Agriculture Reference
In-Depth Information
Transition to forest is well under way in the remaining two sites, abandoned
from agriculture in 1963 and 1964; overstory dominants reflect nearby late
successional deciduous forests and understory dominants include the invasive
shrubs oriental bittersweet (
Celastrus orbiculatus
Thunb.) and glossy
buckthorn (
Rhamnus frangula
L.). KBS LTER sampling began in 1993.
• Late successional Deciduous Forest (DF) stands comprise the endpoint of the
management intensity gradient. Soils of these three hardwood forest reference
sites have never been plowed. Overstory dominants (>10% biomass) are the
native trees red oak (
Quercus rubra
L.), pignut hickory (
Carya glabra
Mill.),
and white oak (
Q. alba
L.); also present are black cherry (
Prunus serotina
Ehrh.), red maple (
Acer rubrum
L.), and sugar maple (
Acer saccharum
Marshall). Understory vegetation is patchy in nature and includes a variety
of native forbs as well as some exotic species such as the shrubs honeysuckle
(
Lonicera
spp. L.) and common buckthorn (
Rhamnus cathartica
L.), the
woody vine oriental bittersweet (
Celastrus orbiculatus
Thunb), and the
increasingly invasive forb garlic mustard (
Alliaria petiolata
M. Bieb.). Two
of the three replicate sites have never been logged, while one was cut prior to
1900 and allowed to regrow. KBS LTER sampling started in 1993.
All MCSE systems and communities are replicated and most are within the same
60-ha experimental area, known as the LTER main site (Fig. 1.2); others, which for
historical or size reasons could not be included in the main site layout, are on the
same soil series within 1.5 km of the other plots (Fig. 1.3). Within the LTER main
site are the four annual cropping systems, the Alfalfa and hybrid Poplar perennial
cropping systems, and the Early Successional community. All are replicated as 1-ha
plots in six blocks of a randomized complete block design (Fig. 1.2), for a total of
42 plots with blocks determined on the basis of an initial analysis of spatial vari-
ability in soils across the site (Robertson et al. 1997).
The Mown Grassland (never tilled) community is located about 200 m to the
south of the LTER main site (Fig. 1.3); four replicated 15 × 30 m plots are located
within a larger 1-ha area of the 10-ha former woodlot. The planted Coniferous
Forests, the Mid-successional communities, and the late successional Deciduous
Forests are each replicated three times in the landscape around the main site
(Fig. 1.3). Within each replicated system, the sampling area is embedded within a
larger area of similar vegetation and land-use history.
Plot sizes for MCSE systems in the main site (Fig. 1.3) are large (90 × 110
m = 1 ha) relative to plot sizes in most agronomic field experiments (Robertson
et al. 2007). By adopting a 1-ha (2.5-acre) plot size, we encompass more of the
spatial variability encountered in local landscapes (Robertson et al. 1997). This
provides greater assurance that patterns discovered are relevant for more than a
single landscape position and avoids statistical problems associated with spatial
autocorrelation. Large plots also (1) allow the use of commercial-scale rather
than plot-scale farm equipment, helping to ensure that agronomic practices are
as similar as possible to those used by farmers; (2) help to ensure the integrity
of long-term sampling by avoiding the danger of sampling the same locations
multiple times years apart; and (3) avoid some of the scale effects associated
