Pleistocene diversification of unifoliolate-leaved Lupinus (Leguminosae: Papilionoideae) in Florida.

The importance and prevalence of recent ice-age and post-glacial speciation and species diversification during the Pleistocene across many organismal groups and physiographic settings are well established. However, the extent to which Pleistocene diversification can be attributed to climatic oscillations and their effects on distribution ranges and population structure remains debatable. In this study, we use morphologic, geographic and genetic (RADseq) data to document Pleistocene speciation and intra-specific diversification of the unifoliolate-leaved clade of Florida Lupinus, a small group of species largely restricted to inland and coastal sand ridges across the Florida peninsula and panhandle. Phylogenetic and demographic analyses alongside morphological and geographic evidence suggest that recent speciation and intra-specific divergence within this clade were driven by a combination of non-adaptive allopatric divergence caused by edaphic niche conservatism and opportunities presented by the emergence of new post-glacial sand ridge habitats. These results highlight the central importance of even modest geographic isolation and short periods of allopatric divergence following range expansion in the emergence of new taxa and add to the growing evidence that Pleistocene climatic oscillations may contribute to rapid diversification in a myriad of physiographic settings. Furthermore, our results shed new light on long-standing taxonomic debate surrounding the number of species in the Florida unifoliate Lupinus clade providing support for recognition of five species and a set of intra-specific variants. The important conservation implications for the narrowly restricted, highly endangered species Lupinus aridorum, which we show to be genetically distinct from its sister species Lupinus westianus, are discussed.

Assessing the success of hydrological restoration in two conservation easements within Central Florida ranchland.

In the USA, the United States Department of Agriculture (USDA) Natural Resource Conservation Service (NRCS) has restored millions of acres of wetlands through its Wetland Reserve Easement (WRE) programs. However, few quantitative studies have explored whether WREs have enhanced wetland hydrology and wetland plant communities. Additionally, USDA Compatible Use Permits for cattle grazing and other management practices are sometimes issued for WREs, but little is known about potential benefits/detriments of such practice on the success of wetland restoration. In this study, we tested if hydrological restoration of previously drained species poor pastures increased water depth and hydroperiod. Restoration involved plugging key ditches, adding water control structures and a berm. We also tested if hydrological restoration increased plant diversity (alpha and beta), floristic quality (using coefficient of conservatism) and increased the cover of wetland species (using species wetland status). Finally, we tested if cattle grazing had an effect on the success of restoration by comparing grazed plots to fenced plots. We studied two conservation easements (a total of 748 acres) located on semi-native pastures in central Florida, USA. We monitored vegetation using permanent transects stratified by vegetation type before (2004-2005) and after (2012) the restoration (2008). We assessed wetland hydroperiod using groundwater wells set up in 2003 and located within and outside the boundaries of these two easements. We used linear mixed models and multivariate analyses to compare vegetation and hydroperiods pre- and post-restoration. Number of flooded days increased following restoration in one of the easements, but we did not detect significant changes in hydrology in the other easement. Floristic quality, beta diversity and cover of obligate wetland species increased in both conservation easements and in most vegetation types. These vegetation changes were likely due to restoration activities since annual rainfall was not significantly different pre- and post-restoration. Cattle grazing did not have a negative or positive effect on the success of restoration, nor did we detect any positive effect of grazing on the success of restoration. Overall, our study shows that hydrological restoration can enhance wetland hydroperiod, water depth and wetland vegetation, but more resources should be allocated to short- and long-term monitoring of the restoration success.

Seasonality of fire weather strongly influences fire regimes in South Florida savanna-grassland landscapes.

Fire seasonality, an important characteristic of fire regimes, commonly is delineated using seasons based on single weather variables (rainfall or temperature). We used nonparametric cluster analyses of a 17-year (1993-2009) data set of weather variables that influence likelihoods and spread of fires (relative humidity, air temperature, solar radiation, wind speed, soil moisture) to explore seasonality of fire in pine savanna-grassland landscapes at the Avon Park Air Force Range in southern Florida. A four-variable, three-season model explained more variation within fire weather variables than models with more seasons. The three-season model also delineated intra-annual timing of fire more accurately than a conventional rainfall-based two-season model. Two seasons coincided roughly with dry and wet seasons based on rainfall. The third season, which we labeled the fire season, occurred between dry and wet seasons and was characterized by fire-promoting conditions present annually: drought, intense solar radiation, low humidity, and warm air temperatures. Fine fuels consisting of variable combinations of pyrogenic pine needles, abundant C4 grasses, and flammable shrubs, coupled with low soil moisture, and lightning ignitions early in the fire season facilitate natural landscape-scale wildfires that burn uplands and across wetlands. We related our three season model to fires with different ignition sources (lightning, military missions, and prescribed fires) over a 13-year period with fire records (1997-2009). Largest wildfires originate from lightning and military ignitions that occur within the early fire season substantially prior to the peak of lightning strikes in the wet season. Prescribed ignitions, in contrast, largely occur outside the fire season. Our delineation of a pronounced fire season provides insight into the extent to which different human-derived fire regimes mimic lightning fire regimes. Delineation of a fire season associated with timing of natural lightning ignitions should be useful as a basis for ecological fire management of humid savanna-grassland landscapes worldwide.

