Seasonal Warm-Water Refuge and Sanctuary Usage by the Florida Manatee (Trichechus manatus latirostris) in

Kings Bay, Citrus County, Florida

By Danielle C. Sattelberger, Joyce M. Kleen, Aarin-Conrad Allen, and Richard O. Flamm

Florida ManateeTrichechus manatus latirostris

•Subspecies of the West Indian manatee (Trichechus manatus)

•Inhabit shallow coastal bays and estuaries throughout Florida

•Status: endangered

•Susceptible to natural and human-related threats

• Cold stress

• Watercraft strikes

• Harmful algal blooms (HABs)

•K-selected species

Reliance on Warm-Water Refugia

•Northernmost habitat range for any species in the Family Trichechidae

• Seasonal thermoregulatory need

•Physiology limits ability to thermoregulate in cold waters

•Water temperatures < 20° C may result in hypothermia, frostbite, pneumonia, immune system suppression, or a combination of all these factors

• Cold stress syndrome (CSS)

•Unusually cold winters in 2009 and 2010

• 282 manatees perished due to cold stress

Critical Habitat: Warm-Water Refugia•Summer range extends along the Atlantic and Gulf coasts

•In winter, Florida manatees seek thermal refuge in artificial and natural sources:

Power plant discharge canals Artesian springs

Artificial Problem•Aging power plants in Florida are being closedand torn down

• Manatees susceptible to cold temperatures

•More manatees dependent on natural sources

of warm water

Critical Winter Habitat:Crystal River•Natural aquifer: Crystal River, Citrus County, FL

•Popular area for fishing, boating and other recreation

•Home to the highest population of manatees in Florida during winter months

Crystal River

•Manatee habitat coincides with recreational areas

•Only area to allow manatee swim-with tours

• Over 100,000 people visit each year

•Dozens of manatees located outside sanctuary boundaries

•As a result, manatees can become subject to harassment

Study Site: Kings Bay•Located in Crystal River, Citrus County, FL

•Springs are the headwaters of Crystal River which discharge into the Gulf of Mexico

Kings Bay

•1983 – Crystal River National Wildlife Refuge (CRNWR) established

•Refuge manages 40 acres of manatee sanctuaries

•Encompasses > 70 natural springs

• 72°F water

• Primary spring: King Spring

•Manatees migrate once Gulf of Mexico

temperatures drop below 68°F

•Wintering manatee population has increased, reaching 758 individuals in 2016, making it the largest winter aggregation of manatees at a natural warm-water source

Manatee Protection in Kings Bay•The U.S. Fish and Wildlife Service provides three levels of protection for manatees within Kings Bay:

• A National Wildlife Refuge: federally owned lands and waters set aside to conserve fish, wildlife, plants and their habitats

• A Manatee Refuge: some activities in the water may be restricted to prevent the taking (by harm or harassment) of one or more manatees.

• A Manatee Sanctuary: a federally designated area which restricts all activities, including human entry, during the winter months. The sanctuaries in Kings Bay are in effect from November 15 through March 31. Seven of these sanctuaries are in place each year.

•The State of Florida also provides protection by establishing and enforcing slow and idle speed zones in public waterways

Kings Bay Sanctuaries1. Buzzard Island

2. Magnolia Springs

3. Three Sisters Springs

4. Tarpon Springs

5. Banana Island

6. Sunset Shores

7. Warden Key

Objectives

1) Compare manatee distribution and abundance by season and over time

2) Compare the distribution and abundance results with current manatee regulations and identify possible deficiencies in protection, and

(3) Present recommendations for enhancing manatee protection

MethodsAerial Survey Counts

Spatial Analysis

Statistical Analysis

Methods: Aerial Survey Counts

•Primary approach to assessing population trends

• Obtain minimum population counts• Environmental conditions and the behavior of themanatees influence the number detected

•CRNWR staff have conducted weekly or biweeklyaerial surveys year round since 1983

•Manatee locations plotted with the count of animals at each location

•Data provided by the USFWS - CRNWR headquarters under Permit No. 13005

Methods: Spatial Analysis•GIS used to analyze and interpret data to understand relationships, patterns, and trends

