Northern Prairie Wildlife Research Center
Surprisingly few papers have addressed aspects of fire-wetlands relations; fewer yet have had this subject as a major focus of investigation. In general, fire has been treated as one of a number of management tools appropriate for wetlands, with its major use that of eradication of undesirable vegetation. Unlike the literature on fire in terrestrial upland communities, however, specific fire prescriptions, knowledge of fire behavior under different fuel loadings and environmental conditions, and the detailed consequences of differing fire frequencies, fire intensities, and fire severities in wetlands are largely unknown. As a physical phenomenon, fire in wetlands has only been studied in detail for deep peat soils, where extinguishing a fire can be difficult or well nigh impossible. However, recent studies have begun to emphasize nutrient release, mineral cycling, and other chemical effects of fire upon the soil and subsequent vegetative vigor and productivity.
The earliest references to fire in North American wetlands are the anecdotal accounts of early travelers in the Upper Midwest and Great Plains [recently compiled by Higgins (1986)], that mention fire in lowland areas in the late 1600's. The first North American reference to the value of fire in managing wetlands for wildlife (waterfowl) appears to be that of Furniss (1938), although L. J. Bennett (1938. The Blue-winged Teal, its ecology and management. Collegiate Press, Ames, IA. 144 pp.) mentioned in passing in the same year that breeding habitat conditions for upland-nesting waterfowl could be improved by selective use of fire. The role of fire in maintaining certain plant communities was appreciated long before this, however. Korstian (1924) and Korstian and Brush (1931) identified fire as a major determinant in the establishment and maintenance of Atlantic white-cedar communities; Lewis and Dowding (1926) and Lewis et al. (1928) identified the role of fire in muskeg communities in the Canadian North; the importance of fire in Louisiana marshes was identified by Viosca (1928, 1931); Beaven and Oosting (1939) identified fire as a great agent of change in baldcypress swamps; Wells (1928, 1931) and Penfound and Hathaway (1938) recognized fire as important in southeastern United States pocosins and coastal marshes; and Bradbury (1938) listed fire as a marsh management tool that benefited wildlife while assisting mosquito control.
Despite appreciation among plant ecologists of the role of fire, early use of fire in wetlands management by land management agencies was apparently stymied by the mind-set of the majority of first-generation wildlands managers who viewed fire as an entirely negative phenomenon (Conway 1938; U.S. Forest Service Library 1938; Cox 1939; Hanson 1939). Fire as a natural occurrence, and as a practical tool for managing coastal marshes for furbearers, was generally appreciated by the public, however, and observation of general fire effects, versus scientific investigation per se, led wildlands managers to several basic conclusions on use of fire in wetlands by the 1940's. Wells (1942) and Garren (1943) identified fire as responsible for the development and maintenance of several wetland communities in the Southeast; Griffith (1941) and Smith (1942) described the value of burning in Atlantic coast marshes; Lay and O'Neil (1942) discussed the value of burning for muskrat management on the Texas Gulf coast; and Cartwright (1942) and Ward (1942) identified the value of burning in maintenance of the Delta Marsh, MB. Lynch (1941) described several "types" of burns in Gulf coast marshes, suggested their potential value in managing other types of marshes, and identified the need for further study of fire in marshes. Lynch's descriptions, embellished and repeated by later authors, became the litany of wetland fire managers, but there was no substantial response to his request for further studies of marsh burning for 20 years, until Hoffpauir's (1961a) thesis appeared as the first of a long series of quantitative studies of fire in the Louisiana marshes.
The frequency distribution of publication dates of literature addressing fire-wetlands relations (Table 2) documents the initially low, but steadily increasing, amount of interest in the subject into the 1980's. In the present decade, interest in investigating fire-wetlands relations has risen dramatically with the result that more than one-third of the papers in this bibliography have appeared since 1979. Since certain categories of fire-wetlands papers, namely limnological studies and most information from the Far North, were purposely limited in the annotated bibliography, the apparent approximately 100% increase in reported results per decade from the 1920's through the 1970's greatly understates the increase in the literature on fire in general and the sum of literature related to fire in all wetland ecosystems worldwide in particular. For example, Vierick and Schandelmeier (1980) listed approximately 750 references on the effects of fire in Alaska and adjacent Canada, a number 2.4 times larger than the material in this annotated bibliography. Detailed studies of fire in the North have increased dramatically since World War II, with the greatest increase in the most recent years.
