A Microbial Observatory in the Duke Forest: Exploring fungal diversity in response to environmental change

DFMO: The Duke Forest Mycological Observatory

IMAGES of some Duke Forest fungi

click here to access DFMO database (note: this link has been moved to the Duke Arts & Sciences server, includes photos, sequences, and other collection data)

Published papers & Datasets

O'brien, H. E., Parrent, J. L., Jackson, J. A., Moncalvo, J. M., Vilgalys, R., 2005. Fungal community analysis by large-scale sequencing of environmental samples. Applied and Environmental Microbiology. 71, 5544-5550.

O'Brien et al (suppl data)

The Duke Forest, lying near the eastern edge of the North Carolina piedmont plateau, is one of the nation's premier outdoor laboratories for studies on effects of environmental change, different land use histories, and the dynamics of naturally evolving forest communities.  In 2000, we established the Duke Mycological Observatory to apply DNA-based approaches for biodiversity assessment of fungi and other microbes . In conjunction with other ongoing studies in the Duke Forest (such as the Free-Air Carbon dioxide enrichment, or FACE experiment), this project is also examining fungal community structure to assess the impacts of various environmental variables such as increases in carbon dioxide concentration on fungal diversity.
Further information about the Duke Forest Mycological Observatory can be obtained from Dr. Rytas Vilgalys (fungi@duke.edu), Dept. of Biology, Duke University (or visit http://www.botany.duke.edu/fungi/mycolab).

Project summary

In many large scale ecological studies the diversity of fungi and their roles in ecosystem function and change are almost totally ignored.  Recent advances in the fields of DNA sequencing, bioinformatics, and phylogenetic analysis have made possible many advances in studies of microbial diversity.  We propose to develop a system for performing routine surveys of fungal biodiversity in Duke Forest (Durham, NC) based on molecular systematics.  The Duke Forest lies near the eastern edge of the North Carolina piedmont plateau and supports diverse plant species (>100 tree species), soils, and land use conditions and research projects currently underway include surveys of arthropod diversity, studies of the effects of environmental change including elevated carbon dioxide, different land use histories, and the dynamics of naturally evolving forest communities.

The Duke Forest Mycological Observatory (DFMO) will permit us to study those fungal communities which form the core group of saprophytic and mycorrhizal fungi in temperate forests.  As a pilot project, we initially propose to survey plots representing a range of forest types in order to establish a baseline data set of basidiomycete diversity from the forest.  We will also include plots located within several long term studies including the recently initiated Free-Air Carbon dioxide enrichment (FACE) experiment.  Early data from the FACE study indicate shifts in the microbial population in the soil, but as yet there have been no surveys of microbial diversity by FACE.  Shifts in the abundance, activity and diversity of mycorrhizal fungi are particularly likely (in response to a greater supply of root exudates in the FACE plots) and may play an important role in long term forest/ecosystem health and stability.

Our study will develop a DNA-sequence database framework (based on ribosomal ITS sequences) linked with sites within the Duke Forest.  Extensive molecular databases are already available for most fungal groups, making it possible to identify isolates to the genus in most instances and often to the species level.  We shall focus on soil, litter and mycorrhizal samples from these sites, which will provide a more unbiased estimate of microbial diversity than estimates based "above ground" sampling of fruit bodies or cultivable diversity.  These data will then form a basis for the long term monitoring of fungal microbial diversity within the Duke Forest and FACE site.

This project will provide baseline data about soil-inhabiting, mycorrhizal, soil, and litter-fungi from a variety of habitats. This project will also be one of the first studies of fungal community structure to assess the impacts of various environmental variables such as increases in carbon dioxide concentration on fungal diversity.

Summary of Objectives and Aims:

1. Protocols for DNA based fungal community analysis.  To develop a microbial observatory, a set of protocols and databases, which can provide for rapid, thorough surveys and robust identification of fungal isolates from diverse environmental sources.  This project will particularly focus on diversity in mushrooms and other basidiomycete fungi that represent a large component of forest biodiversity.