•Spatial analysis performed in ArcInfo version 10.2.2

•Manatee locations as noted by jpg files and GPS locations were exported as point shape files

•In total, 711 surveys from 1983 – 2012 containing 18,793 points were mapped

• Each point = number of manatees in a given location of close proximity

• A group of manatees considered as a single sighting if less than 50m apart

•Group sizes ranged from 1 – 212 manatees

Methods: Spatial Analysis•Data organized into two periods to reflect seasonal shifts:

• Winter (November 15 – March 31)

• Summer (April 1 – November 14)

•Layers were created for each season per year and for all winters and summers (1983 – 2012)

•Environmental and biological data included as attributes

• Date, time, water temperature, air temperature, tide, water clarity, mating, number of adults, and number of calves

•Seven existing manatee sanctuaries and major springs in Kings Bay also projected

Methods: Spatial Analysis

•Kernel density analysis performed to generate areas of high manatee use

• Areas of high and low density are depicted by color intensity

•Output cell size = 2, search radius = 120 sq. meters

•Manatee density reported in square meters

•Existing and future manatee sanctuaries reported as appropriate in areas shown to be occupied by a high to medium abundance of manatees with continued seasonal use and areas with high rates of increase

Methods: Statistical Analysis•Average and peak counts determined for each season

•Simple linear regression

• Assess changes over time

•Generalized Linear Model (GLM)

• Poisson distribution

• Overdispersion

• Detect significant differences in manatee density among sanctuaries and year

ResultsSpatial Analysis

Statistical Analysis

Results: Spatial Analysis

WINTER

• Dense clusters at King Spring, Magnolia Springs, and Three Sisters Springs

• Overlap with existing sanctuaries

• Largest and densest cluster surrounding King Spring

• 0.53 – 0.58 manatees per sq. meter

• Not fully within sanctuary boundaries

• Tarpon Springs, Warden Key, and Buzzard Island do not support a high density of manatees

Kernel density map showing relative densities of manatee group sightings in Kings Bay during winter surveys from 1983-2012

Results: Spatial Analysis

• Consistent pattern of seasonal use across the 30 years

• Highest densities among Banana Island, Magnolia Springs, and Three Sisters Springs sanctuaries

Examples of kernel density maps showing the relative densities of manatee group sightings in Kings Bay during winter periods over time

Results: Spatial Analysis

SUMMER

• Wide distribution

• Dense concentration (0.025 – 0.027 manatees per sq. meter) occupied King Spring

• Dense cluster (0.016 – 0.018 manatees per sq. meter) detected northwest of Parker Island

• Near the Parker Island North Spring complex

Kernel density map showing relative densities of manatee group sightings in Kings Bay during summer surveys from 1983-2012

Results: Spatial Analysis

SUMMER

• No consistent seasonal pattern when analyzing each season individually

• Manatees ranged far beyond the high density areas depicted in the kernel density analysis of all summers from 1983 to 2012

Examples of kernel density maps showing the relative densities of manatee group sightings in Kings Bay during summer periods over time

Results:Statistical Analysis

WINTER

• A total of 58,535 sightings were recorded over 412 surveys

• Significant relationship between manatee abundance and time (p < 0.05)

• Significant increases in both average and peak observed manatee counts

• Average: R = 0.81, p < 0.001• Peak: R = 0.8, p < 0.001

• Largest influx occurred in 2009 with 566 manatees

• Annual average rate of increase of total number of manatees = 7%

• Average manatee counts increased by 4.81 animals per year

y = 4.8175x - 9471.6R² = 0.6633

y = 10.523x - 20750R² = 0.6418

0

50

100

150

200

250

300

350

400

450

500

550

600M

anat

ee C

ou

nt

Year

AVERAGE

PEAK

Linear (AVERAGE)

Linear (PEAK)

Peak and average manatee counts in winter from 1983 through 2012 in Kings Bay, Florida

