Nitrogen deposition and carbon sequestration

Abstract

Nadelhoffer et al.¹ use ¹⁵N-tracer studies in nine northern forests to argue that increasing inputs of combined nitrogen from the atmosphere are unlikely to cause the increase in forest growth that has been postulated as the ‘missing sink’ for atmospheric CO₂. Only about 20% of the tracer ended up in the trees and about 70% remained in the organic and mineral layers of the soil. If only 20% of the nitrogen input from the atmosphere were available for tree growth, then not enough combined nitrogen would be coming into the northern forests each year to explain the missing carbon sink.

Main

We do not agree with this argument for several reasons. First, the ¹⁵N-tracer was applied to the soil surface in all the experiments, minimizing the opportunity for assimilation by soil organisms living in the nitrogen-poor organic litter layer. But much of the combined nitrogen received by trees from the atmosphere will be taken up directly by leaves, without ever reaching the ground, irrespective of whether it arrives in rain, as an aerosol or in gaseous form²,³. The efficiency of use of the combined nitrogen received by forests will therefore be greater than the 20% used by Nadelhoffer et al. in their calculations.

A second problem arises in the interpretation of experiments using ¹⁵N-labelled fertilizers. In unfertilized soil, assimilation of nitrogen by microorganisms that decompose nitrogen-poor plant debris is met from the unlabelled inorganic nitrogen pool. In fertilized soil, the same amount of assimi-lation occurs (assuming that assimilation is driven by microbial demand rather than the size of the pool), but some of this assimi-lated nitrogen is labelled. This means there is more unlabelled nitrogen and less labelled nitrogen left in the pool for the trees to take up, a process known as pool substitution. The net effect is that the percentage of ¹⁵N taken up by the trees underestimates the true uptake. This underestimate is more serious⁴ when the pool of unlabelled mineral nitrogen is small and when the soil contains much decomposing organic matter, as in most of the experiments of Nadelhoffer et al. A decline in the proportion of ¹⁵N recovered by plants as fertilizer application decreases is an indication⁵ that pool substitution has occurred (see their Fig. 1).

Nadelhoffer et al. may be right in saying that not enough combined nitrogen reaches the northern forests to explain the missing sink, but their analysis of ¹⁵N-labelled-fertilizer data is not proof. They claim that a maximum of 20% of the combined nitrogen reaching the forest is absorbed by trees, but we think this is a minimum value.

Finally, Nadelhoffer et al. imply that retention of carbon in tree biomass leads to longer-term sequestration than retention in soil. But this depends on the fate of the trees, as part of the carbon from tree litter entering the soil is likely to be held in soil carbon fractions that have turnover times of several centuries⁶.

See also Sievering Reply - Nadelhoffer et al.