernal pools and their associated fauna has increased dramatically because of the rapid rate of loss of small wetlands to development and the well-publicized declines of amphibians (Alford and Richards 1999; Carey and others 1999; Sala 2000; Young and others 2001), a subset of which breed in vernal pools or other isolated wetlands (Pechmann and Wilbur 1994; Lannoo 1998). Amphibians and other wildlife species breeding in marshes, ponds, and rivers benefit from the regulatory attention generally afforded larger, permanent wetlands. Ephemeral and other isolated wetlands, and the adjacent terrestrial habitat used by wetland-dependent animals, have traditionally received little or no protection by wetland regulatory authorities at both the state and federal level (Fretwell and others 1996).

Vernal Pool Functions

Vernal pools provide biological, hydrologic, and ecosystem functions. In New England, vernal pools are specialized breeding habitat for certain amphibians and invertebrates. Common vernal pool-dependent species in New England include wood frogs (Rana sylvatica), mole salamanders (Ambystoma spp.), and fairy shrimp (Order Chirocephalidae). The term “vernal pool” generally refers to small, isolated, seasonal wetlands that provide the optimal breeding habitat for animals adapted to living in temporary, fishless pools. Periodic draw-down limits the establishment of significant resident populations of other predators (e.g., bullfrogs, green frogs, and predaceous insects) commonly found in larger, permanent wetland systems.

Other biological functions include refugia and foraging pools for non-breeding, wetland-dependent species (e.g., herptiles, waterfowl, wading birds, muskrat, and moose), for non-wetland-dependent species (e.g., bear, deer, shrew, skunks), and for sensitive or listed species (see Table 1) (Semlitsch and Bodie 1998; Gibbs 2000; Snodgrass and others 2000). Many wetland-dependent species exist in separate local populations sustained through occasional migrations which may be supported by a mosaic of isolated wetlands such as vernal pools (Hanski and Gilpin 1991; Semlitsch 1998; Gibbs 2000).

Hydrologic functions include surface water storage and subsurface and groundwater exchange. Ecosystem functions include biogeochemical cycling (see Wyman 1998) and energy exchange (through amphibian and invertebrate production and dispersal) to adjacent terrestrial ecosystems (Pough 1983; Butterwick 1996).

Regulatory Context

Vernal pools receive minimal regulatory protection at both federal and state levels because they are often small (generally less than 0.5 hectares; Brooks and Lyons 1998; Stone 1992; Calhoun and others 2003), and because many are hydrologically isolated wetlands. Dredge and fill activities in freshwater wetlands are regulated at the federal level through Section 404 of the Clean Water Act, and overseen by the United States Army Corps of Engineers (Corps). Section 404(a) grants the Corps authority to issue permits “for the discharge of dredged or fill material into the navigable waters at specified disposal sites.” Navigable waters include “streams (including intermittent streams), wetlands, sloughs, prairie potholes, wet meadows . . . the use, degradation, or destruction of which could affect interstate or foreign commerce.” (33 U.S.C. sec. 1362[7]). The major recent United States Supreme Court decision Solid Waste Agency of Northern Cook County (SWANCC) v. United States Army Corps of Engineers (Corps) (January 9, 2001), significantly limited the scope of federal jurisdiction over isolated intrastate waters under the Migratory Bird Rule. The Court rejected the Corps’ contention that “isolated ponds, some only seasonal, wholly located within two Illinois counties, fall under section 404(a)’s definition of ‘navigable waters’ because they serve as habitat for migratory birds.” In light of this decision, and in the absence of a documented endangered species nexus, it is probable that the Corps will no longer claim jurisdiction over isolated and intrastate waters, such as vernal pools, that are non-navigable.

