The Localized Influence of Fire on Historic Spruce Forest Death

The study of paleobiology provides ecologists the opportunity to see into the past how previous ecosystems in a place were influenced by environmental changes. Not only does paleobiology allow us to see into the past, but also to anticipate how future species and ecosystems will react to the rapidly changing climate today. The relationship between warming climate trends and the transition from boreal species to ones that are better suited for the new climate conditions. In the eastern United States, this shift was characterized by boreal taxa moving primarily northward where climate conditions better suited their environmental needs. However, the impact of fire on this transition is not well studied, so Jensen et al. (2021) set out to investigate the role of both fire and climate in the shift from Picea (spruce) conifer forests to broadleaf deciduous and Pinus (Pine) woodlands from the late Pleistocene to the early Holocene in the southern Great Lakes Region (Fig. 1). 

Vegetation History of the southern Great Lakes Region

The analysis of cores collected from five different kettle lakes across the southern Great Lakes region (Fig. 1) provided pollen and charcoal samples to inform the researchers about the vegetation history of each site. In their research, the first step Jensen et al. (2021) took was to define four different temporal zones based on patterns of species pollen abundance at the sites (Fig. 2). The first zone, Zone A, begins at the record onset, 19-17.5 kyr BP, to 15.8-14 kyr was dominated by Picea (spruce), with some Pinus (pine) and cold tolerant deciduous taxa including Alnus (Alders), Salix (Willows), and Betula (Birches). Transitioning into Zone B (15.8-14 to 13.3–11.6 kyr BP), a marked decline in Picea, resulting in mixed deciduous forests and woodlands, composed of waning populations of Picea and new populations of deciduous hardwood trees. Zone C (13.3–11.6 to 12–10.5 kyr BP) notably occurred during the Younger Dryas, in which all sites saw an increase in Pinus and decrease in deciduous hardwoods. During this time, some sites also saw a second Picea peak where it recovered, but this was not a universal phenomenon at all five sites. Finally, Zone D (12–10.5 to 8.8–7.8 kyr BP) had a sharp decline in Pinus and Picea (where second peaks occurred) as well as a steep  rise of temperate deciduous forests at all of the sites. Quercus (Oak) species showed a significant rise in abundance in this zone, indicating the presence of oak savannas, similar to those restored at Conard Environmental Research Area (CERA) here in Iowa.

So what’s the deal with fire?

Jensen et al. (2021) used their construction of the vegetative history at each site to identify patterns of vegetation change coinciding with signs of fire at that time. The results of combining charcoal analysis and Picea vegetative history divided the sites into two distinct patterns of Picea vegetation change. The first pattern, termed “decline and return”, was observed at Stotzel-Leis and Silver Lake sites. It is characterized by abrupt declines in Picea that coincided with periods of enhanced fire activity (Fig. 3). After that decline, deciduous forest primarily replaced Picea, with some return of Picea and Pinus during Zone C (Fig. 2). In this case, fire significantly accelerated the process of Picea replacement. The second pattern, “stair step” was observed at Appleman Lake, Bonnett Lake, and Triangle Lake Bog sites. It is distinguished by the decline in Picea through several abrupt events, however the relationship between Picea declines and fire activity is less clear (Fig. 3). Picea was primarily replaced by Pinus after the decline, and deciduous forests later on (Fig. 2). These two patterns indicate that the effects of fire vary significantly at the local level. Of the five sites studied, the two that were significantly impacted by fire had different patterns of replacement after significant fire events, pointing to fire significantly affecting what species colonize disturbed areas. 

Why Should We Care?

The rapidly changing climate due to global warming is creating new environmental circumstances that species have little time to adapt to. During the late-glacial period, the warming climate after the Last Glacial Maximum provides scientists an opportunity to see how species reacted to that warming period. Spruce forests and other boreal species habitat zones are farther North now, and may fare similarly as the global temperature continues to rise. Changing patterns of fire frequency are also associated with climate change. Jensen et al. addresses how fire has interacted with climate to produce different ecological outcomes, which can inform how areas with frequent fires will be impacted differently. 

