In the mid-1940s, nine pulp and paper and land holding companies discussed establishing a long term research area for the spruce fir (Picea-Abies) forest type so important to Maine. In 1950 these companies pooled resources and purchased land in east central Maine, about 10 miles (16 km) north of the city of Bangor. Approximately 6 miles (10 km) long and 1 mile (1.6 km) wide, the 3,800 acres (1,540 ha) were leased to the USDA Forest Service's Northern Research Station for 99 years to provide a location for long term forest management research in a mixed northern conifer forest.

The property was named the Penobscot Experimental Forest, often called simply PEF. Its history before 1950 is not well documented. What we do know is that only a small portion was cleared for agriculture or grazing, and that most of the area was cut lightly in the recent past (perhaps 20 to 40 years before 1950) for pine, hemlock and spruce sawlogs. Earlier cutting may have been heavier; presence of charcoal and old burned stumps in some areas indicate fires following cutting of pine stands. In 1950, stands on the PEF were 60 to 100 years old with a few older individual trees scattered throughout the area.

In 1994, the industrial owners donated the PEF to the University of Maine. With the donation, the owners stipulated that the mission of the Forest is "to afford a setting for long term research conducted cooperatively among USDA Forest Service scientists, University researchers and professional forest managers in Maine; to enhance forestry education of students and the public; and to demonstrate how the timber needs of society are met from a working forest." Under a formal agreement between the University of Maine and the Northern Research Station, the Forest Service maintains control of its long term research and any new research is subject to approval by a committee of university and Forest Service scientists.

The climate is cool and humid. The 30 year (1951-1980) normal (i.e. mean annual) temperature for nearby Bangor, Maine, is 43.9^oF (6.6 ^oC). February, the coldest month, has an average daily temperature of 19.3 ^oF (7.1 ^oC) while July, the warmest, averages 68.0 ^oF (20.0 ^oC). Normal precipitation is 41.7 inches (1060 mm), with 48 percent falling from May through October. Annual snowfall averages 94 inches (239 cm). Average growing season is 156 days.

Soil attributes of the PEF are primarily the result of glacial influences, consequently the soils are highly complex and variable. Wisconsin glacial till derived from fine grained, dark colored sedimentary rock forms the principal parent material. Major soil types occurring on the glacial till ridges are well drained Plaisted loams and stony loams, and moderately well drained Howland loams and sandy loams. Flat till areas between the ridges are occupied by poorly and very poorly drained Monarda and Burnham loams and silt loams. Outcroppings of vertically bedded shale covered by a thin mantle of till represent the Throndike stony and very stony loams.

Some of the lowest areas along the present water courses and in depressions are occupied by deposits of lake and marine fine sediments. Common soil types of these parent materials include moderately well drained Buxton silt loam, poorly drained Scantic silt loam, and very poorly drained Biddeford silt loam and silty clay loams.

A treeless flood plain occurs along Blackman Stream, the major water course in the PEF. Soil types here are very poorly drained Saco silt loams and fine sandy loams. Glacio fluvial sands and gravels are limited to a small esker system along Blackman Stream at the northeastern edge of the Forest. Soils on these materials are excessively drained Stetson gravelly sandy loam, well drained Stetson fine sandy loam, and well drained Machias fine sandy loam.

A few organic deposits of muck and peat occur. In general, such soils do not support forest stands, but those on the experimental forest do.

Under a recent ecological land classification, the location of the PEF is within the Central Maine Coastal and Interior Section of the Laurentian Mixed Forest Province. It is dominated by mixed northern conifers, including eastern hemlock (Tsuga canadensis); spruce, mostly red (Picea rubens)with some white (P. glauca); balsam fir (Abies balsamea); northern white cedar (Thuja occidentalis); eastern white pine (Pinus strobus); and, infrequently, tamarack (Larix laricina) or red pine (P. resinosa). The most common hardwoods are red maple (Acer rubrum); paper birch (Betula papyrifera); gray birch (B. populifolia) and aspen, both quaking (Populus tremuloides) and bigtooth (P. grandidentata). Sugar maple (A. saccharum), yellow birch (B. alleghaniensis), American beech (Fagus grandifolia), white ash (Fraxinus americana), black cherry (Prunus serotina), northern red oak (Quercus rubra), and American basswood (Tilia americana) are scattered throughout the Forest. Vegetation types are typically more diverse than the industrial spruce-fir forest farther north.

The PEF is in the Acadian forest, a region covering much of Atlantic Canada and adjacent Maine. An ecotone between boreal and broadleaf biomes, the region is dominated by mixed conifers. Red spruce is the signature species of the Acadian forest, distinguishing it from similar forests around the Great Lakes where white spruce is common and red spruce is absent. Balsam fir, a boreal species, is at its southern limit, while others trees, including eastern hemlock and eastern white pine, are at their northern limit. Stand-replacing fires are less frequent than in the boreal forest or other temperate forests. Natural disturbances are insect epidemics (notably spruce budworm) and windstorms, causing sporadic mortality. Most of the forest around Bangor has been periodically cut for high value products since the 1790s. However, little of the 4,000 ac PEF was ever cleared and cutting for 20-40 years before it became an experimental forest was light.