Each New England state has a wetland protection statute that regulates activities in jurisdictional wetlands, as defined by the state. The specifics of the regulatory program and permit process vary from state to state (and even within states) (Preisser and others 2000), but small wetlands, including vernal pools, receive the least protection under most state regulatory programs. While the majority of New England states have no specific state regulations for protecting vernal pools, some states have amended their legislation to include regulation of smaller wetlands. For example, in Massachusetts, vernal pools receive regulatory review under the Wetlands Protection Act, provided that they are located within another category of regulated wetland and have been certified by the Massachusetts Division of Fisheries and Wildlife. Rhode Island enhanced their protection of vernal pools in 1994 by adding a new wetland category, special aquatic site, to their regulations, and the 1995 revisions to Maine’s Natural Resources Protection Act include vernal pools as Significant Wildlife Habitat, although this designation is still pending an administrative definition and requires individual pools to be mapped prior to designation.

The conservation of vernal pool breeding habitats is further complicated by the upland habitat needs of pool-breeding amphibians. The most overlooked aspect of vernal pool protection and creation is the need for protection of the surrounding upland, or critical terrestrial habitat, where pool-breeding amphibians forage and hibernate in the non-breeding season. For example, ambystomatid salamanders may travel an average of 145 m from breeding pools, while wood frogs may travel, on average, over 1000 m from pools (Berven and Grudzien 1990; Windmiller 1996; Semlitsch 1998).

Vernal pool creation, mandated by regulatory agencies as compensatory mitigation for lost wooded wetlands or vernal pools in larger wetland complexes, has become more common in the last decade (Mahaney, personal communication). Compensatory mitigation refers to the use of wetland restoration, enhancement, creation, and/or preservation to compensate for permitted wetland losses. For two decades of compensatory mitigation, thousands of hectares of wetlands have been restored or created in the United States. Given that for every 100 acres of wetland filled, 178 acres of wetland have been restored, enhanced, created, or preserved (National Research Council 2001), the overall health of our Nation’s wetlands depends on the quality of mitigated wetlands. The National Research Council identified vernal pools as a wetland type difficult to create. If vernal pools continue to be degraded or lost, and if creation attempts are not successful, we have cause for concern regarding the loss of function of isolated wetlands at both local and landscape scales.

The purpose of this paper is to evaluate the documentation on existing vernal pool creation attempts in New England in order to assess whether or not compensatory mitigation is addressing the replacement of vernal pool functions. We examined 15 vernal pool creation attempts in New England that resulted from federal regulatory action. Data on project documentation, including nature of the project and mitigation plans, were collected to determine if pool functions are being adequately addressed in creation attempts. Weaknesses in the mitigation projects we reviewed were noted and recommendations, based on current published literature, are provided for improving project design and monitoring. We recognize that mitigation strategies vary from state to state, and therefore our recommendations are general in nature.

Methods

Project Identification and Documentation

Visits to federal and state agencies in New England were conducted in 1999 and 2000 to inventory and summarize documentation on projects that included vernal pool creation attempts. The survey was limited to pools created as compensatory mitigation required by a Corps permit, or required by an United States Environmental Protection Agency (EPA) enforcement action. In order to identify projects meeting this specification, individuals at New England branches of federal regulatory agencies, including the Corps, the EPA, and the United States Fish and Wildlife Service (USFWS), were contacted to identify vernal pool creation attempts in the New England region. Following our initial survey, we provided state wetland regulatory agencies with a list of pool creation attempts that federal agencies had identified in their states and asked them to verify the list, provide any other data they might have, and to identify any cases that might have been overlooked by their federal counterparts.

We reviewed the vernal pool creation attempts in New England against the standard set by the National Wetlands Policy forum in 1990, who recommended that the federal government adopt the policy of “no overall net loss of values and functions of the nation’s remaining wetlands base” (National Wetlands Policy Forum, 1990).

Mitigation Plans

The nature of the mitigation was determined from project records and/or by interviewing project personnel. Parameters reviewed included number of pools created at each project site, type of wetland lost, functions intended to be replaced, project’s water-dependency, whether mitigation was on-site or off-site, total cost, and year constructed.