Reference

Jensen, A. M., Fastovich, D., Watson, B. I., Gill, J. L., Jackson, S. T., Russell, J. M., ... & 

Williams, J. W. (2021). More than one way to kill a spruce forest: The role of fire and climate in the late‐glacial termination of spruce woodlands across the southern Great Lakes. Journal of Ecology, 109(1), 459-477.

The Driftless Area: Is There A Future Without Knowing The Past?

    There is a common misconception that most of the beauty and allure of the Midwest landscape lays in the awe of the Great Lakes. Well, this idea is likely fueled by the fact that most Midwest states form what is known as the “Corn Belt”. In 2007, the Midwest consisted of 20,360,396 hectares of corn and 14,277,472 hectares of soybean, while the total area of agricultural land spans well over 127 million acres. Flat and homogenous landscapes of corn and soy are quite often what comes to mind when most Americans conceptualize the imagery of the Midwest. Though this may be a predominantly true narrative, there are many relatively small ecoregions that constitute the ecological diversity of the Midwestern landscape, one of which is a distinct region referred to as the Driftless Area. Extending across 4 state borders, this region is constructed by Southwestern Wisconsin, Southeastern Minnesota, Northeast Iowa, and a very small portion of Northwestern Illinois. It is known to have been skipped over by glacial drift at the end of the last Ice Age, which saved this relatively small region from the flattening effects of glaciation notable on the surrounding landscape. Diverse and karst topography allows for the watersheds to run cold with groundwater-dominant streams that support a world-class trout fishery. To the enticement of an angler’s solitude, the extent of national publicity that this region has received is strikingly low considering that the Mississippi corridor that flows through the Driftless Area is home to over 40% of waterfowl in the US, more than 300 bird species, and 260 species of fish. Not to mention that it functions as a massive recreational resource to more than 3 million people annually (more than Yellowstone), which produces a $6.6 billion annual recreational/tourism economy.

   Consequently, agricultural and cattle operation land expansion is undoubtedly a leading factor in ecological degradation and poses the largest threat to the ecological integrity of this region, which doesn’t align with the simultaneously high ecological and ecotourist economic value previously mentioned. Many ecological restoration efforts are actively combating degradation. In hopes to protect what is left and recreate an ecologically diverse and healthy future, it is vital that we consider the impact of perspective in this reconstruction. This may go without saying, but its worth reiterating so I’ll say it here for emphasis, “in order to know where we are going, first, we must know where we have been”.

    So, let's take a look at the past. The Shea et al. (2014) study attempts to reconstruct pre-Euro colonial vegetative composition of the Driftless Area. They bluntly recognize the critical role of “reference conditions and the development of knowledge regarding the processes that contributed to development of those conditions and their variability across landscapes” (Shea et al., 2014). Public Land Survey records were used to collect mid-1800’s land condition data. The data was analyzed in two intersectional groupings:

1. Vegetation cover and structure
• i.e. whether the composition of tree cover and species resulted in the existence of:
1. Prairie (No trees)
2. Savanna (Few trees)
3. Open Forest (Open canopy; substantial sunlight penetration to ground)
4. Closed Forest (Closed canopy; very little sunlight penetration to ground)
2. Statistical analysis of cover class associations with environmental factors, specifically: 
1. Soil Texture (silt, sand, or clay)
2. Topographic Roughness (diversity in elevation of the landscape)
3. Distance From Waterway

The cover classes are important for determining the demographic composition of the vegetative distribution to determine a relative biome type throughout the region. Complimentary to these data is the importance of understanding the conditions which led to the formation of these species compositions.

After collecting the data, Shea et al. found that pre-Euro colonial Driftless Area was predominantly comprised of oak savanna with 72.6% of all recorded trees being bur oak (Quercus macrocarpa), white oak (Q. alba), and black oak (Q. velutina) and 69.4% of the area was dominated by savannas. Importantly, they also found that their work similarly reflects other larger scale statewide assessments in other studies.

