We Didn't Start the Fire: a Paleoecological Approach to Inform Oak Savanna Restoration

 A historical approach helps us work towards oak savanna restoration, such as in the area of Johnson Slough in south-eastern Minnesota discussed in Spencer et al’s article ‘Using Paleoecology to Inform Land Management as Climates Change: An Example from an Oak Savanna Ecosystem’. Besides adding to the regional paleoecological knowledge base, this study sought to understand the natural range of variability of oak savanna and fire activity to evaluate oak savanna restoration and prescribed burning efforts today (such as at Krumm and Conard Environmental Research Area, where we have done plant collections for BIO 305). To understand the factors of fire activity, Spencer et al account for climate, human activity and vegetation. Spencer et al focus on the history of Johnson Slough from the mid/late Holocene period, but we also compare its history with that of Kimble pond and Sharkey lake in southeast Minnesota.

Figure 1. Johnson Slough

    The loss of the ice sheet marked the onset of the Holocene period in the region (ca. 8000 to 4000 cal year BP) (Camill et al. 2003). The mid/late Holocene period had a period of high fire activity from 6500 to 2600 cal year BP, with a shift from prairie to oak savanna occurring over this same period. 2600 cal year BP to 1963 AD, which includes the time of Euro–American settlement. Before that, there were 2.2 million hectares (ha) of oak savanna existed in the state.

This paleological history reconstructs the development and disturbance history of an oak savanna ecosystem using charcoal, pollen, and magnetic susceptibility from lake sediments. Sediment analysis techniques of fossil pollen assemblages as proxies for vegetation changes, and macroscopic charcoal as a proxy for local fire activity. Magnetic susceptibility data can be used to identify changes in depositional and erosional activity, including fire.

    Paleoclimatic records show the transition from prairie to oak savanna in the Late Holocene as it became warmer and wetter. The region was divided into zones and radiocarbon dating was used to set up periods and zones. Two separate age models were developed, one for the post-restoration period (ca. 1963 to 2010 AD), and another for the middle to late (mid/late) Holocene (ca. 8000 to 2600 cal year BP).

Figure 2. Magnetic susceptibility, charcoal concentration and Quercus: Poaceae pollen ratio at Johnson Slough. 

    Quercus macrocarpa was the dominant species and some of it was cleared by human activity in the ‘disturbed period’ reflected in the spiking in the pollen ratio in the period (Figure 2). Used for comparison with the other periods, the post-restoration period with planted oak and curbed fires gives us a comparison to analyze the shifts in vegetation composition were driven by climatic changes, and shows that fire severity was likely influenced by the vegetation. The presence of the oak savanna is most evident in a pollen ratio; low pollen ratio indicates more prairie-like vegetation and higher pollen ratio indicates more savanna-like vegetation.

    Increases in charcoal concentration within the post-restoration period are indicative of prescribed burning. Charcoal increased during humid periods as in the late Holocene; it is believed that even during humid periods, there is sufficient opportunity for the drying of grasses to cause fuel for fires. There is also data suggesting that fire burns in cooler and humid temperatures such as the late Holocene tend to produce charcoal that is more brown than black—a direction for future inquiry. Overall, with data from charcoal increase and pollen data in the mid/late Holocene, we find that there was a dominant role of climate over human interference in shift. The climate being dominant in fire is proven by other studies in the late Holocene in south-eastern Minnesota (Camill et al. 2003).

    This figure from Camille et al article on the history of Kimble pond and Sharkey lake in south east Minnesota strengthens the broader trend of the region with a similar decreasing pollen ratio to observe competition between Quercus (Oak) and Poaceae (Prairie vegetation) as pointed out by the annotated red arrows (Figure 3). The charcoal record of Johnson Slough most closely matches that of Kimble Pond. Charcoal concentrations for Johnson Slough increase at the beginning of zone 2 (ca. 6500 cal year BP), reach maximum levels within that zone and decrease toward the end (ca. 2600 cal year BP) (Figure 2). The same trend was seen in the charcoal influx data at Kimble Pond (Camill et al. 2003).

    As such, we find that the in fires associated with humid conditions and greater fuel production in late Holocene is likely what is determined to have caused prairie taxa to burn away the prairie landscape (Oaks have a lower fire severity), and it is unlikely that human interference significantly contributed to the shift.

Sources

Spencer, Jessica D, Andrea Brunelle, and Tim Hepola. “Using Paleoecology to Inform Land Management as Climates Change: An Example from an Oak Savanna Ecosystem.” Environmental management (New York) 60.6 (2017): 1090–1100. Web.

John R. Tester. “Effects of Fire Frequency on Plant Species in Oak Savanna in East-Central Minnesota.” Bulletin of the Torrey Botanical Club 123.4 (1996): 304–308. Web.

Philip Camill et al. “Late-Glacial and Holocene Climatic Effects on Fire and Vegetation Dynamics at the Prairie-Forest Ecotone in South-Central Minnesota.” The Journal of ecology 91.5 (2003): 822–836. Web.

Clark, James S et al. “Relationships Between Charcoal Particles in Air and Sediments in West-Central Siberia.” Holocene (Sevenoaks) 8.1 (1998): 19–29. Web.