Part of this emphasis on aerial surveys may be because of an early rejection of the capture, then marking (tagging) and subsequent recapture to use in a population model for estimating population size along with other data (Packard and Nichols, 1983). This method is one of the better ways to determine population estimates. Many carcasses are collected each year and given a cause of death. By knowing various manatee characteristics modeled population estimates can be made and sensitivity analyses completed. These characteristics include: general proportions of tagged animals to good minimum winter counts, proportion of tagged animals at specific warm water sites from recapture of live animals, growth, maturation, reproduction and mortality characteristics such as ages and sex of the dead animals with and without tags.

MORTALITY ISSUES
Increases in the manatee population should be expected to show over time increasing numbers of dead manatees from all causes including boats (O'Shea, 1988; Eberhardt & O'Shea, 1995). Increases in boater registration should show a numerical increase for manatee deaths as well. Boaters look to the increasing minimum population estimates and wonder why more regulations are needed given the robust population changes. Environmentalists and most regulatory agencies point to the increasing boat deaths coupled with scientific uncertainty as justification for more regulations. Ackerman et al., (1995, figs 3 &4) plotted boating deaths versus registered boats and both of these variables versus time, but did not discuss these results in the context of population changes.

No statistical models are needed to show the positive changes with time for minimum manatee population and boater registration (Figure 4), boating related manatee deaths per 1000 registered boats and minimum manatee population per 1000 registered boats (Figure 5). Collinearity between population and boats is clear from Figure 4, even though these two variables are independent of one another. Figure 5 suggests that the rate of boating deaths is not changing faster than the number of registrations with respect to the population. Two other graphs with time are helpful in placing the boating related manatee deaths in perspective: minimum population counts and boating related deaths (Figure 6) and total deaths (adjusted for boating deaths) and boating related deaths (Figure 7). Figure 6 suggests that the population is growing much faster than the number of boating deaths. Figure 7 suggests that total mortality adjusted for boat-related mortality is growing faster. This change is consistent with the population growth in Figure 6. If boat-related mortality is adjusted per capita, then there is no apparent change with time (Figure 8).

Mortality information is consistent with an increasing population of manatees. If the natural mortality rate has remained relatively unchanged over the past several decades, then relative changes should reflect trends in the general population. If the perinatal death rate has remained relatively unchanged over the past several decades, then relative changes should reflect trends in the female population.

An examination of per capita manatee deaths more accurately determines relative mortality trend issues than just the use of annual numerical totals. Figure 9 shows that while the annual deaths are increasing, the annual deaths per 100 manatees are obviously not trending up. In Figure 10, annual perinatal deaths are increasing over time, but the annual deaths per 100 manatees are not trending up since about 1983. Figure 11 compares (per 100 manatees) an adjusted annual deaths (total-perinatal) with perinatal deaths. The perinatal line shows some pattern similarity and differences to the adjusted line suggesting that, perinatal deaths may be closely tied with adult mortality in some years (cold winters?, red tides?). Figure 12 shows the relationship between minimum population counts and perinatal deaths which suggest that the increase in perinatal deaths is much less than the increase in counts.

The general State-wide relationships between boater registration and boat-related manatee deaths described above appear to break down in regional areas (Wright et al., 1995). They ascribed these differences to location of the artificial warm-water sources with respect to Intracoastal Waterways, to apparent size of the local manatee populations and to number of boats in the region. Figures 13-16 show the boater registrations in the Tampa Bay area, Indian River Lagoon north of Indian River County, the Charlotte Harbor Complex and Collier County. In each of these areas, there is no clear relationship for registration and boat related deaths. Comparing between regions, the contrasts between the graphs for registrations and boating deaths appear different (all boating deaths are on the same scale).

The age distribution of carcasses (Table 3) show relatively few animals in the older age class. This distribution may be, in part, the result of an increasing population where younger animals are beginning to influence the relative abundance of older (25+) age classes. Only more data with comparisons among age frequencies may resolve such a hypothesis. Females deaths are considered more important than male deaths for population increase because of the relatively long pregnancy and calf association. Therefore, the ages at which females are dying are important to maintaining or increasing the population. Boating related deaths appear proportionally the same, by sex, for the totals. If the synoptic counts were getting better at counting and the increases in the counts just meant, not new young manatees in the population, but rather existing manatees not counted before, then the age frequency distribution should be skewed to older animals.