2. Molecular phylogenetic survey of Duke Forest fungi.  Combined field studies and high-throughput DNA sequencing will allow us to collect between 5,000-10,000 individual fungi during the course of this study.  This will be the first research study of the fungal community from Duke Forest, and one of the first to employ DNA sequences as the primary sampling unit.  Plans are to survey soil, litter and mycorrhizal diversity using a variety of approaches (fruiting censuses, culturing, direct sampling,) in order to obtain multiple estimates of species diversity.

3. Biodiversity informatics.  Development of informatics tools that can integrate DNA sequence collection and databasing, alignment, phylogenetic analysis, and taxonomic identification of entire fungal communities.  Substantial molecular taxonomy databases are being developed by the mycological community for many groups of fungi, with taxonomic resolution at the genus and species level in some cases.

4. Measuring biodiversity.  The combination of molecular phylogenetic and ecological approaches will allow us to estimate and quantify taxonomic and genetic diversity of fungi.  These DNA based approaches will also facilitate comparative studies of biodiversity between ecological communities in time and space.  We are also very excited about applying new approaches for estimating biodiversity that include information from molecular evolution.  The ambitious sampling effort is expected to at least triple the number of fungal species known for the Duke Forest and will provide a lower bound estimate of fungal species diversity for the Duke Forest.

Project Plan

General sampling strategy. Community surveys will include fruit body censuses, culture studies, as well as several direct environmental sampling protocols which will provide independent estimates of species diversity from each area.  Numerous studies have shown that more biodiversity can be recovered using many small samples at many well-separated sites as opposed to exhaustive sampling of a few large plots (Palmer and White, 1994; Nekola and White, 1999).  Therefore, we have chosen to utilize many small (500 mg) samples from a number of well-separated plots as part of our sampling strategy. We will use existing 0.1 hectare permanent sample plots (PSPs) that represent a range of distinctive forest types present in the Duke Forest; individual PSPs will be selected in consultation with Bob Peet of UNC (see Table 1).

Taxonomic studies.   Both formal and informal field trips to sites in Duke Forest are regularly made by members of our laboratory and local mushroom collectors.  These mushroom forays often produce valuable collections of many species, which contribute to our larger taxonomic reference database.  Plans include a series of forays/workshops in which 5-10 taxonomic experts from different fungal groups will be invited to collect and identify fungi from Duke Forest.  At least one annual mushroom foray is also sponsored by the Triangle Mushroom Club, and these will also be integrated with the DFMO.

Component studies and study sites

1. Pilot studies:  Several initial pilot studies use samples collected from forest areas adjacent to campus (Duke Division) or within Duke Campus (specifically, the "Anderson woods" area across the street from the BioSci bldg).   We are currently refining protocols for rapid high volume sampling of DNA sequences from a variety of environmental samples, including fruit bodies, soil and litter mycelia, and mycorrhizal plant roots.
2. FACE site studies. Fruiting censuses are underway for all six FACE rings.  In addition, community surveys of mycorrhiza and litter fungi will be carried out within every ring starting in early 2001, as soon as protocols are available.
3. Fungal communities.  The major effort for this project is an ambitious survey of fungal communities associated with every major forest type from Duke forest (Table 1).  Two 0.1 hecatare PSPs will be sampled for every natural forest community (total: 16 PSPs).  Collecting schedules for fruiting censuses (10-12 visits per year) and environmental sampling (3-4 collecting trips/yr) are planned.

The Duke Forest

Established in 1931, the Duke Forest consists of about 7,900 acres in the North Carolina piedmont plateau managed for sylvicultural, educational, research, and conservation purposes by the School of Forestry and Nicholas School of the Environment at Duke University (Edeburn, 1981).  The forest supports a wide variety of plant species including more than 100 tree species, soils, topography, and past land use histories. Currently these forests are known to support more than 1000 species of vascular plants (Palmer, 1990) in several distinct forest types (summarized in Table 1).