Results:Statistical Analysis

SUMMER

• A total of 13,792 sightings were recorded over 299 surveys

• No significant relationship between manatee abundance and time (p >0.05)

• No significant increases in average and peak observed manatee counts

• Average: R = 0.05, p = 0.76

• Peak: R = 0.24, p = 0.19

• Large fluctuations in counts, however, average manatee counts increased by only 0.109 animals per year

y = 0.1094x - 169.92R² = 0.0033

y = 1.4129x - 2711R² = 0.0607

0

20

40

60

80

100

120

140

160

180

200

220

240

260

Man

atee

Co

un

t

Year

AVERAGE

PEAK

Linear (AVERAGE)

Linear (PEAK)

Peak and average manatee counts in summer from 1983 through 2012 in Kings Bay, Florida

Results:Statistical Analysis

Generalized Linear Model

• Significant relationship between sanctuaries and manatee density (p < 0.001)

• Significant heterogeneity in manatee density among the Magnolia Springs (p < 0.0001), Banana Island (p = 0.01), and Three Sisters Springs (p < 0.0001) sanctuaries

Variables in the model L-R Chi-Square Significance

Dependent Variable

Manatee Density

Independent variable

Sanctuary

Buzzard Island 3.3746 0.0662

Magnolia Springs 22.0881 <.0001*

Tarpon Springs 2.9617 0.0853

Banana Island 6.6263 0.01*

Sunset Shores 0.0222 0.8816

Three Sisters Springs 917.3286 <.0001*

*Indicates a significant variable

Results of the generalized linear model relating manatee density (dependent variable) to sanctuaries and year (independent variables). Significant variables indicate significant differences in densities among sanctuaries and years across the 30 year timespan.

Results:Statistical Analysis

• Significant relationship between manatee density and year (p < 0.001)

• Significant years: 1997-1998, 1998-1999, 2000-2001, 2003-2004, 2006-2007, 2009-2010, 2010-2011, and 2011-2012

• Coincided with severe winter events

• Exhibited higher manatee densities

Variables in the model L-R Chi-Square Significance

Year

1983-1984 0.0491 0.82471984-1985 0.5085 0.47581985-1986 0.1033 0.74791986-1987 0.0768 0.78171987-1988 0.0831 0.77311988-1989 0.1284 0.72011989-1990 0.0286 0.86571990-1991 0.0017 0.96751991-1992 0.0709 0.791992-1993 0.0122 0.91221993-1994 0.0097 0.92161994-1995 0.066 0.79731995-1996 0.0033 0.95431996-1997 0.0079 0.92921997-1998 4.2161 0.04*1998-1999 4.2577 0.0391*1999-2000 1.1021 0.29382000-2001 7.6639 0.0056*2001-2002 0.099 0.7532002-2003 0.0292 0.86432003-2004 5.0383 0.0248*2004-2005 0.1959 0.65812005-2006 0.0789 0.77892006-2007 6.2268 0.0126*2007-2008 2.0437 0.15282008-2009 1.7761 0.18262009-2010 10.0466 0.0015*2010-2011 11.9106 0.0006*2011-2012 29.7893 <.0001*

DiscussionSpatial and Temporal Distributions

Management Implications

Spatial and Temporal Distributions

•Wintering manatee use continued to increase over time

• Peak winter use increased at a greater magnitude

• High winter manatee abundance attributed to increased manatee survival and population growth and increasingly severe winter events

•Summer manatee abundance also increased

• Peak usage typically attributed to late cold fronts

•Magnolia Springs, Banana Island, and Three Sisters Springs sanctuaries exhibited significantly higher manatee densities

• Suggests differences in quality among habitat type

•Manatee density was not equal among years, likely influenced by cold weather events

Management Implications

•Spatial distribution varies seasonally and is influenced by several natural and anthropogenic factors:

• Abundance and distribution in vegetation

• Condition and availability of warm-water refuges

• Altered distribution and behavior

• Increased mortality and injury

• Location of artesian springs

•With the onset of colder weather, manatees concentrate in the southern and eastern portions of the bay, aggregating near springs to keep warm

