more on fire

From 1997 through 2000, I participated in a prescribed fire study with The Nature Conservancy of Missouri. Teamed with Beth Churchwell of TNC and assisted by a cadre of volunteers, mostly graduate students from Washington University, we monitored the effects of prescribed fire on more than 20 species of amphibians and reptiles on the 6,000 acre Chilton Creek Preserve, near Van Buren.

The preserve is located in the Ozarks, on the Current River. Mostly it's oak woodland, with areas of shortleaf pine, glades (openings of exposed bedrock and sparse grasses and herbaceous vegetation located mostly on xeric ridgetops), and graminoid fens bordering high-gradient streams. There is a small area of floodplain forest along the river, and at least one cave. The woodlands range from relatively mesic areas on lower north or east facing slopes, to stunted and gnarled post oaks on well drained south or west facing ridges with loose cherty soils. Layers of dolomite and sandstone alternate and are exposed in various parts of the preserve, which consists of alternating ridges and valleys with about 300 feet of local elevation change.

TNC had already designed their burn program when I became involved, and we had only a short time to gather pre-burn baseline data. The preserve had been divided into six burn units using existing roads and trails as firebreaks, with control units located around the periphery of the site and not scheduled for fire. The treatment units were burned on various rotations; annually or bi-annually in some units, or on a randomly determined and variable frequency in others.

We established sample plots in both treatment and control units, and sampled all of them pre-treatment and for two years after the onset of fire management. Thus, some plots were burned once and some twice during our study, and the control plots remained unmanaged.

We sampled three times per year, in early spring, late spring, and fall. Typically we'd spend three days on site each visit, camping within the preserve to maximize field time. Given the rugged nature of the site and the fact that plots were randomly located and thus often far from trails, it was grueling work. We couldn't spare time off for bad weather, so we sampled in heat, cold, thunderstorms, and more.

I obviously can't go into a detailed analysis here, in this limited space. Suffice it to say that after two years, we could see the differences in the vegetation structure... mostly a reduction in the shrub layer and an increase in the herbaceous layer. On part of one plot, a too-intense fire left a shrub thicket two years later, but in most of the preserve it became easier to walk through the woods because of the more open understory.

In treatment units, we saw reductions in numbers of only two amphibian species, two types of Eurycea which are characteristic of streams and seeps, which were and still are among the most abundant amphibians on the site, and which we hypthesize had been allowed to expand beyond their original habitat niche by tree canopy closure. If we're right, then fire basically forced these species to pull back toward their original stream-associated habitat and abandon the drier uplands.

Some other species increased in numbers. These included, unsurprisingly, certain types of lizards and snakes adapted to drier and more open and sunlit conditions. They also included at least one type of frog, possibly as a result of increased ground cover associated with expanding and denser herbaceous vegetation. However, the duration of the study wasn't really great enough to determine the full extent of potential benefit to some species. This is particularly true for longer-lived species, which respond more slowly to environmental change.

One conclusion can be stated with certainty: There were no catastrophic losses of amphibians or reptiles, despite the relatively aggressive fire frequency. Burning was done both spring and fall (unlike many Midwestern sites, Native American fire was apparently mostly in the spring here, and associated with the hunting migrations of native peoples), but was done under appropriate conditions and by experienced burn crews. Most of the fires were low ground fires, and exceptions were limited to small areas.

It appears that returning prescribed fire to the oak-pine woodlands of the Missouri Ozarks resulted in increased diversity, both in vegetation communities and in the herpetofauna. In the latter case, it did this by reducing numbers of some (but not all) of the most abundant species, while increasing numbers of some less common species.

Perhaps someday I'll return to the site. We recorded GPS coordinates for all sample plots, so even if the landscape changes, which it likely has after more than a decade of fire, it will not be difficult to find the same locations. Whether that happens or not, I have many fond memories from that study, had an opportunity to work with some great people, and learned a lot that I've since applied on other sites.