<
 
 
 
 
×
>
hide
You are viewing a Web site, archived on 03:49:37 Nov 01, 2004. It is now a Federal record managed by the National Archives and Records Administration.
External links, forms, and search boxes may not function within this collection.
Home


Vegetation Change in National Parks


by
Kenneth L. Cole
National Biological Service
Natural ecosystems are always changing, but recent changes in the United States have been startlingly rapid, driven by 200 years of disturbances accompanying settlement by an industrialized society. Logging, grazing, land clearing, increased or decreased frequency of fire, hunting of predators, and other changes have affected even the most remote corners of the continent. Recent trends can be better understood by comparisons with more natural past trends of change, which can be reconstructed from fossil records. Conditions before widespread impacts in a region are termed "presettlement"; conditions after the impacts are "postsettlement."
Fossil plant materials from the last few thousand years are used to study past changes in many natural areas. Pollen buried in wetlands, for example, can reveal past changes in vegetation (Faegri and Iversen 1989), and larger fossil plant parts can be studied in deserts where the fossilized plant collections of packrats, called packrat middens, have been preserved (Betancourt et al. 1990).
This article summarizes the rates of vegetation change in four national park areas over the last 5,000 years as reconstructed from fossil pollen and packrat middens. These four national park areas from different ecological regions (Fig. 1) demonstrate the flexibility of these paleoecological techniques and display similar results.

Fig. 1. Four national park units studied.

Northern Indiana Prairie
Experimental prairie fire at Indiana Dunes National Lakeshore. Courtesy K. Cole, NBS
A 4,500-year history of vegetation change was collected from Howes Prairie Marsh, a small marsh surrounded by prairie and oak savanna in the Indiana Dunes National Lakeshore near the southern tip of Lake Michigan. Only 40 km (25 mi) from Chicago, this area has been affected by numerous impacts from settlements but still supports comparably pristine tall-grass prairie vegetation as well as the endangered Karner blue butterfly (Lycaeides melissa samuelis). Although this site has experienced more disturbances than any of the others described here, it is a most valuable site because of its many species (Wilhelm 1990) and its tall-grass prairie vegetation that has been nearly eliminated elsewhere.
The many historical impacts to this area make it a good source for studying past changes. Past amounts of pollen from the primary plant taxa are illustrated in Fig. 2. Many changes occurred before settlement, but more rapid changes occurred in the last 140 years.
Fig. 2. Selected taxa of fossil pollen recorded from Howes Prairie in the Indiana Dunes. The percentage of total pollen representing each plant is graphed along a vertical time axis. The dotted line shows the sedimentary horizon representing settlement of the region (about A.D. 1850). Major changes indicated by letters: A -- decline in pine and increases in oak and grasses due to plant succession and climate change; B -- decline in pine due to logging of white pine in mid-1800's; C -- increase in ragweed from cleared farm fields and increase in fly ash from the development of the steel industry in Gary, IN (22 km away) in the late 1800's (Cole et al. 1990); D -- increase in charcoal particles as steam railroads ignite nearby drained wetlands and subsequent decline in charcoal as steam power ends and wildfires are controlled; and E -- decline in oak as frequent fires top-kill mature trees followed by increase in oak as periodic prairie fires are extinguished.
Past rates of change in vegetation can be measured by summing the relative change in each plant type between successive samples and then dividing by the number of years between samples. The technique is similar to that used by Jacobson and Grimm (1986).
Fig. 3a. Rates of change from a tall-grass prairie and oak savanna in the Indiana Dunes, IN, based on pollen from tree species.
Although these changes had been occurring throughout the last 4,500 years, the postsettlement rates of change are at least 10 times greater than the presettlement rates of change (Fig. 3a). The rates of change have been declining over the last 50 years, but are still far greater than the presettlement rates of change.

Northern Michigan Forest
Fig. 3b. Rates of change from a boreal pine forest at Pictured Rocks, MI, based on pollen from tree species.
A similar analysis was carried out on pollen from a small bog (unofficial name: 12-Mile Bog) surrounded by pine forest along the southern shore of Lake Superior (Fig. 3b). This site, within Pictured Rocks National Lakeshore, was more severely affected by logging and slash burning in the 1890's than by the periodic wildfires that characterized this forest earlier, but it has been protected for the last 80 years. The magnitude of change caused by the crude logging and slash burning of the logging era was far greater than any recorded during the 2,500 years since Lake Superior receded to create the forest of white and red pine (Pinus strobus and P. resinosa).
Coring a small pond at Pictured Rocks National Lakeshore. Courtesy W. L. Loope
As in the Indiana Dunes, rates of change have declined during the last 60 years, and the forest is now very similar to the forest of 2,000 years ago. Thus, although the area is still changing at a rate far above normal, it has begun to recover through protection.

