Larval Dispersal Modeling Suggests Limited Ecological Connectivity Between Fjords on the West Antarctic Peninsula

Larval dispersal is a key process for community assembly and population maintenance in the marine environment, yet it is extremely difficult to measure at ecologically relevant spatio-temporal scales. We used a high-resolution hydrodynamic model and particle-tracking model to explore the dispersal of simulated larvae in a hydrographically complex region of fjords on the West Antarctic Peninsula. Modeled larvae represented two end members of dispersal potential observed in Antarctic benthos resulting from differing developmental periods and swimming behavior. For simulations of low dispersing larvae (pre-competency period = 8 days, settlement period = 15 days, swimming downward) self-recruitment within fjords was important, with no larval settlement occurring in adjacent fjords <50 km apart. For simulations of highly dispersing organisms (pre-competency period = 35-120 days, settlement period = 30-115 days, no swimming behavior), dispersal between fjords occurred when larvae were in the water column for at least 35 days, but settlement was rarely successful even for larvae spending up to 150 days in the plankton. The lack of ecological connectivity between fjords within a single spawning event suggests that these fjords harbor ecologically distinct populations in which self-recruitment may maintain populations, and genetic connectivity between fjords is likely achieved through stepping-stone dispersal. Export of larvae from natal fjord populations to the broader shelf region (>100 km distance) occurred within surface layers (<100 m depth) and was enhanced by episodic katabatic wind events that may be common in glaciomarine fjords worldwide.

Bias Correction of Climate Model’s Precipitation Using the Copula Method and Its Application in River Basin Simulation

During the last few decades, the utilization of the data from climate models in hydrological studies has increased as they can provide data in the regions that lack raw meteorological information. The data from climate models data often present biases compared to the observed data and consequently, several methods have been developed for correcting statistical biases. The present study uses the copula for modeling the dependence between the daily mean and total monthly precipitation using E-OBS data in the Mesta/Nestos river basin in order to use this relationship for the bias correction of the MPI climate model monthly precipitation. Additionally, both the non-corrected and bias corrected data are tested as they are used as the inputs to a spatial distributed hydrological model for simulating the basin runoff. The results showed that the MPI model significantly overestimates the E-OBS data while the differences are reduced sufficiently after the bias correction. The outputs from the hydrological models were proven to coincide with the precipitation analysis results and hence, the simulated discharges in the case of copula corrected data present an increased correlation with the observed flows.