A synthesis of ecosystem management strategies for forests in the face of chronic nitrogen deposition

Abstract

Total nitrogen (N) deposition has declined in many parts of the U.S. and Europe since the 1990s. Even so, it appears that decreased N deposition alone may be insufficient to induce recovery from the impacts of decades of elevated deposition, suggesting that management interventions may be necessary to promote recovery. Here we review the effectiveness of four remediation approaches (prescribed burning, thinning, liming, carbon addition) on three indicators of recovery from N deposition (decreased soil N availability, increased soil alkalinity, increased plant diversity), focusing on literature from the U.S. We reviewed papers indexed in the Web of Science since 1996 using specific key words, extracted data on the responses to treatment along with ancillary data, and conducted a meta-analysis using a three-level variance model structure. We found 69 publications (and 2158 responses) that focused on one of these remediation treatments in the context of N deposition, but only 29 publications (and 408 responses) reported results appropriate for our meta-analysis. We found that carbon addition was the only treatment that decreased N availability (effect size: −1.80 to −1.84 across metrics), while liming, thinning, and prescribed burning all tended to increase N availability (effect sizes: +0.4 to +1.2). Only liming had a significant positive effect on soil alkalinity (+10.5%–82.2% across metrics). Only prescribed burning and thinning affected plant diversity, but with opposing and often statistically marginal effects across metrics (i.e., increased richness, decreased Shannon or Simpson diversity). Thus, it appears that no single treatment is effective in promoting recovery from N deposition, and combinations of treatments should be explored. These conclusions are based on the limited published data available, underscoring the need for more studies in forested areas and more consistent reporting suitable for meta-analyses across studies.