Wednesday, 1 February 2012

Restored wetlands are ghosts of their former selves

As a result of urbanization and increasing intensity of anthropogenic land use, an astonishing number of wetland ecosystems have been lost worldwide over the last century. This includes marshes, peatlands, floodplains, mangroves, depressional wetlands, and lacustrine wetlands, which, cumulatively, comprise one of the  most productive and economically valuable types of ecosystem on the planet. Wetland areas provide a number of goods and services, ranging from fish production to erosion control to carbon storage.

 (Lebanon's Aammiq Wetland)

Scientists have long realized the benefits of wetland areas, and, as a result, have encouraged both the restoration of degraded wetlands and the creation of new ones to replace those that have been permanently lost. Over the past several years, studies attempting to assess the efficacy of these mitigation efforts have been stymied by the fact that wetland "health" can be measured in a variety of ways, that different characteristics may reach "natural," or "undisturbed" levels at different rates, and that a number of factors can influence this process. However, a recent analysis by collaborators from UC-Berkeley, the Pyrenean Institute of Ecology, and the Muséum National d’Histoire Naturelle takes all of these factors into consideration in an attempt to identify factors associated with the structure and functioning of restored and newly-generated wetlands, as compared with those at reference sites.

The study was a massive undertaking; the authors compiled data published on 621 wetland sites from around the world. They were able to locate 401 reports on the biology of restored wetlands, and another 220 papers covering the biology of manmade wetlands. From each study, the authors extracted information about hydrologic, biological, and biogeochemical properties of altered/created wetlands, as well as undisturbed/reference wetlands that could be used as an indication of "ideal" or "normal" wetland settings. Over 100 studies reported measurements taken at multiple time points during the restoration process, which allowed the researchers to generate trajectories for each of the three broad categories of variable. Analyses of these trajectories allowed them to investigate how fast different wetland properties were changing in both disturbed and undisturbed areas, as well as to visualize whether disturbed wetlands were poised to become "natural-like," or to reach equilibrium at an entirely different state. Finally, the scientists looked at abiotic variables such as wetland size, source of wetland water, and local temperatures in order to see how these environmental variables can influence wetland restoration efforts.

(Chisago County, Minnesota's Hawkinson Wetland)

Overall, the results of the analysis provide both good and bad news for those who are interested in undoing some of the damage that anthropogenic disturbance has done to wetlands. For example, hydrological features--which can be physically engineered by manipulating topography, soil, and water flow--can be restored almost immediately. Another positive finding was that phosphorus storage seems unaffected by wetland degradation, perhaps because it is more closely tied to underlying geological conditions than biological patterns. Large wetlands and those in warmer (particularly tropical) climates were able to recover some characteristics at fairly decent rates; this was also the case for riverine and tidal wetlands.

On the whole, though, many of the patterns were discouraging. Biological structure in disturbed wetlands (including abundance, richness, and diversity of native animals and plants) only recovered to 77% of reference values, even as long as 100 years after some restoration efforts. Plant assemblages were the slowest to recover, taking an average of 30 years to reach levels similar to those in reference sites. For all classes of resident organisms, absolute values never reached those of reference sites--probably reflecting the difficulties associated with colonizing a new site, sensitivity of some species or age classes to disturbance, and the presence of exotics capable of outcompeting native individuals. Further, the storage and cycling of both carbon and nitrogen were reduced following wetland degradation; even after 50-100 years, restored wetlands only achieved, on average, about 74% of the biogeochemical functioning observed in undisturbed areas. Taken together, the biological and biochemical results suggest that recovery of ecosystem structure might be a necessary first step in achieving functional recovery.

(Scotland's Old Castles Wetland)

The authors point out that two major factors could have been responsible for the fact that restored and created wetlands "lag" behind undisturbed areas. First, the disturbed areas may simply not have had time to achieve the characteristics seen in natural sites. It may be the case that wetland restoration requires a period of centuries rather than decades; we may not reap the benefits of current mitigation projects, but perhaps our children and grandchildren will. Second, restored and created wetlands may only be capable of achieving "alternative states" that will never look like the original habitats. If this is the case, then managers interested in extracting particular ecosystem goods and services will have to figure out how to do so given a whole new set of environmental variables, rather than counting on an eventual return to previous conditions.

Alternatively, it may be possible to come up with totally different--and improved--methods of wetland restoration and creation, leading to habitats that better resemble the ideal that we are trying to achieve. In order to develop more effective management techniques, the authors say, it will be necessary to perform "more realistic, long-term evaluations to find better ways to alleviate constraints limiting the recovery of wetland ecosystems."

Moreno-Mateos, D., Power, M.E., Comín, F.A., and Yockteng, R. 2012. Structural and functional loss in restored wetland ecosystems. PLoS BIOLOGY 10(1):e1001247.

Thanks to the following websites for providing the images used in this post: 

No comments:

Post a Comment