USGS - science for a changing world

Upper Midwest Environmental Sciences Center

The Long Term Resource Monitoring Program

An element of the Upper Mississippi River Restoration - Environmental Management Program

Ecological Assessment of High Quality UMRS Floodplain Forests

Upper Mississippi River Restoration - Environmental Management Program U. S. Army Corps of Enineers


Floodplain forests associated with the Upper Mississippi River System (UMRS) have been significantly reduced in acreage from pre-settlement levels due to timber harvesting, conversion to agricultural cropland, and other land use changes (Yin, 1997). Currently, remaining floodplain forests are significantly affected by the altered hydrological regimes associated with river impoundment. These historical and present day disturbances have contributed to long-term changes in the composition and structure of UMRS floodplain forests (USGS, 1999), as well as negatively affecting their overall health and general condition. A loss of tree species diversity, most importantly the hard mast component (e.g., oaks), has been identified as a prominent management concern in this ecosystem (USACE, 2006). A lack of successful regeneration has also been documented in many areas, which could result in further losses in forest habitat over time. This loss of forest cover and tree species diversity has been projected to continue into the future unless active management of floodplain forests can reverse this trend (Urich et al., 2002).
Floodplain forest restoration is therefore a premier ecological issue in the Upper Mississippi River System, and the lack of natural regeneration of oak and other desirable tree species (Battaglia et al., 2002) is often cited by resource managers as justification for planting targeted species in various floodplain reforestation efforts. However, tree plantings in floodplains using bare root seedlings or direct seeding have often met with mixed results (Stanturf et al., 2001; Dey et al., 2003). In the UMRS, initial assessments indicate that land managers appear to have had a great deal of success using RPM® (Root Production Method) trees for reforestation purposes (Grossman et al., 2003). Regardless of the stock used, but especially with RPM trees, tree plantings can be very expensive for land management agencies working with limited resources. It is therefore essential that methods for rapidly assessing site quality and viability for reforestation and restoration efforts be established for the UMRS. There is also a recognized need for determining when, where and how to apply active management practices in UMRS floodplain forests so as to achieve the degree of success required to justify these actions.
The use of high quality reference sites as targets for ecological restoration efforts is well documented (Allen et al., 2000). There is undeniably a need to delineate and inventory the biological and physical characteristics of remaining high quality URMS floodplain forests for restoration purposes. Complicating this issue, however, is the fact that the underlying landscape (soils, hydrology, etc.) in which UMRS floodplain forests are situated has been so radically altered from a functional standpoint that "restoration" per se may be entirely unfeasible in certain locations. For example, mast species may be entirely unsuitable for reforestation in pooled reaches immediately upriver of locks and dams due to their inability to tolerate the continuously elevated water tables encountered in those locations. Species used for reforestation purposes may therefore be best selected by their ability to inhabit specific locations regardless of whether or not they did so historically. In this respect, the use of high quality remnant forests as reference sites may be most useful when their application as such is constrained by a set of diagnostic modifiers appropriate to the functional classification of present day UMRS ecosystems (Comer et al., 2003). What remains unclear is the ability to successfully match potential vegetation communities and individual species with different environmental variables and conditions at localized and/or regional spatial scales. Related to this issue, assessing tree regeneration success is important in determining which set of environmental factors are most important to the continued survival of a diverse assemblage of floodplain tree species.