Fine-scale, spatial and temporal assessment methods for feral swine disturbances to sensitive plant communities in south-central Florida.

Feral swine are estimated to annually cost hundreds of millions of dollars in economic loss to property and agriculture in the USA, while their ecological consequences remain largely unmeasured. Using submeter-accurate Global Positioning System technology over a multiyear project, we are quantifying in a novel way the spatial and temporal attributes of swine rooting damage within 587 ha of ecologically sensitive wetland plant communities at Avon Park Air Force Range in south-central Florida. We delineated damage polygons from 0.0023 to 4,335 m(2) and were able to document recurrent damage through time at most sites during each assessment. For each polygon, we also estimated the age of damage and assigned to it a severity index, qualities of the rooting in which we detected changes in proportions over time. Spatially explicit damage assessments at fine scales conducted over several years can assist land managers in determining effects of rooting on rare plant populations, and will allow investigators to hypothesize what factors are driving patterns of this disturbance across ecologically sensitive plant communities.

Impacts from control operations on a recreationally hunted feral swine population at a large military installation in Florida.

Feral swine were targeted for control at Avon Park Air Force Range in south-central Florida to avert damage to sensitive wetland habitats on the 40,000-ha base. We conducted a 5-year study to assess impacts from control to this population that had been recreationally hunted for many years. Control was initiated in early 2009. The feral swine population was monitored from 2008 to 2012 using a passive tracking index (PTI) during the dry and wet seasons and using recreational hunter take rates from the dry season. All three indices showed substantial feral swine declines after implementing control, with indices leveling for the final two study years. Military missions and recreational hunting seasons impacted temporal and spatial consistency of control application, thereby limiting further impacts of control efforts on the feral swine population. The PTI was also able to monitor coyotes, another invasive species on the base, and detect Florida black bear and Florida panther, species of particular concern.

A structural equation model analysis of relationships among ENSO, seasonal descriptors and wildfires.

Seasonality drives ecological processes through networks of forcings, and the resultant complexity requires creative approaches for modeling to be successful. Recently ecologists and climatologists have developed sophisticated methods for fully describing seasons. However, to date the relationships among the variables produced by these methods have not been analyzed as networks, but rather with simple univariate statistics. In this manuscript we used structural equation modeling (SEM) to analyze a proposed causal network describing seasonality of rainfall for a site in south-central Florida. We also described how this network was influenced by the El Niño-Southern Oscillation (ENSO), and how the network in turn affected the site's wildfire regime. Our models indicated that wet and dry seasons starting later in the year (or ending earlier) were shorter and had less rainfall. El Niño conditions increased dry season rainfall, and via this effect decreased the consistency of that season's drying trend. El Niño conditions also negatively influenced how consistent the moistening trend was during the wet season, but in this case the effect was direct and did not route through rainfall. In modeling wildfires, our models showed that area burned was indirectly influenced by ENSO via its effect on dry season rainfall. Area burned was also indirectly reduced when the wet season had consistent rainfall, as such wet seasons allowed fewer wildfires in subsequent fire seasons. Overall area burned at the study site was estimated with high accuracy (R (2) score = 0.63). In summary, we found that by using SEMs, we were able to clearly describe causal patterns involving seasonal climate, ENSO and wildfire. We propose that similar approaches could be effectively applied to other sites where seasonality exerts strong and complex forcings on ecological processes.

Optimizing line intercept sampling and estimation for feral swine damage levels in ecologically sensitive wetland plant communities.

Ecological sampling can be labor intensive, and logistically impractical in certain environments. We optimize line intercept sampling and compare estimation methods for assessing feral swine damage within fragile wetland ecosystems in Florida. Sensitive wetland sites, and the swine damage within them, were mapped using GPS technology. Evenly spaced parallel transect lines were simulated across a digital map of each site. The length of each transect and total swine damage under each transect were measured and percent swine damage within each site was estimated by two methods. The total length method (TLM) combined all transects as a single long transect, dividing the sum of all damage lengths across all transects by the combined length of all transect lines. The equal weight method (EWM) calculated the damage proportion for each transect line and averaged these proportions across all transects. Estimation was evaluated using transect spacings of 1, 3, 5, 10, 15, and 20 m. Based on relative root mean squared error and relative bias measures, the TLM produced higher quality estimates than EWM at all transect spacings. Estimation quality decreased as transect spacing increased, especially for TLM. Estimation quality also increased as the true proportion of swine damage increased. Diminishing improvements in estimation quality as transect spacings decreased suggested 5 m as an optimal tradeoff between estimation quality and labor. An inter-transect spacing of 5 m with TLM estimation appeared an optimal starting point when designing a plan for estimating swine damage, with practical, logistical, economic considerations determining final design details.