As a result of their ecologically distinct small-sale study Shea et al. was able to reconstruct a more precise representation of the tree cover and species composition of the area.  In alignment with greater precision was their ability to distinguish narrower assessments of environmental factors leading to the resulted vegetative composition. For example, this precision exposed that the prevalence of oak and the diverse oak cover classes, and being distinctly associated with savannas suggest a critical presence of low intensity fire in pre-Euro colonial eras as an Indigenous ecological management technique. Relevantly, they propose that variance of fire intensity and frequency has been known to have direct impacts on tree species composition; exemplified here by the absence of fire in oak savanna leading to mesophication. 

    This leads to comparatively reference present-day land cover in the Driftless Area. We know that today ~47% of the Driftless Area is agricultural land, 13% is developed, and 34% is forested, mostly in small fragments of closed forest. So, understanding that this study, as the first of its kind for this region, is intended to further inform management decisions is vital to recognize as a tool for future restoration efforts in the Driftless Area. The largest public concern and strongest political draw would undoubtedly be trout stream management and restoration. Shea et al. has already given us a outlook at potential future impacts by suggesting that hackberry (Celtis occidentalis), boxelder (Acer negundo), and honeylocust (Gleditsia triacanthos) are likely to increase in dominance. Strong efforts have been made to remove encroaching riparian boxelder communities and replace with prairie grasses, and more recently prairie cord grass strips have been planted as riparian buffers. This one area of interest where I could see Shea et al. offering a more critical look at restorative management. Although open to further interpretation, it is vital to critically and collectively inform future restoration. 

Work Cited

- Shea, M. E., Schulte, L. A., & Palik, B. J. (2014). Reconstructing vegetation past: pre-Euro-American vegetation for the midwest driftless area, USA. Ecological Restoration, 32(4), 417-433.

Image Sources

- https://www.extension.iastate.edu/news/driftless-region-beef-conference-jan-24-25

- https://www.mnopedia.org/thing/oak-savanna

- Shea, M. E., Schulte, L. A., & Palik, B. J. (2014). Reconstructing vegetation past: pre-Euro-American vegetation for the midwest driftless area, USA. Ecological Restoration, 32(4), 417-433.

Mysteries in the Sediment: Piecing Together the History of Zizania palustris

 A woman stands on the edge of a canoe, whipping a bawa'iganaakoog against the tall stalks of manoomin (Zizania palustris or northern wild rice). Most of the grains fall onto the base of the boat, but some are lost to the estuary. The hydrophilic seeds sink into the mud and grow into next year’s crop. A man sits in the back of the canoe, moving the pair forward along the water’s edge. Later, the rice will be roasted over a fire and danced over in a ceremony. The Ojibwe tribe’s relationship with Zizania palustris dates back to their arrival when a prophecy foretold their new home where plants grow on water. Since European settlement, they’ve adjusted cultural practices with the ever-evolving landscape marked by settler disturbance. Paleo-ecologists are piecing together evidence from resilient populations of native species like this. Herbarium specimen like Henry Conard’s collections of Zizania palustris in the Grinnell College Herbarium are some of the limited accesses we have to these historical landmarks (Figure 1). 

Figure 1: Henry Conard’s collections of Zizania plaustris in the Grinnell College Herbarium. Left: 1923, Swan Lake; Right: 1922, Rice Pond, Jasper County, IA

In 2015, Nurse et al published a deep analysis of the shifting populations of Zizania palustris on the heavily trafficked St. Louis River Estuary, a waterway leading to Lake Superior between Duluth, MN and Superior, WI, since Euro-American settlement 250 years ago. For the last two centuries, economically driven industries like timber, paper and travel have held precedent over preserving natural landmarks. By the mid-1800s, excessive logging in the region caused massive soil erosion. The completion of the Duluth ship canal in 1871 and growing number of mills scattered along the river compounded the soil disturbance, adding pollution and removing viable habitat. Although the state of Minnesota declared the area contaminated in 1928, the Clean Water Act was not passed until 1972, mandating that sewage deposited in the river must be treated. 

 

Evidence of these events still exist in the estuary’s sediment. The researchers studied pollen and phytoliths from soil cores at five separate locations to determine the population shifts of Zizania palustris and reconstruct a vegetation history of the region (Figure 2).

