Potential effects of mowing prior to summer burning on the Eastern Massasauga (Sistrurus c. catenatus) at Squaw Creek National Wildlife Refuge, Holt County, Missouri, USA.

Abstract:

We conducted a pilot study to begin evaluating the use of pre-burn mowing to reduce or eliminate Eastern Massasauga (Sistrurus c. catenatus) mortality during summer prescribed fires. Drift fencing was used to determine the presence of Eastern Massasaugas on a 8.1 ha portion of native wet prairie at Squaw Creek National Wildlife Refuge during the summer 2001. Twenty-two days of trapping with 10 randomly located drift fence traps (n=220 trap nights) resulted in the capture of 96 snakes representing 8 species, including 9 Eastern Massasaugas. After trapping was completed the site was mowed to a height of 20 cm and then burned 13 days after mowing was completed. Post-burn transect searches of the site revealed one yellow-bellied racer (Coluber constrictor) mortality due to the fire. The low mortality rate suggests that pre-burn mowing may potentially reduce fire related mortality of Eastern Massasaugas and other snake species by negatively modifying occupied habitat which in turn forces snakes to leave the area or seek rejuge below ground.

Key Words: Eastern massasauga, rattlesnake, Sistrurus catenatus catenatus, prescribed fire, Missouri

1. Introduction

The Eastern Massasauga is a state endangered species in Missouri (Missouri Natural Heritage Program, 2003) as well as a candidate species for listing under the United States Endangered Species Act (Szymanski, 1998 and Federal Register, 1999). The general ecology of this species is well summarized in Johnson et al. (2000). in Missouri massasaugas mainly utilize wet prairie and to a lesser extent drier upland prairie habitat with minimal shrub and tree cover (Seigel, 1986 and Johnson et al., 2000). Preservation of this early successional habitat is imperative for the continued survival of this species in Missouri.

A common problem for land managers trying to maintain prairie habitat is encroachment of woody vegetation. An effective tool for controlling this problem is the utilization of prescribed fire during the summer months which may either kill or temporarily set back woody vegetation (Anderson, 1997 and Adams et al., 1982). Other means of controlling woody growth include herbicides and mechanical manipulation, although controlled burning appears to be the most efficient and cost effective method.

Unfortunately, prescribed fire during the summer months coincides with the middle of the seasonally active period for massasaugas, as well as other reptile species, and can result in direct mortality. Durbian (2001) reported a 21% mortality rate for massasaugas after a spring bum conducted on 18 April at Squaw Creek National Wildlife Refuge and Frese (2003) reported an 85% mortality rate for all snakes found after a fall bum on tallgrass prairie in northern Missouri. Mortality and injury to reptiles from seasonal fires has also been described by Babbitt and Babbitt (1951), Erwin and Stasiak (1979) and Heinrich and Kaufman (1985). The potential risk for direct mortality of massasaugas from summer bums creates a tenuous management situation when dealing with prescribed fire as a management tool. For this reason it is vitally important to explore management techniques that can be used to reduce or eliminate mortality due to summer burning. A potential solution may be mowing prior to conducting prescribed fire. The reduction of cover through mowing, during the warm summer months, may result in habitat conditions that are intolerable to massasaugas, due to higher surface temperatures, which may force them to temporarily leave the area or go under ground until the habitat conditions are once again suitable. Burning habitat that snakes have temporarily abandoned or during a period when snakes are under ground should prevent direct mortality from fire. We initiated a pilot study during the summer 2001 to begin evaluating the potential effects of pre-burn mowing on summer burn mortality of Eastern Massasaugas.

2. Materials and Methods

This study was conducted on the 3,012 ha Squaw Creek National Wildlife Refuge located on the Missouri River floodplain in Holt County, Missouri. Squaw Creek National Wildlife Refuge consists of 6 main habitat types including old field (24 ha), developed land (102 ha), agricultural cropland (234 ha), forest (558 ha), prairie (643 ha), and wetland (1,451 ha)(Figure 1). The refuge currently harbors one of three known populations of Eastern Massasauga Rattlesnakes in Missouri (Johnson, 2000). The main wet prairie (380 ha), located in the center of the refuge, supports the largest portion of the population.