The rate of increase of reports of fire-wetlands studies appears to be leveling off if the data available through July 1987 can be taken as an indication. Nonetheless, the number of studies is still increasing and a firm basis for much more detailed work has been developed in many parts of North America. As one indication of the "arrival" of fire-wetlands studies in the consciousness of land managers throughout North America, the number of texts and reviews that addressed fire-wetlands relationships grew dramatically in the 1970's and this demonstration of level of interest has been sustained. (Table 2).
The geographic distribution of studies of fire-wetlands relations (Table 3) only partially reflects the distribution of major wetlands in North America. Thus, one can recognize several areas of emphasis already addressed in fire-wetlands studies as well as other areas in need of basic investigation. Within the Atlantic states and provinces, 27 of the 95 papers (28%) have addressed freshwater pocosins, bogs, Carolina bays, and canebrakes of the Coastal Plain. These increasingly rare communities are dependent upon fire to maintain a mixture of seral stages and currently remain the object of intense botanical and conservation interest. Two other large, well known wetlands on the Atlantic coast have also been the subject of substantial studies, specifically of the relation of fire to many aspects of ecosystem functioning. The Okefenokee Swamp of Georgia with 16 studies and the Great Dismal Swamp on the North Carolina-Virginia border with 6 studies have received some of the most sophisticated analyses in the East. A third area, the Pine Barrens of New Jersey, once known for extreme conflagrations, now is burned much less often and studies have concentrated on upland sites despite the great number of wetlands and surrounding coastal marsh in the Barrens. The five studies on the Pine Barrens are thus but a fraction of the literature on the area's relation to fire. As an indication of interest in these four major wetland ecosystems of the Atlantic states, each has been the subject of recent monographic treatment: pocosins (Richardson 1981; Sharitz and Gibbons 1982; Ash et al. 1985); Okefenokee Swamp (Cohen et al. 1984); Great Dismal Swamp (Kirk 1979); and the Pine Barrens (Forman 1979).
Studies from Georgia and North Carolina and those that address the entire mid-Atlantic and Southeastern region dominate the available publications on fire-wetlands relations in the eastern United States and Canada. These areas include the numerous previously mentioned Okefenokee and pocosin studies, but additional contributors to this geographic concentration of effort appear to be the long-standing interest by plant ecologists in the flora of the Atlantic Coastal Plain and substantial interest in economically efficient management of coastal marshes and eastern pine and lowland forests. In the latter cases, both industrial and wildlife management reasons have provided incentives for research on fire-wetlands relations.
Within Florida, 20 of the 32 studies (62%) have addressed the Everglades ecosystem or nearby natural areas. With increasing human pressure upon natural landscapes in southern Florida, and the ongoing research programs of the Everglades National Park and nearby preserves, substantial work on this system has ensued. Monographic treatment of the entire system includes Loveless (1959) and Duever et al. (1986). Cypress swamps and forested wetlands in general have been recent subjects of intense study because of their value to fish, wildlife, and water quality, and in some areas, their potential for treating wastewater. Cypress swamps have been covered in two recent monographs (Ewel and Odum 1984; Duever et al. 1986); Atlantic white-cedar swamps in two older reviews (Korstian and Brush 1931; Little 1950); and bottomland hardwood and other forested wetlands have been addressed by a number of Fish and Wildlife Service and other government publications (e.g., Wharton et al. 1982).
In the Gulf states, almost all of the studies have addressed coastal areas, with emphasis being placed on the coastal prairies of Texas (26%) which have been of substantial interest to livestock producers as well as Louisiana salt and brackish marshes noted for their forbearer and waterfowl resources (44%). These latter marshes have been subject to manipulation for so long (200+ years) that management schemes are relatively straightforward, but recent coastal subsidence and loss of marsh threatens coastal Louisiana, suggesting the need for further studies.