Road and topographic maps are available, some in digital format. A soils survey in the 1960s resulted in a map that was recently digitized. In 1988, forest stands in those areas not in long term research were typed and mapped from aerial photography and that map has also been digitized. The 1988 data were updated with field sampling in 1996.

Long-term databases are associated with silvicultural studies established in the early 1950s. Forest Service researchers measure growth, yield, stand structure and species composition in a number of treated and control areas, called compartments, covering almost 30 percent of the total PEF area. About one half of these compartments date from the 1950s, the rest have been established at various times since then.

Periodic inventories are taken on a network of approximately 580 permanent plots. These plots consist of three concentric circles, the whole plot 1/5 acre (0.08 ha) in size, an inner plot 1/20 acre (0.02 ha) and a second inner plot 1/50 acre(0.008 ha ). Tree species and quality information are recorded and diameter breast height (DBH) at 4.5 ft (1.37 m) measured on living trees. Condition of dead trees, or snags, is also recorded. All trees and snags greater than or equal to 4.5 in (11.4 cm) DBH are measured on the 1/5 acre plots. Live trees greater than or equal to 2.5 in (6.35 cm) are measured on the 1/20 acre plots. Live trees greater than or equal to 0.5 in (1.27 cm) and less than 2.5 in (6.35 cm) are measured on the 1/50 acre plots. Regeneration (living trees taller than 6 in (15 cm) but with DBH less than 0.5 in) is measured for several height classes on three milacre (4.05 m²) plots located on the circumference of the 1/20 acre plot.

Since the 1970s, individual trees have been numbered and followed. Regeneration inventories started in the 1960s. Inventories are taken every 5 years and before and after each stand entry. Volumetric data are calculated from a local volume table. All inventory data from this long term research are stored in electronic format. There are a number of additional databases from shorter-term studies overlaid on the long-term silvicultural experiment, and from the university's Forest Ecosystem Research Program (FERP), which was initiated in 1996.

Beginning in 2003 additional data collected on permanent plots have included total tree height, live crown height and crown radii for trees greater than or equal to 4.5 in. These same trees are now being stem mapped as well. Additionally, new data on compartment understory vegetation are being taken.

Research traditionally focused on management of mixed conifer stands for timber. Over the past 20 years, research broadened to include fundamental studies of tree growth and maturation, spruce budworm predation, biodiversity, root structure and function, coarse woody material, economics, and various growth and yield modeling efforts.

There are presently 46 individual Forest Service research areas comprising about 445 ha (1,100 acres) on the PEF. The largest study, a replicated experiment comparing a number of silvicultural treatments, was established between 1952 and 1957 on about 162 ha (400 acres). The experimental design is completely random; each of eight treatments is replicated twice for a total of 16 compartments. Treatments include three intensities of the selection system (5 , 10 and 20 year cutting cycles), two uniform shelterwood strategies (2 stage and 3 stage overstory removals), two diameter limit cutting methods (fixed diameter and flexible diameter removal limitations), and a clearcut harvest. In addition, two compartments were established in a "natural" area to serve as an experimental control. Although there was no evidence of recent harvesting when the controls were selected, cutting likely occurred at some time in the past.

Over the years, additional experiments have been overlaid onto the long term silviculture study. For example, studies of soils, pre commercial thinning, wildlife, vegetation, and entomology have been completed or are ongoing. These experiments all tested or are testing hypotheses about responses to silvicultural treatment.

A number of additional compartments are managed under the selection system with a 10 year cutting cycle. The purpose of these compartments is to provide areas for small scale, short term research and to keep the stands in a "state of ready" for additional long term silvicultural research.

Within a year of the PEF being leased to the Forest Service, an unreplicated demonstration was installed and has been maintained since. Called the Management Intensity Demonstration, this 40 acre (16 ha) area was divided into four equal blocks. Four silvicultural treatments are demonstrated: an unregulated "logger's choice" harvest, a periodically cut to prescribed diameter limits, a single tree and small group selection harvest every 15 years, and a similar harvest every 5 years. These stands are inventoried every 5 years with all trees over 4.5 inches (11.4 cm) DBH measured and saplings and seedlings sampled.

There are a small number of genetic test plantings on the PEF. In the 1960s, provenance tests of both eastern white pine and white spruce were planted. A half sib progeny test of white spruce was planted in 1974-75.

Faculty and students at the University of Maine account for an increasing number of areas on the PEF used for research, currently totaling about 300 acres (120 ha). The university's major effort on the PEF is the long term Forest Ecosystem Research Program (FERP). FERP is investigating effects of innovative silvicultural prescriptions on ecosystem structure and function. The silvicultural prescription is a variation of shelterwood with retention designed to mimic natural disturbance at two scales. In addition to managing existing stands, FERP treatments also include plantations of spruce and hybrid larch (Larix spp.) Response of trees, other flora, fauna and soil processes are being evaluated.