 Figure 2: Left: Common species on the St. Louis River Estuary found in sediment collections. Right: Map of collection sites used by Nurse et al.

Poaceae pollen grains are notoriously hard to identify. The authors mention Calamagrostis canadensis (Bluejoint grass),Phragmites americanus (American reed), and Glyceria canadensis (Rattlesnake manna grass) as species easily confused with the study subject (Figure 2). The team used four herbarium specimens and eight fresh samples of Zizania palustris to develop a key for verifying pollen samples. They listed a psilate surface visible under light microscopy and polar-to-equatorial ratio of 1.2 as the best indicators of Zizania palustris pollen (Figure 3). 

 

The pollen analysis was paired with an investigation of remanent silica particles, known as phytoliths, to verify results amid sediment redistribution. The team identified inflorescence bracts as a reliable source of rondels, an identifiable morphotype of phytoliths from Zizania palustris. When the rice falls from the plant into the water, the inorganic silica from the inflorescence bracts remains in the sediment. Unfortunately, phytoliths from northern wild rice are commonly found in the exact location of a historical stand and there was no precise record for North of Clough and Pokegama Bay (Figure 3). 

 Figure 3. Left: Micrographs of phytolith morphotypes from the St. Louis River Estuary. Box A in the top right shows the rondels from Zizania palustris inflorescence bracts. Right: SEM images of pollen differentiate species (A) Calamagrostis canadensis: scabrate texture and spherical shape (B) Zizania palustris: psilate surface and polar-equatorial ratio of 1.2  

Despite mass erosion, the dual methods proved effective. The researchers found pollen in all five core samples. Sediment cores from the North Bay, Billings Park and Minnesota Point had inconstant amounts of phytoliths from Zizania palustris dating back to the 18^th century, although it did not appear in the samples from North of Clough and Pokegama Bay (Figure 2). 

 

The authors pieced together the vegetation shifts from Zizania palustris data, microcharcoal remnants, sediment accumulation and paleo-environmental evidence from other species. Evidence of increased sediment accumulation, increased microcharcoal and decreasing pine stands in the mid-1800s coincided with the arrival of settlers. The presence of Ambrosia (ragweed) pollen in the samples furthered the evidence of disturbance. As species like Pinus disappeared, pollen from hardwood trees like Betula (birch), Fraxinus (ash), and Quercus (oak) expanded. Sediment accumulation from the dying forests ran into the estuary, lowering water levels and pressuring the population of Zizania palustris. This effect was mirrored in the further development of dams and shipping canals coupled with natural events like droughts and wildfires. 

 

Environmental impacts caused by settlers reached all corners of the Midwest. The Nurse et al paper demonstrates effective species identification strategies for populations marked by mass disturbance and highlights the biological and cultural resistance of Zizania palustris. Northern wild rice stands still exist across the Midwest today and documentation of these populations is necessary to track its vitality into the future. 

Andrea M. Nurse, Euan D. Reavie, Jammi L. Ladwig, Chad L. Yost, Pollen and phytolith paleoecology in the St. Louis River Estuary, Minnesota, USA, with special consideration of Zizania palustris L., Review of Palaeobotany and Palynology, Volume 246, 2017, Pages 216-231, ISSN 0034-6667, https://doi.org/10.1016/j.revpalbo.2017.07.003.

 

Lost in the Sands of Time: Attempting to Construct Climate and Vegetation Narratives in the NWSP

 

One of the challenges researchers face in reconstructing vegetation history is understanding why vegetation changes occur. Climate changes can be inferred from evidence of vegetation changes such as fossil pollen records, but these records do not prove that climate changes alone cause vegetation to change and cannot serve as independent climate proxies (Calcote et al. 2020). Finding alternative reference points for climate would be especially useful in the Upper Midwest, where changes in forest structure are well documented but drivers of vegetative change are not entirely understood (Calcote et al. 2020). In their study, Calcote et al. compare lake level data from sediment cores against fossil pollen records to determine if lake sediment can serve as a valid independent proxy for climate and moisture levels in the Northern Wisconsin Sand Plain. 

 

 

Figure 1a. Mesic and xeric tree species found at Cheney Lake. Images courtesy Peter M. Dziuk. 