California Coastal Sage Scrub
Fossil pollen was analyzed from an estuary on Santa Rosa Island off the coast of southern California (Cole and Liu 1994). The semi-arid landscape around the estuary is covered with coastal sage scrub, chaparral, and grassland. This site, within Channel Islands National Park, is one of the least affected areas in this region of rapidly expanding urbanization, although the island's native plants and animals were not well adapted to withstand the grazing of the large animals introduced with the ranching era of the 1800's. This island, which had no native large herbivores, became populated with thousands of sheep, cattle, horses, goats, pigs, deer, and elk. The National Park Service is removing many of the large herbivores, although most of the island remains an active cattle ranch.
Fig. 3c. Rates of change from coastal sage scrub on Santa Rosa Island, CA, based on pollen from an estuary.
All pollen types from 33 samples spread over the last 4,600 years were analyzed. The rates of change in the pollen were similar to those observed from the other sites (Fig. 3c).

Southern Utah Desert
Because fossil pollen is usually preserved in accumulating sediments of wetlands, different paleoecological techniques are necessary in arid areas. In western North America, fossil deposits left by packrats (Neotoma spp.) have proven a useful source of paleoecological data (Betancourt et al. 1990). Past desert vegetation can be reconstructed by analyzing bits of leaves, twigs, and seeds collected by these small rodents and incorporated into debris piles in rock shelters or caves. These debris piles can be collected, analyzed, and radiocarbon dated.
The vegetation history of a remote portion of Capitol Reef National Park (Hartnett Draw) was reconstructed through the analysis of eight packrat middens ranging in age from 0 to 5,450 years (Cole 1995). The vegetation remained fairly stable throughout this period until the last few hundred years. The most recent deposits contain many plants associated with overgrazed areas such as whitebark rabbitbrush (Chrysothamnus visidiflorus), snakeweed (Gutterezia sarothrae), and greasewood (Sarcobatus vermiculatus), which were not recorded at the site before settlement.
Conversely, other plants that are extremely palatable to grazing animals were present throughout the last 5,450 years, only to disappear since settlement. Plant species preferred by sheep and cattle, such as winterfat (Ceratoides lanata) and rice grass (Stipa hymenoides), disappeared entirely, while many other palatable plant species declined in abundance after 5,000 years of comparative stability.
Fig. 3d. Rates of change from desert vegetation at Capitol Reef, UT, based on plant fossils in packrat middens.
The past rates of vegetation change for this site were calculated in a manner similar to the fossil pollen records (Fig. 3d). Although the rate of change calculation is less precise than the fossil pollen records because there were fewer samples, the results show a similar pattern. The rate of vegetation change is highest between the two most recent records.
Although this area is still grazed by cattle today through grazing leases to private ranchers from the National Park Service, much of the severe damage was probably done by intensive sheep grazing during the late 1800's when the entire region was negatively affected by open- land sheep ranching. We cannot yet demonstrate whether the grazing effects are continuing or if the site is improving, although reinvasion of palatable species is unlikely in the face of even light grazing. Severe overgrazing is required to eliminate abundant palatable species, but once they are eliminated, even light grazing can prevent their restoration.

Implications
Twenty-eight thousand-year-old packrat midden at Capitol Reef National Park; orange notebook is 6 inches high. Courtesy K. L. Cole, NBS
Wise land management decisions are more likely to be made if land managers understand a site and are able to place the status quo into a historical perspective. Because most of the damage to these four sites occurred before the 20th century, land managers might not even be aware of the tremendous changes that have occurred were it not for these fossil records. Since the ultimate goal for the management of many areas is to mitigate settlement impacts and return the land to its presettlement status, detailed knowledge of the effects of settlement is imperative.
In all study areas, postsettlement rates of change were at least 10 times higher than the presettlement rates of change. Thus, the changes now being observed in even remote natural ecosystems are unlike former natural changes. Some areas are continuing to change at rapid rates, while other areas, which have not been disturbed as recently, are stabilizing. The climatic warming projected for the next 50 years may exacerbate these ongoing changes, but will be only one of many variables operating in the unplanned redesign of our natural ecosystems. Land managers need to understand the nature and severity of the effects of settlement to return the land to its presettlement condition.
For further information:
Kenneth L. Cole
National Biological Service
Cooperative Park Studies Unit
University of Minnesota
Department of Forest Resources
115 Green Hall
St. Paul, MN 55108
References
Betancourt, J.L., T.R. Van Devender, and P.S. Martin, eds. 1990. Packrat middens: the last 40,000 years of biotic change. The University of Arizona Press, Tucson. 467 pp.

Cole, K.L. 1995. A survey of the fossil packrat middens and reconstruction of the pregrazing vegetation of Capitol Reef National Park. National Park Service Res. Rep. In press.

Cole, K.L., D.R. Engstrom, R.P. Futyma, and R. Stottlemyer. 1990. Past atmospheric deposition of metals in northern Indiana measured in a peat core from Cowles Bog. Environmental Science and Technology 24:543-549.

 Cole, K.L., and G. Liu. 1994. Holocene paleoecology of an estuary on Santa Rosa Island, California. Quaternary Res. 41:326-335

Faegri, K., and J. Iversen. 1989. Textbook of pollen analysis. Wiley and Sons, New York. 328 pp.

Jacobson, G.L., Jr., and E.C. Grimm. 1986. A numerical analysis of Holocene forest and prairie vegetation in central Minnesota. Ecology 67:958-966.

Wilhelm, G.S. 1990. Special vegetation of the Indiana Dunes National Lakeshore. Indiana Dunes National Lakeshore Research Program, Rep. 90-02, Porter, IN.


Home