Accomplishments : PEF is recognized internationally for silvicultural research. It has the only examples of uneven-aged management of conifers in the northeastern USA and eastern Canada where cutting cycles have been sustained over decades. Pioneering research on shelterwood removals, strip cutting, and precommercial thinning was conducted. Analyses of stand growth and development and economics of various silvicultural treatments have influenced management decisions throughout the region. The range of stand composition and structure in the long-term silvicultural study attracts cooperating scientists interested in the effects of forest management on hydrology, soils, plants, and animals. Technology transfer is a major undertaking with numerous tours annually for students and professionals.

Because of ownership and close proximity to the University of Mane, most cooperators are faculty and graduate students from there. RWU-NRS-4155 is fully responsible for Forest Service research at PEF. With the exception of FERP, other PEF research is guided jointly by a team of university faculty and Forest Service scientists. Collaborators also include researchers from other Northern Research Station projects, Canadian Forest Service, University of Vermont, and University of New Hampshire, staff of state forestry and wildlife agencies, and members of the Small Woodland Owners Association of Maine.

Research opportunities are nearly unlimited in all aspects of ecology and management of conifer dominated and mixed conifer-hardwood stands of the Acadian forest. New research within the long-term silvicultural study is encouraged and facilitated by Forest Service scientists. Research in conjunction with FERP is with approval of the director of that program. All other research proposed for the PEF must be presented to the University-Forest Service team that manages the research program.

The Penobscot Experimental Forest is one of four experimental forests assigned to the Northern Research Station's Research Work Unit 4155, "Ecology and Management of Northern Forests," based in Durham, New Hampshire.

Currently all the permanent structures on the PEF are Forest Service buildings. Near the center of the Forest there is a large two story storage barn/workshop/bunkhouse constructed in the 1950s that depends on a generator for electricity. There are also two smaller buildings, one used for storing non perishable items and one to house the generator.

A 2240 ft² (208 m²) multi-function Forest Service administrative building with commercial electrical service was completed in 1997 near the entrance to the PEF. It has offices, a library/conference room, work space, bedrooms, bathrooms and a basement with workshop and storage areas. It is fully accessible to individuals with physical disabilities.

The Forest has 6 miles of good quality gravel road traversing it. There are also over 4 miles of additional gravel loop roads providing access to most of the research areas.

The PEF is located in the towns of Bradley and Eddington, Maine, across the Penobscot River from Orono and the University of Maine. State Highway 178 parallels the eastern bank of the river and provides access to the Forest. The Forest Service administration building is on the access road about 1/4 mile (0.4 km) from Highway 178. It is located at approximately 44 degrees 52.7 minutes North and 68 degrees 39.2 minutes West. Winter access beyond the Forest Service office may be limited to skis, snowshoes or snow machines.

Brissette, J.C. 1996. Effects of intensity and frequency of harvesting on abundance, stocking and composition of natural regeneration in the Acadian Forest of eastern North America. Silva Fennica 30(2):301-314.

Brissette, J.C., R.M. Frank, T.L. Stone, and T.A. Skratt. 1999. Precommercial thinning in a northern conifer stand: 18 year results. For. Chronicle 75(6): 967-972.

Frank, R.M. and J. C. Bjorkbom. 1973. A silvicultural guide for spruce-fir in the Northeast. Gen. Tech. Rep. NE-6. Broomall, PA: U.S. Department of Agriculture, Forest Service, Northeastern Forest Experiment Station. 29 p.

Frank, R.M.; T.L. Stone, H.S. Crawford. 1996. Tree/wildlife diversity responses to silvicultural practices in northern conifers. In: Proceedings of the 1995 Society of American Foresters Convention. Society of American Foresters, Bethesda, MD. p. 413-414.

Kenefic, L.S.  2002.  Research on the Penobscot Experimental Forest.  SWOAM News. 27(2):9.

Seymour, R.S. and L.S. Kenefic. 1998. Balance and sustainability in multi-aged stands: a northern conifer case study. J. For. 98(7):12-17.

Safford, L.O., R.M. Frank, E.L. Little, Jr. 1969. Trees and shrubs of the Penobscot Experimental Forest, Penobscot County, Maine. Research Paper NE-128. USDA Forest Service, Northeastern Forest Experiment Station, Upper Darby, PA 27 p.

Sendak, P.E., J.C. Brissette, and R.M. Frank. 2003. Silviculture affects composition, growth, and yield in mixed northern conifers: 40-year results from the Penobscot Experimental Forest. Can. J. For. Res. 33:2116-2128.

Seymour, R.S. 1995. The northeastern region. In: Barrett, J.W. (ed.). Regional silviculture of the United States, third edition. John Wiley & Sons, New York. p. 31-79.