Mitigation plans, including pool design criteria, monitoring protocols, performance criteria, and remediation plans, were reviewed when available. We assessed completeness of regulatory files by looking for the presence of a final mitigation plan, as well as final reports for each year of monitoring.

Pool design criteria.

The following parameters were discussed in project reports regarding pool design: water regime (i.e., seasonal depth, duration, and timing of inundation and saturation), determination of pool depth and area, seeding/planting of vegetation, soil characteristics (i.e., source, depth), utilization of egg mass attachments, addition of woody debris, preservation of critical terrestrial habitat (i.e., adjacent upland), and transfer of eggs or adult amphibians to created pools. We summarized the percent of projects that addressed each of these parameters in their mitigation reports.

Monitoring protocols.

A review of monitoring protocols was similarly conducted. Monitoring variables that were considered by project managers and regulators included vegetation, depth, water quality, amphibians, target amphibian species, invertebrates, soils, and other wildlife. Frequency and total length of monitoring period was documented.

Performance criteria and remediation plans.

Presence of performance criteria and remediation plans was recorded. Performance criteria were defined as specified measures to be met in order for project success to be declared. Remediation plans were defined as measures intended to remedy project deficiencies identified during the monitoring period in order to increase the chance of meeting performance criteria.

Results

Project Identification and Documentation

Fifteen vernal pool creation projects resulting from federal action (representing a total of 36 created pools between 1991–2000) were identified by federal and state agencies: Maine (7), New Hampshire (3), Connecticut (3), Massachusetts (1), and Vermont (1) (Rhode Island reported having no vernal pools created as compensatory mitigation). All the projects impacted wooded wetlands; eight directly impacted a vernal pool or pond (Table 2). Almost half of the development impacts were transportation related (e.g., road construction), while the remaining projects were a result of commercial and residential development (e.g., commercial center, prison, subdivision, and golf course.) Only five of the 15 projects specifically identified “vernal pool creation” as the statement of functions to be replaced, although that was a required action for all 15 projects.

A memorandum of agreement between the Corps and the EPA allows permit issuance for only the least environmentally damaging practicable alternative (LEDPA) (USCORPS and USEPA 1989), yet this standard was not rigorously met for these projects. According to Section 230.10 of the Clean Water Act Section 404(b)(1) guidelines, there is the rebuttable presumption that “alternatives for non water-dependent activities that do not involve special aquatic sites are available.” However, of the 15 Corps-approved projects evaluated in this study, only two were for a water-dependent activity (cargo terminal and boat ramp). Cost information could not be determined or was not available for any of the 15 creation attempts, due in part to the fact that most were embedded within a larger wetland mitigation project.

Mitigation Plans

Various key pieces of documentation were missing from all project files. Mitigation plans were available for nine projects (60%) and missing or never created for five projects. One project compiled all available file documentation, which included regulatory and consultant letters and memos, into what became known as the “mitigation report” at project end.

Pool design criteria.

Pool design criteria were not well documented. For example, in one project, the regulator reported that “design specifications were simply a seat-of-the-pants idea that the consultant/landowner came up with.” No rationale was provided for design specifications for any of the pools. Parameters considered in pool design are summarized in Table 3.

Although improper hydrology is a common cause of wetland creation failure, only seven projects in this study reported an assessment of water regime (i.e., by use of test pits/monitoring wells) of the site prior to pool construction. Area and depth of the created pools varied considerably across projects. Nine projects (60%) had a pre-determined area for their pools which had a range of 436 ft2 to 56,000 ft2. Ten projects stated pre-determined depth limitations, ranging from a maximum depth of three feet to the stipulation that pools be 12–24 in deep from February to May.

Most projects addressed vegetation planting strategies, although there was considerable variation among approaches taken, ranging from planting/seeding of aquatic bed areas and around pools, to allowing the regeneration of native grasses from propagules of neighboring wetlands. Native species were preferred for plantings, however, cultivars were used as well.

