6. Concluding remarks

Both anecdotal evidence and scientific investigations have led to the suggestion that the Florida peninsula witnessed increasing summertime maximum temperatures and decreasing warm season convective rainfall during the twentieth century. This study has expanded upon the previous work of Pielke et al. (1999) by undertaking a more comprehensive set of numerical simulations with the RAMS model to further investigate the question of whether these trends can be attributed, at least in part, to land-cover change. The results of the simulations support the earlier findings that, in general, summertime maximum temperatures increased and convective rainfall decreased when natural vegetation was replaced with current land cover. When averaged over the model domain, the diurnal cycle was amplified, with higher afternoon maximum and lower nighttime minimum. This signature was especially apparent in areas of the interior peninsula that were drained of standing water and converted to agricultural land during the twentieth century. Over those areas, the model afternoon maximum temperatures warmed by as much as several degrees because of land-cover change alone. Furthermore, changing the land cover to 1993 conditions resulted in a decrease in the grid-average convective rainfall.

The analysis presented in this paper is significantly more comprehensive than that provided by the previous study in examining the relationships among land-cover change, surface fluxes, the detailed character of surface-forced circulations, diurnal cycles of shelter-level temperature, and the spatial distribution and amount of convective rainfall. In particular, a physical-dynamical link between the land-cover change and the overall regional trend of decreasing rainfall is provided. In this present work, it has been shown that changes in the land surface directly impacted the structure and strength of the modeled sea-breeze circulations. In the pre-1900 land cover scenario, the minimum in sensible heat flux over the inundated floodplain of the Kissimmee River basin was associated with an outward-directed mesoscale flow that served to constructively reinforce the sea breezes. Changing the land surface to 1993 conditions effectively removed this feature, resulting in weakening of the upward branches of the sea-breeze circulations over areas immediately on either side of the Kissimmee basin, with a weakening of the compensating subsidence over the center of the basin itself. This change in the low-level wind field was associated with an increase in precipitation over the immediate interior of the peninsula, with general decreases elsewhere. When expressed as a spatial average over the model domain, the precipitation decreased by 10% to 12% of the pre-1900 total. The changes in both the spatial distribution and the grid-average percentage decrease were remarkably consistent among all three simulated periods. These differences within a given pair of simulations were also present when the simulations were repeated with a variety of sensitivity factors, despite varying magnitudes among the resulting rainfall totals. The consistency in the changes between land-cover scenarios among the various pairs of sensitivity experiments lends greater credence to the suggestion that anthropogenic land-cover changes could be responsible for significant changes in the warm season precipitation climatology of the Florida peninsula.

Limited observational data were analyzed to provide long-term regional trends during the period spanned by the two land-cover scenarios. The observed trends indicate warming of July-August daytime maximum temperature and decreased rainfall. However, the observational trend of nighttime minimum indicates warming, which is inconsistent with the model results. It is difficult to know whether this discrepancy results from model error and uncertainty, observational error and uncertainty, or some combination of these factors.

These results could have important implications for land use and water resource management interests in south Florida, including the ongoing efforts to preserve and protect the Everglades ecosystem. The findings presented in this paper imply that restoration of a more natural flow regime and the resulting land-cover changes would alter the distribution of typical low-level wind patterns and associated convective rainfall, as well as change the surface diurnal cycle of shelter-level temperature. Furthermore, these results provide additional evidence to support the suggestion that land-cover change should be considered as a potentially significant factor in studies that provide information regarding climate trends.

Acknowledgments.

We thank David Zierden, Jim O'Brien, and Melissa Griffin of the Florida Climate Center for their assistance in providing daily and monthly rainfall and temperature data for selected surface weather stations in Florida and for their recommendations concerning anomalously wet, average, and dry rainfall years in south Florida. The contributions of John W. Jones, Tom Smith III, and Tom Cronin of the U.S. Geological Survey and Jim Irons of the NASA Goddard Space Flight Center are appreciated. Adriana Beltran and Chris Castro are credited for the implementation of the Kain-Fritsch scheme in the RAMS model. Bob Walko provided invaluable help with incorporating the USGS vegetation datasets into the RAMS model. Funding for this research was provided by USGS Grant 1434- CR-97-AG-00025, Task 7, and the USGS Geographic Research Applications Prospectus Activity.

Go back to Comparison: Model & Observed Trends | Go ahead to References