The inland Midwest and Great Plains were only addressed by 20% of the papers despite the large area encompassed. The extensive inland wetlands of the interior thus seem severely understudied from the standpoint of their interaction with fire. Given their importance to migratory birds and resident wildlife, and their recently much diminished areal extent, experimentation with fire and all other available management options seems imperative if the resource is to be sustained. The high number from Wisconsin includes numerous general studies, but also reflects concentration on management of Horicon Marsh. Six of the nine papers from Utah were by the same set of authors, and comprise detailed studies of the Great Salt Lake marshes. The prairie pothole country of the Dakotas, Minnesota, and the Canadian Prairies has received little concentrated study with the notable exception of the Delta Marsh, MB. The Delta Marshes were the site of early attempts to manipulate emergent marsh vegetation with fire (Ward 1942, 1968) and continue to be an important site for experimental studies of alternative marsh management strategies (Neckles et al. 1985; Thompson and Shay 1985; Shay et al. 1987).
There is no clear explanation for the lack of studies of fire-wetland relations in the West. The paucity of wetlands in many areas may be one factor, but continued interest in maintaining riparian zones suggests that studies addressing grazing, fire, and other range management practices are still needed. The effects of fire on high mountain wetlands appear completely unstudied, as does the effect of fire on Pacific coastal marshes.
The concentration of studies in relatively few geographic areas seems to be at least partially attributable to development of graduate programs emphasizing marsh ecology and fire at only a few universities. The geographic distribution of both study sites and issuing institutions that confer degrees for studies of fire-wetlands relations (Table 4) shows the clear numerical superiority of Louisiana State University, and to a lesser extent, Texas A&M University, the University of Georgia, and Iowa State University. All four institutions have access to nearby State and Federal refuges and wildlife management areas, and have ongoing programs that integrate range, wildlife, and botanical studies. In general, however, the subject has not been addressed by student degree programs as often as we had expected, especially given recent emphasis at many universities on whole-ecosystem studies, successional relations in plant communities, and the flow of energy and cycling of nutrients in ecosystems. The most critical finding, however, was that 14 of the 35 theses and dissertations that addressed fire-wetlands relations apparently resulted in either no published manuscripts or published papers that did not discuss the fire-wetlands aspects of the research. Given the general difficulty of locating unpublished Masters theses, especially, it seems clear that those interested in site-specific studies will have to make personal inquiries to locate older, unpublished material.
The subject index shows that several major groups of organisms have received little attention with regard to fire. For example, studies of the effect of fire upon wetlands used for livestock range universally ignore the effects of burning upon grourid-nesting nongame birds, reptiles, amphibians, and invertebrates. Invertebrates, as a group, have received almost no mention in the fire-wetlands literature although they have been addressed in studies of fire in upland sites. Fire in and adjacent to prairie wetlands has been assessed with regard to ground-nesting waterfowl, but the effects of fire on marsh and wading birds dependent upon cattails, common reed, bulrush, and other emergent aquatic plants for cover, food substrate, and nest sites have been barely mentioned. Only in studies of coastal marshes in Louisiana has enough data been collected to provide broad managinent prescriptions. The conclusion from these Louisiana studies is that it is not possible to use fire to manage these coastal areas simultaneously to benefit livestock, waterfowl, muskrats, and other marsh wildlife. This suggests that use of fire in wetlands elsewhere should be for specific management purposes and not applied as a tool of supposed benefit to all marsh wildlife and plants.
Eighty-one (27%) of the entries in the annotated bibliography were reviews of broad geographic scope, or texts discussing numerous aspects of fire, habitat, or species management. As expected, there was much overlap in these review papers, since the original literature is so limited. Nonetheless, despite these reviews, no wetland management handbook exists that fully integrates fire management with other wetland management practices and that presents specific management guidelines. There is a need for such a synthesis, but as this bibliography makes clear, much is yet unknown about the use of fire in marsh management and the effects of fire on wetland ecosystems. Furthermore, the literature reviewed in this bibliography emphasizes more the uniqueness of each wetland system than the similarities of all wetlands in their responses to fire. This suggests that the entire field of study, except for those wetlands the subject of multidisciplinary research (Great Salt Lake marshes, Delta Marsh, Okefenokee Swamp, the Everglades, etc.), is still at the basic data-gathering stage for most of North America and that synthesis might occur more appropriately some time hence. Several areas in need of major inquiry can be defined. We particularly need data to develop fire prescriptions for various wetland types, further knowledge of the effects of fire upon marsh nutrient cycling, a better understanding of the use of fire in managing wetland complexes in the Great Plains, and development of optimal schemes to meet air and water quality objectives while simultaneously meeting wetland management objectives through appropriate use of fire.