The objective of this project is to assess the physical and biological characteristics of remaining high quality floodplain forests in the Upper Mississippi River System (UMRS). This assessment will be an integral step towards establishing a suite of baseline ecological reference information that will in turn facilitate future floodplain forest restoration efforts in the UMRS. It is widely recognized among scientists and resource managers that long-term alterations in the functionality of UMRS floodplain forest ecosystems, primarily related to navigation channel management, have impacted the ecological health of these forests. For example, many UMRS floodplain forests are degrading in quality and progressing towards monotypic stands of silver maple (USGS, 1999). From a scientific perspective, the relationship between various environmental factors that affect ecosystem functionality in floodplain forests and the response of vegetation communities to long-term changes in those factors remains unclear. These relationships need to be clarified in order to facilitate floodplain forest restoration efforts. To address this issue, we first need to define and assess high quality remaining floodplain forests as reference sites to provide targets for reforestation/restoration efforts by land management agencies. Examination of remnant communities that have successful regeneration is particularly desirable as it is critical to determine the environmental requirements for floodplain species in the modern floodplain before developing restoration guidelines. This will require surveying locations with remnant forest communities that represent the desired direction of restoration efforts (e.g., mature bottomland hardwoods with an acceptable level of species diversity).
From a management perspective, agencies responsible for restoring bottomland hardwoods and other vegetation communities in the UMRS need to know which suite of environmental variables are best suited to serve as indicators for appropriate vegetation types in conjunction with the restoration process. However, successfully matching potential vegetation communities and individual species to different environmental conditions requires a stronger foundation of baseline ecological information. This project will facilitate restoration efforts by providing a means to rapidly assess the suitability of particular species to potential sites based on an assessment of site conditions. This will in the long run reduce costs by targeting specific species and/or communities to potential restoration sites, thereby avoiding costly "trial and error" approaches to restoration efforts. This project directly addresses the floodplain forest theme by incorporating the issues of loss of floodplain forest diversity, long-term alterations in ecosystem functionality, long-term effects of river management on floodplain forests, lack of natural regeneration, reforestation, restoration, active forest management, and the needs of resource managers.


Site selection: Criteria for the selection and categorization of study sites in selected portions of the UMRS and the lower Illinois River will be established (e.g., high quality forests should have an assemblage of species that is characteristic of historical precedents for that forest type in terms of composition and diversity, evidence of successful natural regeneration, and little or no evidence of impaired ecosystem functionality). It is anticipated that initial selection of high quality reference sites will likely rely to some degree on recommendations of experienced site managers and other field personnel. It is also anticipated that high quality sites may be difficult to find depending on the threshold set forth in the selection criteria. In this event it may be more feasible to include some sites that are impaired to a certain degree, but that still comprise a functional ecosystem. The inclusion of high profile sites that have been targeted for restoration planning is also a distinct possibility (e.g., Ted Shanks Conservation Area in Pool 24).

Site assessment:

For the purposes of this project, biological characteristics will refer to metrics used to describe the vegetation present on study sites. Vegetation to be sampled at each site will consist of overstory trees, understory trees and shrubs, tree regeneration (saplings and seedlings), and herbaceous species. Biological characteristics described by the study will utilize common ecological metrics such as species composition, species richness and diversity, density, overstory basal area, age, structure, and relative presence of invasive species. Physical characteristics will refer to quantitative and qualitative environmental parameters that are considered to play an important role in maintaining floodplain forest ecosystem functionality (e.g., hydrologic regime, connectivity to river, elevation, soil texture, soil total Nitrogen and OM content, photosynthetically active radiation (PAR) available to tree regeneration, and canopy openness).
Data collection related to biological and physical characteristics will utilize a nested plot protocol, replicated at each study site as logistical considerations and spatial constraints permit. Our goal is to locate a minimum of twelve sites in the UMRS that contain high quality floodplain forest, within which we will establish three permanent vegetation plots. The corners of each plot will be marked with rebar, capped with PVC, and tagged with a unique identifying plot code. Plots will be 20x50 m and will contain nested 1x1 m, 3x3 m, and 10x10 m subplots. Cover of herbaceous vegetation will be measured in the 1m2 subplots; height of woody seedlings will be measured in the 9m2 subplots; diameter at breast height (DBH) of saplings will be measured in the 100m2 subplots; and DBH of remaining trees will be measured in the remainder of the plot.

To characterize the canopy conditions in our reference sites, we will take color fisheye canopy photographs 1.5 m from the ground surface at 25 randomly selected points in each 20x50 m plot. We will use the GLA software package for calculating for calculating percent canopy openness of each image, as well as relative PAR available to understory vegetation. A hand-held light meter (AccuPAR Linear PAR/LAI ceptometer, Model PAR-80) will also be utilized to acquire real-time absolute PAR measurements at each of these points.
Eighteen soil samples will be taken from random locations within each 20x50 m plot. Six of those will be subjected to textural analysis using the Bouyoucas hydrometer method in the lab to determine percent sand, silt and clay. Six samples will be analyzed for total Carbon and Nitrogen. Soil bulk density will be determined for the remaining six samples.
Each study site will be characterized by its degree of connectivity to the river by determining its position in the landscape relative to levees and other flood control structures. Hydrologic regime at each site will be characterized by a combination of factors, including flood periodicity, relative position within pool reach (e.g., lower, middle or upper pool), and degree of fluctuation of corresponding average seasonal river elevations.
Elevational gradients are hypothesized to be a statistically significant environmental parameter influencing the composition and structure of floodplain vegetation, but consistently detailed elevational records are generally lacking throughout the URMS floodplain. Therefore, fine-scale elevation data will be obtained at each study site through the use of sophisticated survey equipment such as a Topcon Total Station.

