Agriculture Reference
In-Depth Information
(Hatfield et al. 2011). When rainfall is adequate, KBS LTER corn yields are ~12.5
Mg ha
−1
at standard 15% moisture (199 bu acre
−1
). In most years, however, yields
are constrained by rainfall, as they were during a 2008 local drought when rain-fed
corn yields in the Resource Gradient Experiment (Fig. 1.9) were only 7.5 Mg ha
−1
(120 bu acre
−1
), as compared to irrigated yields of 12.5 Mg ha
−1
(199 bu acre
−1
)
and national average corn yields of 9.7 Mg ha
−1
(155 bu acre
−1
). In contrast, 2011
saw favorable precipitation; rain-fed corn yields were 13.4 Mg ha
−1
(215 bu acre
−1
)
against a national average of 9.1 Mg ha
−1
(147 bu acre
−1
) and with less response to
irrigation (Fig. 1.9).
Over all years during the 1989-2009 period, MCSE corn yields averaged 6.4 ±
0.7 Mg ha
−1
(102 bu acre
−1
). This is lower than county (7.4 ± 0.2 Mg ha
−1
; 118 bu
acre
−1
) and national (8.4 ± 0.2 Mg ha
−1
; 134 bu acre
−1
) averages for the same period.
For 4 of the 9 MCSE corn years in this period, yields were at or above county and
national yields; for 3 years, corn yields were not significantly different from (but
lower than) county and national yields; and for 2 years, corn yields were signifi-
cantly lower than county and national averages. However, both county and national
yields include those from irrigated acreage, which inflate yield comparisons rela-
tive to rain-fed MCSE yields. In Kalamazoo County about 38% of corn acreage is
irrigated and, nationally, about 15% (NASS 2012b). Overall KBS LTER corn yields
and variability are thus fairly typical of those experienced by rain-fed Kalamazoo
County growers, and reflect how rain-fed corn will vary with the year-to-year vari-
ability in growing season rainfall that is typical for farms within the region.
Landscape and Regional Observations
As noted earlier, certain important ecosystem services that may not be evident
at the field scale emerge at the scale of landscapes. Prominent examples include
biodiversity-mediated services that require landscape-level habitat configura-
tions (Gardiner et al. 2009) and recreational and aesthetic services that emerge
from a landscape of varied vegetation and topography (Bolund and Hunhammar
1999, Swinton et al. 2015a, Chapter 3 in this volume). Likewise, the provision of
high-quality water is an important service delivered by well-managed agricultural
landscapes.
Experiments and observation networks designed to address landscape-level
questions are by necessity specialized and do not lend themselves to a one-size-fits-
all design (Robertson et al. 2007). Biogeochemical questions, for example, may
require a diversity of flow paths and discrete watersheds to address (e.g., Hamilton
et al. 2007). In contrast, questions about insect biodiversity may require a multi-
county region that includes a variety of landscape patterns, crop rotations, or inten-
sities (e.g., Landis et al. 2008, Landis and Gage 2015, Chapter 8 in this volume).
And economic questions may require a social or market setting that encompasses
scales from the regional (e.g., Jolejole 2009, Chen 2010, Ma et al. 2012) to the
national (e.g., James et al. 2010) and international.
Consequently, there is no single landscape scale that is the focus for KBS LTER
landscape-level research. Rather, our landscape research setting expands outward
