Hurricane disturbance and recovery of energy balance, CO₂ fluxes and canopy structure in a mangrove forest of the Florida Everglades

Authors

Jordan G. Barr, Everglades National Park
Vic Engel, Everglades National Park
Thomas J. Smith, U.S. Geological Survey
José D. Fuentes, The Pennsylvania State UniversityFollow

Document Type

Article

Date of this Version

2012

Citation

Agricultural and Forest Meteorology 153 (2012) 54– 66

Comments

This document is a U.S. government work and is not subject to copyright in the United States.

Abstract

Eddy covariance (EC) estimates of carbon dioxide (CO2) fluxes and energy balance are examined to investigate the functional responses of a mature mangrove forest to a disturbance generated by Hurricane Wilma on October 24, 2005 in the Florida Everglades. At the EC site, high winds from the hurricane caused nearly 100% defoliation in the upper canopy and widespread tree mortality. Soil temperatures down to −50 cm increased, and air temperature lapse rates within the forest canopy switched from statically stable to statically unstable conditions following the disturbance. Unstable conditions allowed more efficient transport of water vapor and CO₂ from the surface up to the upper canopy layer. Significant increases in latent heat fluxes (LE) and nighttime net ecosystem exchange (NEE) were also observed and sensible heat fluxes (H) as a proportion of net radiation decreased significantly in response to the disturbance. Many of these impacts persisted through much of the study period through 2009. However, local albedo and MODIS (Moderate Resolution Imaging Spectro-radiometer) data (the Enhanced Vegetation Index) indicated a substantial proportion of active leaf area recovered before the EC measurements began 1 year after the storm. Observed changes in the vertical distribution and the degree of clumping in newly emerged leaves may have affected the energy balance. Direct comparisons of daytime NEE values from before the storm and after our measurements resumed did not show substantial or consistent differences that could be attributed to the disturbance. Regression analyses on seasonal time scales were required to differentiate the storm’s impact on monthly average daytime NEE from the changes caused by interannual variability in other environmental drivers. The effects of the storm were apparent on annual time scales, and CO₂ uptake remained approximately 250 g C m⁻² yr⁻¹ lower in 2009 compared to the average annual values measured in 2004–2005. Dry season CO2 uptake was relatively more affected by the disturbance than wet season values. Complex leaf regeneration dynamics on damaged trees during ecosystem recovery are hypothesized to lead to the variable dry versus wet season impacts on daytime NEE. In contrast, nighttime CO₂ release (i.e., nighttime respiration) was consistently and significantly greater, possibly as a result of the enhanced decomposition of litter and coarse woody debris generated by the storm, and this effect was most apparent in the wet seasons compared to the dry seasons. The largest pre- and post-storm differences in NEE coincided roughly with the delayed peak in cumulative mortality of stems in 2007–2008. Across the hurricane-impacted region, cumulative tree mortality rates were also closely correlated with declines in peat surface elevation. Mangrove forest–atmosphere interactions are interpreted with respect to the damage and recovery of stand dynamics and soil accretion processes following the hurricane.