Data analysis:

We will use multivariate statistical techniques such as non-metric multidimensional scaling to examine compositional trends in these forest communities. Then we will apply vector fitting and corresponding Monte Carlo permutation tests to test for significance of different environmental variables to the ordinations. These procedures are readily available in several commercial statistical software packages (e.g., DECODA, PRIMER, and PC-ORD). It is hypothesized that elevation, hydrological regime and soil texture (e.g., percent clay) will likely be significantly related to community trends in these ordinations. As they are also likely the most readily accessible and easily measured environmental variables in site specific locations, this will in effect serve to streamline preliminary site assessments by resource managers seeking to re-establish and/or restore vegetation as part of an environmental management program in the UMRS. At any rate, multivariate community ordination should enable us to identify a set of significant, explanatory variables that provide insight into community structure. This will be of use not only to resource managers, but also to other scientists interested in incorporating such information into models designed to facilitate future restoration efforts at local and/or regional scales.


Allen, J.A., B.D. Keeland, J.A. Stanturf, A.F. Clewell, and H.E. Kennedy, Jr. 2000 (revised 2004). A guide to bottomland hardwood restoration. USDA Forest Service General Technical Report SRS-40. 133 p.

Battaglia, L.L., Minchin, P.R., and D.W. Pritchett. 2002. Sixteen years of old-field succession and reestablishment of a bottomland hardwood forest in the Lower Mississippi Alluvial Valley. Wetlands 22: 1-17.

Comer, P., D. Faber-Langendoen, R. Evans, S. Gawler, C. Josse, G. Kittel, S. Menard, M. Pyne, M. Reid, K. Schulz, K. Snow, and J. Teague. 2003. Ecological Systems of the United States: A Working Classification of U.S. Terrestrial Systems. NatureServe, Arlington, Virginia.

Dey, D.C., J.M. Kabrick, and M.A. Gold. 2003. Tree establishment in floodplain agroforestry practices. In: Proceedings of the 8th North American Agroforestry Conference, June 22-25, Corvallis, OR. Association for Temperate Agroforestry, p. 102-115.

Grossman, B.C., M.A. Gold, and D.C. Dey. 2003. Restoration of hard mast species for wildlife in Missouri using precocious flowering oak in the Missouri River floodplain, USA. Agroforestry Systems 59: 3-10.

Stanturf, J.A., S.H. Schoenholtz, C.J. Schweitzer, and J.P. Shepard. 2001. Achieving restoration success: myths in bottomland hardwood forests. Restoration Ecology 9(2): 189-200.

U.S. Army Corps of Engineers. 2006. Upper Mississippi River System Forest Management Plan – Draft. U.S. Army Corps of Engineers: St. Paul, Rock Island and St. Louis Districts.

U.S. Geological Survey. 1999. Ecological Status and Trends of the Upper Mississippi River System 1998: A Report of the Long Term Resource Monitoring Program. USGS Upper Midwest Environmental Sciences Center, La Crosse, WI. LTRMP 99-T001. 236 p.

Urich, R., G. Swenson, and E. Nelson. 2002. Upper Mississippi and Illinois River floodplain forests: desired future and recommended actions. Upper Mississippi River Conservation Committee. 35 p.

Yin, Y., J.C. Nelson, and K.S. Lubinski. 1997. Bottomland hardwood forests along the Upper Mississippi River. Natural Areas Journal 17(2): 164-173.

Principal Investigator:

Name: John Chick
Address: Great Rivers Field Station
8450 Montclair Ave.
Brighton, IL 62012
Telephone: 618-466-9690
Fax number: 618-466-9688


Name: Lyle Guyon, National Great Rivers Research & Education Center
Name: Loretta Battaglia, Southern Illinois University – Carbondale

Accessibility FOIA Privacy Policies and Notices

Take Pride in America logo logo U.S. Department of the Interior | U.S. Geological Survey

Page Contact Information: Contacting the Upper Midwest Environmental Sciences Center
Page Last Modified: February 1, 2011