Nitrogen and phosphorus stoichiometry in a northern hardwood forest

Background/Question/Methods

Stoichiometric relationships between foliar nitrogen (N) and phosphorus (P) have been identified across forest ecosystems at the global scale. Although theory suggests similar relationships may occur at smaller spatial scales, a wide range of foliar N:P ratios have been observed within forest ecosystems. For other forest components  (e.g. soil, herbs), both the stoichiometric constraint  and variation in N:P is unknown. Despite this lack of knowledge, N:P ratios are frequently used to assess nutrient limitation in terrestrial ecosystems. We examined N:P stoichiometry among and within forest components (soil, soil microbes and autotrophs) of a northern hardwood forest ecosystem. We also examined whether soil N:P influences N:P of biotic components.   In 2006 and 2008, tree foliage, roots, herbaceous vegetation, soil and soil microbial biomass were sampled from single-species plots dominated by red oak (Quercus rubra), sugar maple (Acer saccharum), eastern hemlock (Tsuga canadensis), American beech (Fagus grandifolia), and yellow birch (Betula alleghaniensis) in the Catskill Mountains, NY. Half the plots were fertilized with N by adding NH₄NO₃ (50 kg ha^-1 yr^-1) from 1997 to 2008.  Samples were analyzed for total N:P and standardized major axis regression was used to detect stoichiometric relationships between N and P and between soil and biotic N:P.

Results/Conclusions

We found strong correlation between N and P across all components (r²=0.80, p<0.001), but a weaker correlation within components.  Significant, but moderate correlations between N and P were found for foliage (r²=0.22, p<0.001), roots (r²=0.31, p<0.001), herbs (r²=0.27, p<0.001), soil microbes (r²=0.34, p<0.001), and mineral soils (r²=0.39, p<0.001). N:P variation was greatest in mineral soil (C.V.=1.84), and lower in organic soil, roots, soil microbes, and foliage (mean C.V=0.24).  In unfertilized plots, autotrophic N:P was similar across components; e.g., foliar N:P averaged 34.5 ± 1.3 and roots averaged 29.3 ± 1.4. In fertilized plots, foliage (F=5.5, p=0.02), roots (F=12.9, p=0.001), and herbs (F=10.3, p=0.003) had significantly greater N:P compared to controls.  Tree species significantly affected N:P in organic soil (F=10.5, p<0.0001), roots (F=6.6, p=0.0009) and herbs (F=5.2, p=0.007) due to species differences in both N and P.  Soil N:P was not significantly correlated with autotrophic N:P (mean r²=0.08) or microbial N:P (r²=0.18). Microbial N:P was unaffected by species or fertilization and, similar to global-scale studies, appeared homeostatic. Results suggest that interspecific differences in factors like nutrient availability, and uptake contribute to variation in autotrophic N:P. At small spatial scales (<20 m)  these differences may uncouple total soil N:P from autotrophic N:P.