Although 11 of the projects suggested source material or characteristics of soil to be used in pool basins, we were often not able to determine what was actually applied in final pool construction (e.g., one project reported “either 12 in clean, silty clay soil or 4–6 in topsoil with at least 12% organic content”). Considerable variation was found among planned soil types (e.g., topsoil, composted leaf mulch, loam) as well as substrate depth which had a range of 4–12 in.

Egg mass attachment sites and woody debris surrounding the pools were rarely provided. Several projects indicated that buffers would be preserved around the pools to provide vernal pools species with critical terrestrial habitat. However, there was great variation in what qualified as a buffer, ranging from a “550-foot preservation radius to accommodate post-breeding migrations of amphibians to adjacent uplands” to “buffer plantings will be installed in the parking lot, surrounding the detention basin and in an island at the front of the parking lot.” Four projects (27%) included the intentional transfer of vernal pool amphibian egg masses, larvae, or adults to the new pools, while another four included the idea within their remediation plans.

Monitoring protocols.

The availability of monitoring reports was scattered and inconsistent, ranging from a total lack of monitoring documentation to an occasional complete file. In general, stated monitoring periods were three to five years. It is not possible to summarize monitoring frequency, as even within projects, monitoring frequency varied with parameter being monitored. For example, in one project vegetation was to be monitored annually while hydrology would be monitored monthly. Furthermore, documentation on monitoring implementation was missing or incomplete.

Vegetation, water depth, and soil characteristics were monitored for a majority of projects (Table 3). Level of monitoring however, was quite variable. For example, some projects assessed only survival of plantings, while others determined species composition, percent cover, hydrophytic dominance, and presence of invasive species.

Monitoring of amphibians was performed for 80% of the projects evaluated. Strategies ranged from incidental observations to egg mass counts and larval sampling. The majority of projects listed only incidental observations. Just two projects specifically monitored eggs, larvae, juveniles and adults (one project was successful, one was not). Only six projects specifically monitored for the use of the pools by indicator species such as the wood frog or spotted salamander, while the remainder referred only to “tadpoles” and “frogs” without recording species.

Monitoring of soil conditions, invertebrate populations, and other wildlife was extremely limited (Table 3).

Performance criteria and remediation plans.

Performance criteria (determinants of success) were very difficult to locate in the documentation. Most projects did not clearly state the intended functions of the pools other than to provide “amphibian habitat.” Although five projects specifically addressed the creation of vernal pools, not one project required the use of the pools by vernal pool indicator species as a criterion for success.

Since performance criteria were not clearly stated, it was impossible to determine whether success had been achieved, unless pools were explicitly declared failures, as was the case for seven of them (47%). Many had insufficient follow-up monitoring and/or too much documentation missing from project files to be able to determine performance criteria or whether they had been met.

Although 11 of the projects (73%) mentioned remediation plans (Table 2), evidence of follow-up and/or remedial action was not available in the files. One project consultant reported failure of the pools to function as vernal pools at every monitoring visit over a three year period. However, the consultant consistently argued against remediation throughout the follow-up period, stating that the developer had acted in good faith. The final monitoring report for that project indicated that the pools had improper hydrology, failed to provide amphibian breeding habitat, and that the surrounding wetland buffer had also failed to develop as intended. Only one project specifically planned remediation measures, if needed, to “establish appropriate hydrology and enhance/ensure success of the site and use by vernal pool species.”

Discussion

A review of 15 vernal pool creation projects illustrates that the majority of the projects fail to meet the goal of replacing lost vernal pool functions, or indeed even state that as a goal. There is an immediate need to develop standards for all aspects of vernal pool creation including record-keeping, development of detailed mitigation plans including defining criteria for success, and for development of remediation plans if functions fail to be replaced. Only when such standards are developed and implemented will we be able to approach the goal of a no-net-loss of vernal pool functions. Based on the results of this review of project documentation, we provide suggestions for improving compensatory mitigation for vernal pools based on current published literature in each of the areas in which we found weaknesses.