[FIGURE 1 OMITTED]

The study site encompassed an 8.1 hectare portion of native grass located in the northwestern corner of the main wet prairie (Figure 1) which consists of saturated soil that has been managed for native warm season grasses including Eastern gamma grass (T. dactyloides), Indian grass (S. nutans), switch grass (P. virgatum) and big bluestem (A. gerardi) since 1996. This area is immediately adjacent to a seasonally flooded wetland, on the east and south sides, that is principally managed for waterfowl and shorebirds (Figure 1). The wetland is dry during the summer months and does not provide a barrier to snake movement to and from the main wet prairie. During 1999 and 2000 the site was encroached by cottonwood (populus sp.) and willow (salix sp.). A vegetation survey conducted in 2001 indicated that the percent frequency of woody vegetation on this site was 14.8% for cottonwood and willow combined (F.E. Durbian, Unpublished Data).

Snakes were captured using 10 drift fences (Fitch, 1960) randomly located on the site (Figure 2). Each 0.6 m x 7.3 m fence was constructed from 1.3 cm plywood. A single 0.6 cm hardware cloth funnel trap was placed at either end of each fence for a total of 20 traps. Trapping took place 4-5 days per week during the period of 11 June-13 July. Traps were checked daily, typically late morning, and twice daily when temperatures exceeded 32[degrees]C. Grass was placed over the traps to provide shade and reduce heat stress/mortality. The number of each species captured and released was recorded daily.

All drift fences were removed on 16 July and the entire site, excluding approximately 1.2 ha of mature trees located on the east side, was mowed to a height of approximately 20 cm on 17 July. This height was chosen based on recommendations by Johnson et al. (2000) who suggested that mowing at heights greater than 10-15 cm can "reduce or eliminate mortality from mowing." The controlled burn took place on 30 July (13 days post-mowing) and followed protocol within the refuge fire management plan (Speer, 2001). A combination of backing (25%), flanking (25%) and head (50%) fire was utilized to achieve a safe and effective burn. The BehavePlus Fire Modeling System (Systems for Environmental Management, PO Box 8868, Missoula, Montana, 59807) was used to calculate the maximum burn parameters for the fire (Table 1) using variables collected on the day of the fire. Maximum values were used to provide the most extreme conditions that could have occurred on the fire. On the morning of 31 July, a transect search for burned snakes was conducted by 3 refuge staff members who walked 10 meter wide grids on the entire burn site. All three staff members were experienced in looking for both live and burned snakes.

3. Results

A total of 8 species, including 9 massasaugas, were captured during 220 drift fence nights (1 fence with 2 traps/24 hrs) (Table 2). Other species captured and released unharmed included two prairie voles (M. ochrogaster), one least shrew (C. parva), one yellow warbler (D. petechia) and one indigo bunting (P. cyanea).

The controlled burn took approximately 3 hours to complete and burned 100% of the site down to mineral soil. A post-fire inspection revealed that a majority of the standing woody vegetation was affected (scorched) by the fire.

The transect search effort took approximately 2 hours to complete. One dead yellow-bellied racer and one dead western painted turtle were located. The snake appeared to have been burned while the turtle was likely dead prior to the fire, based on body decomposition.

4. Conclusion

The mortality rates, based on total captures, of 0% for massasaugas and 1.04% for all snakes were much lower than expected relative to the figures reported by Durbian (2001) and Frese (2003) who found post-burn mortality rates of 21% for massasaugas and 85% for all snakes, respectively, after active season prescribed fires. This information suggests that mowing prior to summer burning has the potential to reduce the mortality of massasaugas, and other snake species, that occurs during active season burns. These low mortality rates may possibly be attributed to the creation of an environment inhospitable to snakes. Mowing vegetation to 20 cm greatly reduced the amount of shade present and reduced humidity at the surface level. These factors combined with high daytime ambient temperatures (~32[degrees]C) likely created a very hot micro-climate near the ground, forcing snakes to leave the area and/or seek shelter below ground during the hotter portion of the day. Alternatives to this hypothesis include: 1) Many snakes were burned but not located. This is highly unlikely as the staff members involved in the search effort all had experience finding both live and dead snakes in burned areas. 2) General weather conditions, regardless of mowing, prompted snakes to emigrate to cooler areas. This is also unlikely because snakes were captured in the area prior to burning under the same weather conditions. 3) Direct mortality of snakes was incurred by the mowing that took place prior to burning. No post-mowing mortality search was conducted based on the recommendation that mowing to heights greater than 10-15 cm during the hotter portions of the day will have minimal impact on massasaugas (Johnson et al., 2000). Although a survey conducted immediately after mowing would have provided an opportunity to identify the potential impact to massasaugas and other reptiles, it is doubtful that mortality due to mowing would be 100% and there were likely snakes still on the site immediately after mowing.

