Agriculture Reference
In-Depth Information
(Table 7.4). Native C
4
grasses that dominate tallgrass prairies of Konza LTER in
Kansas and show a strong positive response to fertilization (Clark et al. 2007) are
rare at KBS LTER, likely reflecting their absence in the surrounding landscape
(Foster 1999).
Cross-Site Analyses of Fertilization Effects on Grasslands
Many sites in the LTER Network have established and maintained long-term N
addition experiments in grasslands and similar herbaceous communities, provid-
ing opportunity for cross-site analysis of the relationship between productivity and
diversity across wide geographic and climatic gradients (Gross et al. 2000, Gough
et al. 2000, Suding et al. 2005). An initial synthesis of these data showed a uni-
modal relationship between productivity and plant species diversity across sites
(Gross et al. 2000) and that N addition had similar effects on herbaceous com-
munities ranging from Arctic heathlands to tallgrass prairie and coastal marshes,
although the magnitude of their responses differed (Gough et al. 2000, Suding et al.
2005). Although these experiments differed in sampling area, similar amounts of N
were added (10-12 g m
−2
), so it was possible to identify mechanisms that drive the
magnitude of the response to fertilization across communities (Suding et al. 2005,
Clark et al. 2007, Gough et al. 2012).
On average, N addition resulted in a 50% increase in aboveground production
and a consistent decline in species richness across sites (except for coastal marshes)
despite a broad range in initial aboveground productivity (Suding et al. 2005, Clark
et al. 2007). The magnitude of the productivity increase was strongly correlated
with the magnitude of the decrease in species richness, except in several of the
coastal marsh systems (Suding et al. 2005). Although functional groups differed in
their probability of being lost from a fertilized plot, overall species abundance in
unfertilized control plots was the strongest predictor of species loss in response to
fertilization. Species that were rare in the unfertilized community were more likely
to be excluded in fertilized plots, regardless of their functional group (Suding et al.
2005). Subsequent analyses of this dataset showed that the loss of species following
N addition was greatest in communities with lower soil cation exchange capacity,
colder regional temperature, and a larger production increase following N addition
(Clark et al. 2007).
Species composition also was an important determinant of the productivity
response, specifically the abundance of C
4
grasses (Clark et al. 2007); however, the
photosynthetic pathway (C
3
vs. C
4
) did not appear to be the causal factor (Suding
et al. 2005). In a recent meta-analysis, Gough et al. (2012) found that the form of
clonal growth (having a spreading or clumping growth form vs. nonclonal) com-
bined with height (relative position in the canopy) were strong predictors of both
species and community responses to N addition. However, neither clonality nor
height alone predicted the probability of species loss following N addition (Suding
et al. 2005).
A shift from soil resource limitation to light limitation is often assumed to be
important in determining plant species composition following nutrient enrich-
ment. However, that plant communities become less diverse with N addition,