Manatees in the Charlotte Harbor Complex suffered a natural mortality event (but aggravated by the location of the artificial warm-water site in the Caloosahatchee River) in 1996 that was attributed to the toxic effects of a sustained red-tide bloom during the late winter. About 149 deaths during a 2-3 month period were attributed to this cause. Just prior to this event as part of the synoptic aerial survey some 469 manatees were counted in Lee County on January 10, 1996. The day before, 434 were in the vicinity of the Florida Power and Light's artificial warm water discharge into the Orange River and then to the Caloosahatchee River. During the synoptic aerial survey of 5-6 January 2001, some 498 manatees were counted in Lee County, with 434 manatees located at the warm water discharge. These data suggest that manatees recouped their losses from the 1996 red tide and all other sources of mortality in just four years.

Variables used in a general linear model were examined (Appendix 3) for relationships with boat-related deaths. This analysis demonstrate that the size of the manatee population will have a significant influence on boat-related deaths as suggested by Wright et al., (1995). However, because of collinearity or multicollinearity among boat registration, boating-related deaths, and minimum population counts the parameter estimates are unstable and have high standard errors. This collinearity or multicollinearity problems are not removed easily with satisfactory results.

CONCLUSIONS

1. All of the biological data suggests that the West Indian Manatee in Florida has a sustained increasing population. The minimum count increase has been at a rate of 6-7% per year for the last 25 years. Much of that early increase came in the absence of wide-spread speed zone regulations and later (~1991- present) large-area speed zones in some counties.

2. Increasing boat related deaths of manatees is a complex issue deserving of further analysis. A manatee per capita analysis shows a stable, but variable trend, without increasing. Increases in the population have a significant effect on the annual number of deaths. Unfortunately, Florida should expect the increasing annual numerical trend to some extent. It is not a sign of failure to adequately protect manatees, but is a positive indication of successful population growth as long as there is no sign of increasing per capita mortality rate. The perceived failure (measured only by total body count) is a repeating loop for more regulation that must be re-thought. Only the folks on the water will be the ones to influence what happens on the water. New or existing rules perceived to be erroneously based, or as part of a legal settlement without all interests' participation will lead to disrespect and failure of well-intended rules.
3. The biological status of manatees and their continued existence in Florida was unknown when they were included in the Marine Mammal Act and the Endangered Species Act by the U. S. Congress. No one knew the population level of Florida manatees when this species began its recovery, aided by City, County, State or Federal management. No one knows today what the population is in 2001. Florida should develop its own population model(s) before adding more restrictions based solely on rising boat related manatee deaths and use instead a per capita manatee evaluation.
5. A population model and estimates from such a model are needed for Florida to adopt quantitative biological goals which provides measures for healthy and sustainable population of manatees. The minimum population counts have not been considered by scientists to be useable for estimating the size of Florida's existing population. Performance criteria for existing and new regulatory zones should be established to: 1), determine types of zones needed, 2) determine effectiveness of existing zones, and 3) determine need for new zones. Additional regulatory zones should be adopted only after determining the status of the existing State manatee population.
6. A review of the information in Appendix 4 concerning listing or reclassifying of species in Florida indicates that the time for a petition to reclassify manatees in Florida is appropriate. Florida's manatee population is growing at a healthy rate, occupies virtually all habitats available and has forage and freshwater sources.
7. A vigorous capture - tag and recapture program should be implemented by the State to gain much better information about population size and change. Winter synoptic counts should be done only in cold-winter years. There is enough data available to set the criteria for triggering these counts. Mild and warm winters provide ambiguous data not useful for minimum counts or estimated future status of manatees.
8. The building of water-cooled electric generating power plants along Florida's coast line beginning in the 1950's have provided manatees with artificial winter refuges north of historical wintering areas. This change has likely reduced winter mortalities and aided increases in manatee populations. However, these areas are becoming a risk area for disease with the huge numbers of manatees that crowd together in these ?non-natural tiny ecosystems'.
9. Loss of seagrasses in Tampa Bay has not limited an astonishingly rapid increase in manatees using this clearly impacted estuary. Manatees are evidently very flexible in their ability to find food, water and reproductive sites.
10. Manatees showed a spectacular recovery in Lee County from all causes of mortality, especially the 149 red-tide related deaths in 1996, based on synoptic aerial counts in 1996 and 2001. This suggests that manatees are, and will be, resilient to effects of cold mortality, hurricanes and red-tide events in Southwest Florida. The record high State-wide count implies that all areas of the State with sustained populations are also resilient to these known causes of episodic mortality.
11. Perinatal deaths, when examined on a manatee per capita basis, are remaining stable and without evidence of an increasing rate.
12. As long as the per capita manatee deaths among the various major categories (boat-related, perinatal, and natural+undetermined) remain without positive time trends, Florida should expect manatees to continue to increase in population. Minimum manatee synoptic counts during cold winters should continue to rise.