 

Figure 1b.  Cheney Lake (the white lines represent past shorelines) (Calcote et al. 2020).

           

Figure 2. Fossil pollen data from one of the sediment cores (from Calcote et al. 2020). 

 

 

Figure 3. LOI and magnetic susceptibility data from one of the sediment cores (from Calcote et al. 2020).  

The Northern Wisconsin Sand Plain (NWSP) is an area in Wisconsin that features forests with a mixture of coniferous and broadleaf trees growing in poor, sandy soils. The vegetation makeup of the NWSP responds to changes in moisture, with xeric species such as jack pine (Pinus banksiana) and red pine (Pinus resinosa) increasing during drier conditions and mesic species such as white pine (Pinus strobus) and oak (Quercus) increasing during wetter conditions (Figure 1a). Vegetation varies with soil richness and abundance of fire according to location, with fire having an important stabilizing effect on jack pine populations, as Bjorn Larson writes in this blog post. 

The authors took sediment cores from the research site, Cheney Lake, and analyzed fossil pollen of mesic and xeric species along charcoal data (Figure 2). To determine the suitability of Cheney Lake for climate records, they estimated the depth of the lake and its response to climate changes in the 20th century from photographs and past shorelines (Figure 1b). They then estimated lake level over time based on percent organic matter (through LOI), magnetic susceptibility, and 14C dating of sediment cores (Figure 3). Sediments dating from 7200 cal. years BPA to 2010 CE were studied; sections of sediment core were divided into zones based on composition and appearance, as shown in Figure 3.  

The authors found that the lake level and pollen records did not match as closely as expected. Between 3300 and 1500 cal. yr BP, for example, the lake levels (and therefore moisture levels) at Cheney Lake were high, but the percentage of fossil pollen from mesic species did not increase, and fire frequency did not decrease as expected; there was also not as much organic matter in the sediment as expected. A drought beginning around 1500 cal. yr BP led to decreased lake levels, but xeric species’ pollen percentages did not increase and fire frequency did not increase.  The authors theorize that a shift from mesic species to jack pine did not occur during the 1500 cal. yr B.P. drought at Cheney Lake because jack pines were already established there due to a lack of fire breaks. The most significant correlation at Cheney lake was a slight increase in white pine pollen at 500 cal. yr BP, when moisture increased.

Interestingly, at other locations such as Ferry Lake (the lake in Bjorn Larson’s blog post), vegetation followed the climate trends at Cheney Lake, with jack and red pine pollen percentages increasing and oak pollen percentages decreasing during drier periods. These differences between the locations could be due to differing soil composition or presence of fire breaks. Here, oak pollen percentages did not recover even after moisture levels increased after the 1500 cal. yr BP drought, which also suggests that jack pine populations remain stable after they have been established. Records of lake level in sediment were not reliable predictors of vegetation change at Cheney Lake, but they matched regional trends in climate, growing drier during the Medieval Climate Anomaly and growing wetter during the Little Ice Age. The drought at 1500 cal. yr B.P. was previously undocumented. 

The study may have not been able to construct a cohesive narrative about vegetation history out of sediment records, but it did something just as valuable: it showed that lake level changes do not consistently correlate with vegetation changes, and that local variations can greatly affect vegetation composition. It also brought up new questions—the authors introduce theories, but it is not clear why the vegetation at Cheney Lake did not match the vegetation at other sites in the NWSP. As our understanding of vegetation history develops, we will learn how to better use proxies for climate to understand the drivers of vegetative change.  

Works Cited

Calcote, R., Nevala-Plagemann, C., Lynch, E. A., & Hotchkiss, S. C. (2021). Late-Holocene climate changes linked to ecosystem shifts in the Northwest Wisconsin Sand Plain, USA. The Holocene, 31(3), 409-420. 

Image sources (for Figure 1a)

    https://www.minnesotawildflowers.info/tree/jack-pine 

    https://www.minnesotawildflowers.info/tree/northern-pin-oak

    https://www.minnesotawildflowers.info/tree/red-pine

    https://www.minnesotawildflowers.info/tree/white-pine