The SWANCC vs. Corps decision limits the scope of the Corps to regulate isolated waters, including vernal pools. Potentially removing much of the Clean Water Act protection for 30% to 60% of the Nation’s wetlands (Kusler 2001), this decision shifts more responsibility for regulating small wetlands on to state and local governments. Currently, vernal pools and their adjacent critical terrestrial habitat are inadequately protected. Regulatory strategies that focus on protecting only the breeding pools will most likely fail to maintain healthy amphibian populations—protection of critical terrestrial habitat must also be a priority (Gibbs 2000; Babbitt and Tanner 2000; Pope and others 2000; Marsh and Trenham 2001; Calhoun and Klemens 2002). Pools created which are surrounded by parking lots, roads, or degraded upland habitat will not perform the functions of vernal pools with intact adjacent habitat. None of the recent state statutes addressing protection of vernal pools or isolated wetlands specifically addresses protection of the adjacent critical terrestrial habitat, with the exception of Maine’s Significant Wildlife Habitat designation.

To ensure that the landscape-scale functions of vernal pools, including support of local populations of herptiles, small mammals and birds, and energy transport to terrestrial ecosystems (Wilbur 1980; Pough 1983; Ernst and Barbour 1989) is maintained, regulators at all levels need to track and assess isolated wetland losses in spatial and temporal contexts and be certain that compensatory mitigation replaces these multiple scale functions (Dodd and Cade 1998; Semlitsch and Bodie 1998; Snodgrass and others 2000).

In addition to critical terrestrial habitat, amphibians (and many other wetland-dependent species) also need suitable habitat connecting wetlands (Joyal and others 2001). Semlitsch (2002) suggests that management plans incorporate additional terrestrial habitats for corridors of movement from ponds, and Gibbs (2000) recommends retaining minimum wetland densities in human-dominated landscapes to conserve wetland-dependent species. Clearly, mitigation plans should take into account the density and class of wetlands within the watershed before planning on-site or off-site mitigation.

Improving Project Documentation

Lack of documentation of vernal pool creation methods and monitoring constrain the progress of wetland mitigation through vernal pool creation. Files of project documentation at federal and state regulatory agencies reflected inadequate record keeping. Because statements of project intent and monitoring reports were often absent or missing, it was difficult, if not impossible, to assess project success. Brown and Veneman (1998) report similar difficulties in their assessment of compensatory mitigation in Massachusetts where the poor quality of record keeping made determination of wetland replication success nearly impossible. They suggest development of a statewide reporting and record-keeping system that would track wetland impacts and corresponding mitigation requirements. Implementation of such programs at the state level throughout New England would facilitate the flow of information among regulators, consultants, and developers, and would facilitate project tracking and subsequent enforcement, thereby increasing compliance and project success. Careful project monitoring and tracking of information for other vernal pool projects across New England, and the rest of the country, would provide valuable case-studies, in lieu of research projects, of factors that lead to project success. A clearinghouse (perhaps at the federal level) for cataloging this information would be helpful for improving and guiding future vernal pool creation attempts.

This tracking approach would also allow regulators and resource managers to assess compensatory mitigation projects in light of landscape scale issues of wetland connectivity and the ecological integrity of isolated pools.

Mitigation Plans

Project monitoring.

Most projects require a three to five year monitoring period to establish that functions have been successfully replaced. To date, monitoring of created pools has been haphazard and often monitoring procedures do not assess the specific functions of vernal pool habitats as select breeding sites for species intolerant of permanent fish populations and other resident predators.

This problem is currently being addressed by the New England District Corps. They have recently modified a document that will address standards for monitoring vernal pools. This revised Checklist for Review of Mitigation Plan (US Corps 2002) calls for monitoring of “obligate and facultative vernal pool species weekly for four weeks from the beginning of the vernal pool activity in the spring, and then biweekly until the end of July for the entire monitoring period.” Identification of amphibians and invertebrates using the pools require that frogs be identified to species level, salamanders to genus level, and macroinvertebrates to the level of order. Data on pool depth, substrate type and plant species in and around the perimeter of the pool(s) is part of the checklist.