Starting in 1933, scientific research in the forest has been facilitated by the establishment of a set of permanent sample plots (PSPs, 0.1 to 1 acre) in stands of various ages and cover types.  Within these plots each tree has been individually number, identified, and a variety of measurements have been taken (position, diameter, height, crown length, etc.) and re-sampling has continued at 4-9 year intervals (most recently in 1997).  In 1977, more than 200 additional plots (0.1 hectare) were established throughout the Duke Forest which represent a sample of nearly all the environments in the forest.  These plots are valuable since they incorporate data on herbaceous species and soil chemistry along with the measurements made for the originally established PSPs.

Much of the scientific value of the Duke Forest lies in data collected from hundreds of studies and surveys completed over the history of the forest, including important early studies in ecology, botany, and mycology (Keever, 1950; Oosting, 1942; Wolf et al., 1938).  In addition, the forest has served as a resource for many more recent projects (see http://www.env.duke.edu/forest/pubs_new.html for a list).  Recent ecological studies in the Duke Forest have included the effects of various forestry practices (Benjamin, 1991), acid deposition and ozone (Richardson et al., 1992), and increased carbon dioxide concentration (DeLucia et al., 1999; Ellsworth et al., 1995; Johnson, 1999).

Table 1. Summary of the dominant forest types in the Duke Forest (source: Duke Forest Office).
Forest Type  Location
Oak - Hickory
Post Oak - White Oak
Blackjack Oak - Post Oak 
Birch - Sycamore
Yellow Poplar - Sweetgum - Ash - Elm - Red Maple
Beech - Maple - White Oak - Red Oak
Loblolly Pine 
Shortleaf Pine 
Virginia Pine
Upland climax
Dry ridges
Dry south-facing bluffs
Early successional floodplain
Stable talus slopes and above flood plains
Dry upland soils
Dry or very poor soils

Fungi of the Duke Forest

Surveys of the fungi from the Duke Forest began in the 1930s (Wolf et al., 1938) and have continued until the present day (B. Clark and R. Vilgalys unpubl.data).  These data show that the forest possesses a diverse collection of fungi from all major groups.  However, based on estimates from other well studied temperate areas such as Switzerland, Finland, and the British Isles (Hawksworth, 1991) it is obvious that the current list (about 700 species) is an underestimate, but is consistent with the state of knowledge of fungal diversity from North America in general.  Based on the data presented in Table 2, we can determine that for most well studied areas the ratio between the number of vascular plant species to the number of fungal species ranges between 1:4 and 1:5.7.  These data are consistent across the best studied areas of Europe.  When the number of vascular plant species is taken into account the expected number of species for the Duke Forest should be between 4,000 and 6,500 (84,000 to 120,000 for North America).  However, these numbers are probably significantly underestimated for two reasons: 1) in many well studied genera the number of species in North America is significantly higher than the number for Europe often approaching 6-7 times (Dennis, 1986); and 2) these estimates do not include insect associated fungi which are very poorly known and potentially very diverse.  Therefore, in a complete survey of the Duke forest we could expect to find more than 4,000 to 6,000 new fungi.  Based on our poor overall knowledge of biodiversity in the fungi, a significant portion of these could be expected to be species new to science (Hawksworth, 1991).  At a 5% recovery rate of new species, (which has been reported for diverse habitats (Petersen, 1979), we can expect to find more than 300 species new to science during this extensive survey of the Duke Forest.
Table 2. Estimated numbers of fungi for North America and the Duke Forest based on data from Hawksworth (1991) (These estimates do not include estimates for insect associated fungi).

Species of
vascular plants
Species of fungi

British Isles
North America

Duke Forest 







(11-15% known)
4,500 - 6,500 
(10-14% known)
E. Müller in Hawksworth, 1991
Rassi and Väisänen, 1987; T. Ahti in Hawksworkth, 1991
Clapham et al., 1987; S. L. Jury in Hawksworth, 1991
Flora of North America Editorial Committee, 1993; Farr et al., 1989

Wolf et al., 1938; Palmer, 1990; B. Clark, and R. Vilgalys, unpubl. data


go to the Duke Mycology home page
go to the Duke Forest home page