Short-term response and long-term trajectory of hurricane disturbance: Impacts of Hurricane Maria on forest biogeochemistry in Puerto Rico

Christine S. O'Connell, Environmental Studies, Macalester College, St. Paul, MN, Omar Gutierrez del Arroyo and Whendee L. Silver, Environmental Science, Policy, and Management, University of California Berkeley, Berkeley, CA, Christina Kiehl, Environmental Studies, Macalester College, Saint Paul, MN, Alicia Sadowski, University of California, Berkeley

Background/Question/Methods

Climate change is likely leading to an increase in the frequency and intensity of tropical Atlantic hurricanes. Uncertainties surround how ecosystems will respond to an increase in intense hurricanes, while changes to C and nutrient cycling could have broad-scale implications for local, regional, and global biogeochemical cycles. Hurricane Maria, a Category 4 storm, passed over Puerto Rico’s Luquillo Experimental Forest (LEF) in Sept. 2017. The storm led to widespread canopy damage and organic debris deposition on the soil surface. We used research plots from the Canopy Trimming Experiment (CTE), a hurricane-simulation experiment in the LEF that has been ongoing since 2002, as well as a sensor field measuring belowground variables to investigate how Hurricane Maria shifted biogeochemistry. We quantified the litter deposited on the soil surface as a result of Hurricane Maria (four litter collections, Nov. 2017 to Oct. 2018), conducted monthly soil sampling and recorded soil abiotic conditions. Monthly soil samples (0-10 cm depth; 1 m depth quarterly) were sampled for C, N and P variables, Fe species, and pH. Litter samples were used to calculate debris decomposition and were sampled for C and N concentrations. Soil moisture, temperature and oxygen probes tracked belowground abiotic conditions (0-15 cm).

Results/Conclusions

Hurricane Maria led to a large deposition of green leaf litter and woody debris (approximately 10x of baseline litter mass), 64% of which decomposed within 1 year of the hurricane, with the largest portion of that decomposition occurring within the first five weeks post-Maria. Non-palm leaf litter saw statistically higher decomposition rates than palm leaf litter (p<0.05). There were also significant trends in litter N over time (p<0.01), initially increasing before decreasing, but no significant differences in litter N across CTE treatment. Previously installed soil sensors showed significant drops in soil moisture (p<0.01) and increases in soil oxygen content (p<0.01) in the immediate wake of Maria, resolving within 4 weeks. However, soil temperature increases persisted for much longer, with daily maximum temperatures remaining more variable than pre-hurricane for up to six months. Soil iron and pH data suggests a reducing soil environment in the initial months after Maria, particularly in the CTE treatment (high hurricane frequency simulation plots), with implications for the biogeochemical impacts of intensifying hurricane regimes. The presence of research plots with known disturbance history that could be monitored post-Maria provide an unprecedented opportunity to test hypotheses regarding controls on biogeochemical cycling after single and repeated disturbances.