The Corps Checklist addresses the issue of achieving success by requiring “evidence that adequate hydrology will be provided to support at least one obligate vernal pool species (mole salamanders, wood frogs, or fairy shrimp).” The Corps is considering including a more specific indication of “adequate hydrology.” The draft Corps Checklist requires documentation of water regime.

This cook-book approach to monitoring will standardize performance criteria. This is particularly important as many consultants or monitoring personnel are not familiar with the science.

Standardizing performance criteria for functional assessments.

Compensatory mitigation in the United States generally occurs without thorough analysis of wetland functions (Brinson and Rheinhardt 1996). The results of this study reflect this trend. Project documentation often did not state which wetland functions were being mitigated through pool creation. Original wetlands were not assessed for functions and prior to construction, nor were comparisons made to existing natural vernal pools.

One goal of this project was to determine whether mitigated vernal pool creation attempts are successfully replacing vernal pool functions. In order to achieve a goal of no-net-loss of functions, it is necessary to determine wetland functions prior to development. Brinson and Rheinhardt (1996) recommend the use of reference wetlands, sites within a specific geographic region that encompass the known functional variation of a class of wetlands, in compensatory mitigation. Reference wetlands offer a way to standardize compensatory mitigation by (1) making the project mitigation goals explicit by identifying reference standards that typify sustainable conditions in a region, (2) providing design templates for wetland restoration and creation, and (3) providing a framework for estimating a recovery or decline in functions after restoration. Additionally, use of reference wetlands removes bias, providing an objective mechanism for assessment of functional success at project completion. For example, Brinson and Rheinhardt (1994), suggest assessing functions in four categories: hydrologic, biogeochemical, plant community maintenance, and animal community maintenance. A guidebook for Californian vernal pools has been developed which list ten major vernal pool functions across the above four categories (Butterwick 1996). Projects to develop guidebooks for vernal pools in other parts of the country were recommended, but to date remain unfunded (M. Schweisberg, personal communication). Some guidance, however, is provided in the published literature and is discussed below to address the shortcomings in projects as outlined in our review.

One key function of vernal pools in the northeastern United States is as breeding habitat for amphibians adapted to fishless environments. Clearly, one key performance criterion should be the maintenance of viable populations of these target amphibians (including maintaining a fish-free environment). Therefore, it is necessary to consider the life-history characteristics of the amphibians for which the breeding pool is being created. Government personnel cite creation of “amphibian breeding habitat” rather than “vernal pool habitat” as a common practice by permittees to skirt replacing specific vernal pool functions for indicator amphibians. Breeding habitat for amphibians that breed in permanent waters, such as green or bull frogs, is easier to attain. The Massachusetts Natural Heritage and Endangered Species Program (MA NHESP) plan to evaluate “success” of the creation attempts based on successful metamorphosis of target amphibian species (M. Burne, personal communication).

In cases of pool restoration through translocations, it is recommended that only after multiple years of breeding, or after a cohort of second-generation adults is found breeding, should the pool be considered a success (Semlitsch 2002). In restored pools (through translocations), if adults fail to return after five to ten years, pools are considered a failure (Denton and others 1997). Monitoring of colonized created pools should follow a similar regime. For example, in one of the created pool sites assessed in this review, three pools were created. Three different hydroperiods were incidentally created, temporary (dries every year), intermediate (dries once in three years), and permanent. In the permanent and intermediate pools, wood frogs and spotted salamanders breed every year. However, after three years, green frog populations built up sufficiently to render both pools sinks for wood frogs and spotted salamanders. The failure of these two pools did not become evident until the third or fourth year of monitoring, and even so, adults continue to return to the sites (Vasconcelos and Calhoun, unpublished data).

Replacement of vernal pool functions should also include consideration of both spatial and temporal issues. Ideally, replacement pools will be located within the same watershed (“on-site”), and will perform the functions of the original pools (“in-kind”). However, if adjacent critical terrestrial habitat is not available on-site, pools should be sited in more appropriate settings. To offset temporal loss of functions, wetland impacts should be prevented until full functional establishment of the mitigation areas.