The results of this pilot study are encouraging and warrant additional research efforts on a larger and more intensive scale. Utilization of radio instrumented snakes and post-mowing searches will provide more conclusive data on the direct effects of both mowing and burning.

Table 1. Fire behavior parameters based on maximum environmental
variables. Parameters and variables are reported in English units of
measure to maintain consistency with standard fire behavior reporting.

Input Variables                      Value

Fuel Model                              3.0
1 Hour Fuel Moisture (%)                5.0
20 Foot Wind Seed                      12.0
Wind Adjustment Factor                  0.3
Slope (%)                               0.0
Aspect (deg)                          180.0

Output Variables (Maximum Values)

Rate of Spread (chains/hour)           85.0
Heat/Unit Area (Btu/[ft.sup.2])       783.0
Fireline Intensity (Btu/ft/sec)      1227.0
Flame Length (ft)                      12.0

Table 2. Total number of each species captured prior to the burn
treatment and associated capture rate.

                                                   Number     Capture
Species                                           Captured    Rate (%)

Red-sided Garter Snake (Thamnophis sirtalis)          41.0        18.6
Western Plains Garter Snake (Thamnophis radix)        23.0        10.5
Western Ribbon Snake (Thamnophis proximus)            10.0         4.5
Eastern Massasauga (Sistrurus catenatus)               9.0         4.1
Western Fox Snake (Elaphe vulpina                      7.0         3.2
Yellow-bellied Racer (Coluber constrictor)             4.0         1.8
Prairie Kingsnake (Lampropeltis calligaster)           1.0         0.5
Diamond-backed Water Snake (Nerodia rhombifer)         1.0         0.5

Total                                                 96.0        43.6

5. Literature Cited

Adams, D., R. Anderson and S. Collins. 1982. Differential response of woody and herbaceous species to summer and winter burning. Southwest Naturalist. 27:55-61.

Anderson, R.C. 1997. Summer fires. Pages 245-249 in S. Packard and C.F. Mutel editors. The tallgrass restoration handbook. Island Press, Washington, D.C.

Babbitt, L.H. and C.H. Babbitt. 1951. A herpetological study of burned-over areas in Dade County, Florida. Copeia. 1951:79.

Durbian, F.E. 2001. 2000/2001 Eastern Massasauga Rattlesnake surveys on Squaw Creek National Wildlife Refuge and associated geographic information system database. Unpublished Report. 8 pp.

Erwin, W.J. and R.H. Stasiak. 1979. Vertebrate mortality during the burning of a reestablished prairie in Nebraska. American Midland Naturalist. 101:247-249.

Frese, P.W. 2003. Tallgrass prairie amphibian and reptile assemblage fire mortality. Herpetological Review. 34:159.

Fitch, H.S. 1960. Autecology of the copperhead. University of Kansas Publication, Museum of Natural History. 13:85-288

Heinrich, M.L. and D.W. Kaufman. 1985. Herpetofauna of the Konza Prairie Research Natural Area, Kansas. Prairie Naturalist. 17:101-112.

Johnson, T.R. 2000. The amphibians and reptiles of Missouri. Missouri Department of Conservation. Jefferson City, MO. 400 pp.

Johnson, G., B. Kingsbury, R. King, C. Parent, R. Seigel, and J. Szymanski. 2000. The Eastern Massasauga Rattlesnake: A Handbook for Land Managers. U.S. Fish and Wildlife Service, Ft. Snelling, MN. 52 pp + appendices.

Missouri Natural Heritage Program. 2003. Missouri species and communities of conservation concern checklist. Missouri Department of Conservation. Jefferson City, MO. xv + 29 pp.

Seigel, R.A. 1986. Ecology and conservation of an endangered rattlesnake, Sistrurus catenatus, in Missouri, USA. Biological Conservation. 35:333-346.

Speer, R.T. 2001. Squaw Creek National Wildlife Refuge Fire Management Plan. U.S. Fish and Wildlife Service, Mound City, MO. 63 pp + appendices.

Szymanski, J. 1998. Status assessment for the eastern Massasauga (Sistrurus c. catenatus) 1998. U.S. Fish and Wildlife Service, Ft. Snelling, MN. 30 pp + appendices.