Developing pool creation standards.

A review of the scientific literature on vernal pool creation attempts yields very little information, with none specific to New England. The only design recommendations for vernal pool creation attempts that could be located were proposed by Richter (1997) working in the northwestern United States. He describes requirements for amphibian breeding, feeding and refuge habitat, dispersal habitat, migration corridors, and a hierarchy of explicit wetland attributes, including current velocity, minimum water depth and water level fluctuation, wetland size/orientation/configuration, and buffer condition. Richter (1997) serves as a starting point and guide for those venturing into the science of vernal pool creation. Semlitsch (2002) provides biological guidelines for effective recovery and restoration plans for aquatic-breeding amphibians that may have applicability to vernal pool creation in New England. Below we review strategies for creating functional vernal pools.

Hydrology.

In their evaluation of 116 wetland mitigation projects, Breaux and Serefiddin (1999) found that the most measured feature of compensatory mitigation sites was vegetation, with hydrology “a distant second.” The most crucial parameter to be given consideration during pool design is water regime (seasonal depth, duration, and timing of inundation and saturation) (Kusler and Kentula 1990; Rowe and Dunson 1995; Snodgrass and others 2000). Functional vernal pools are characterized by periodic drying that will prevent breeding populations of fish and establishment of significant resident populations of other predators. The amount of time the pool must hold water varies among species, with fairy shrimp at the short end of the continuum (as little as six weeks), and ambystomatid salamanders at the other end (three to four months). Additionally, some amphibians exhibit distinct minimum and maximum water depths in which they will oviposit. Richter (1997) has inferred that oviposition of most temperate amphibians is at depths between 10–100 cm. These figures are corroborated by research on successful natural vernal pools in Massachusetts and Maine (Brooks and others 1998; Calhoun and others, in press; Kolozsvary and others, unpublished data). Medium-depth water may offer optimal temperatures to accelerate egg development (Duellman and Trueb 1986), and may also indicate the wetland is only semi-permanently flooded, therefore eliminating the threat of predation.

There is not much guidance on how to replicate hydrology in created pools. Use of shallow groundwater monitoring wells are useful in some situations to assess long-term hydrologic conditions at a specific site by determining depth to the water table (Antonelli 1997). Water budget or sources of hydrologic inputs and outputs to the system, should also be evaluated. Allen and others (S. Allen, personal communication) state that a variety of hydrologic conditions can support successful vernal pool creation, including systems dominated by groundwater contact, perched conditions, and ephemeral surface flows.

Winston (1997) suggests that the use of groundwater models in the design of groundwater-dominated wetlands may provide a better chance of success than designs based solely on pre-construction hydrology. Such models may increase the likelihood of success as they can incorporate changes in hydrology due to construction. Such topographic changes that may alter groundwater dynamics are often not accounted for otherwise. When combined with seasonal variation in recharge, this alteration in hydrology can cause wetland-creation efforts to fail. Additionally, Bonner and others (1997) found that hydroperiod is influenced by many environmental factors, some of which can be determined during pool design (e.g. soil composition of the basin substrate and degree of canopy cover).

Vegetation.

There are differing opinions in the literature regarding the best method for introducing vegetation into wetland design. Mitsch and Wilson (1996) describe two common approaches. The “designer” method emphasizes a botanical engineering approach, via seeding the wetland with specific species, the survival of which becomes a measure of ecological success. “Self-design,” on the other hand, emphasizes introduction of as many species as possible, knowing that natural forces will choose the most suitable design. Mitsch and others (1998) describe the use of self-design for introducing any plant, protist, invertebrate, or vertebrate, stating that “the system itself will optimize its design by selecting for the assemblage of plants, microbes, and animals that is best adapted for existing conditions.” Since water, wind, and animals can carry propagules from one wetland to another, they suggest that the introduction of plant species may not be necessary in developing a functional ecosystem. On the other hand, Galatowitsch and van der Valk (1996) report that, due to the isolated nature of most restored wetlands, seed dispersal may not occur as needed, and therefore active planting may be required. We suggest that decisions regarding both animal and plant establishment in a created pool be based on the quality of the surrounding landscape. For example, created pools in highly developed landscapes may be more susceptible to invasion by exotic species and may warrant extensive management including plantings. Pools in relatively undisturbed environments close to native source populations may be self-design candidates.

Other pool characteristics.

Additional details that should be given consideration during planning for vernal pool creation include pool basin soil characteristics, microtopography of pool bottom, and availability of attachment sites for egg masses. Until wetland vegetation develops in newly created wetlands, oviposition sites can be added in the form of thin, dead branches and twigs (Richter 1997). DeWeese (2000) recommends that site selection include historic vernal pool soils to ensure success, as well as creating pool bottoms with microtopography that will enhance plant distribution and invertebrate habitat.

The Massachusetts Natural Heritage and Endangered Species Program (MA NHESP) is in the early stages of developing vernal pool design recommendations which include establishing the correct hydroperiod by evaluating soils and hydrological inputs and outputs, and salvaging organic layers of lost pools to help “seal” the bottom and colonize the new pools with an invertebrate food base and seeds from native. They also recommend locating the pool near a known source population of target amphibians, and preferably within an existing cluster of pools. Regular monitoring and remediation of problems, such as invasive species control, is of utmost importance to project success.

Remediation.

Less than 15% of the projects reviewed implemented a remedial or adaptive management plan when pool functions failed to be met. Remedial efforts included modification of hydrology to promote ephemeral conditions, control of invasive vegetation and of mechanical disturbance from all-terrain-vehicles, and translocation of eggs, larvae, or adults of target amphibian species. Little research has been done to determine the success or concerns surrounding such tranlocations and, owing to disease and other issues (Calhoun and Hunter, 2003) the practice is still hotly debated (Lubow 1996; Fagan 1999; Seigel and Dodd 2002; Semlitsch 2002; Trenham and Marsh 2002).

In general, we recommend that mitigation plans include plans for monitoring performance criteria, and when it is clear that key vernal pool functions have not been successfully reproduced that adaptive management plans are implemented. Maintaining the proper hydrology is the most challenging aspect of vernal pool creation and remedial efforts should address this issue.

Conclusion

The results of this study show considerable variation in consultants’ understanding of vernal pools and pool-dependent species. Mitsch and Wilson (1996) suggest that many engineers, consultants, scientists, and landscape architects claim to be experts in wetland restoration and creation, while actually possessing little knowledge or experience in wetland ecology. Implementing certification standards for those involved in wetland creation could help to decrease the relatively high number of mitigation project failures which may be attributed in part to a lack of understanding of the principles of wetland science in general, and the life-history requirements of vernal pool species in particular.

Conservation of isolated wetlands, including vernal pools, can best proceed with recognition by land-use managers and regulators of their functions at multiple scales. They provide unique breeding habitats for specialized species, but they also function together as a suprawetland in an upland matrix providing support to a wide variety of wildlife species.

Scientists generally agree that no wetland can be replicated exactly (Kusler and Kentula 1990; Pechmann and others 2001). For this reason, we recommend stricter adherence to mitigation sequencing (that is, avoidance, minimization, and compensation), and urge that projects are permitted for only the least environmentally damaging practicable alternative. We do not recommend creation of vernal pools, even with high success rates, as a fix for the rapid loss of naturally occurring small, isolated wetlands. However, steps can be taken to improve the likelihood of success in an effort to achieve the goal of no-net-loss of functions when mitigation calls for vernal pool creation.

Table 1 Vernal pool facultative species and state conservation status1
Full size table
Table 2 Summary of project documentation for 15 New England vernal pool creation projects
Full size table
Table 3 Summary of parameters considered in project design and monitoring protocols for 15 New England vernal pool creation projects
Full size table