•This information has important management implications

King Spring•Located within Banana Island sanctuary boundaries

•Dense cluster surrounding the spring overflows outside of sanctuary boundaries, extending to Sunset Shores sanctuary

•Banana Island and Sunset Shoressanctuaries designated in 1980 afterreports of manatee harassment

Banana Island Sanctuary

• Since 1983, manatee density has increased by 0.003 manatees per sq. meter per year (p = 0.0087)

y = 0.0003x + 0.007R² = 0.2285

0

0.01

0.02

0.03

Man

atee

Den

sity

(p

er s

q. m

)

Year

Manatee densities within the Banana Island sanctuary from 1983 – 2012

Sunset Shores Sanctuary

• Positive increase in manatee density until 1994 when the Buzzard Island, Warden Key, and Tarpon Springs sanctuaries were designated• Suggests a shift in usage

• Heavier reliance on King Spring located within Banana Island sanctuary

• Overall negative trend in manatee density with -0.0002 manatees per sq. meter per year (p < 0.0001)

y = -0.0002x + 0.0058R² = 0.5578

0

0.005

0.01M

anat

ee D

ensi

ty (

per

sq

. m)

Year

Sunset Shores Linear (Sunset Shores)

Manatee densities within the Sunset Shores sanctuary from 1983 – 2012

Banana Island Addition

Magnolia Springs

•Human disturbance at King Spring may be the source of groups developing at Magnolia Springs

• Other springs within the bay are less favorable

• Smaller size, limited discharge, or narrow passes

Magnolia Springs Sanctuary

• A positive trend in manatee density has been observed with an increase of 0.0007 manatees per sq. meter per year (p = 0.0072)

y = 0.0007x + 0.0122R² = 0.2382

0

0.01

0.02

0.03

0.04

0.05

0.06

Man

atee

Den

sity

(p

er s

q. m

)

Year

Magnolia Springs Linear (Magnolia Springs)

Manatee densities within the Magnolia Springs sanctuary from 1983 – 2012

Magnolia Springs Addition

Three Sisters Springs

•Only confined-water body open to thepublic while wintering manatees are present

•Sanctuary made permanent in October 1998 due to reports of increasing harassment and observations of increasing numbers of manatees, recreational divers, and snorkelers

•Manatee densities have significantly increased over time

•As many as 100 swimmers and paddlers enter Three Sisters Springs per hour to see the manatees

•USFWS has proposed the creation of two expanded no-public entry areas

Three Sisters Springs Sanctuary

• Significant increases in manatee densities over time with 0.0432 manatees per sq. meter per year (p < 0.001)

Manatee densities within the Three Sisters Springs sanctuary from 1983 – 2012

y = 0.0432x - 0.2697R² = 0.6142

-0.5

0

0.5

1

1.5

2

2.5

Man

atee

Den

sity

(p

er s

q. m

)

Year

Three Sisters Springs Linear (Three Sisters Springs)

Three Sisters Springs Additions

Underutilized Sanctuaries•Warden Key, Tarpon Springs, and Buzzard Island sanctuaries subject to reevaluation

•Sanctuaries became established in 1994

• The number of manatees in the area had doubled since 1980

• Large increase in the number of visitors, the inability of the existing sanctuaries to provide sufficient shelter for manatees, and reports of increasing manatee harassment had been cited

•Exhibited significantly lower densities since they were created

• Do not contain any springs within their boundaries to provide thermal refuge

• Change in distribution of resources, encouraging manatees to relocate to newer sanctuaries

• Increase in salinity causing manatees to move to fresher waters or higher quality food sources

• Human disturbance

Warden Key Sanctuary• Negative trend in manatee

density with -0.00002 manatees per sq. meter per year

y = -2E-05x + 0.0007R² = 0.1346

0

0.0002

0.0004

0.0006

0.0008

0.001

0.0012

0.0014

0.0016

0.0018

Man

atee

Den

sity

(p

er s

q. m

)

Year

Warden Key Linear (Warden Key)

Manatee densities within the Warden Key sanctuary from 1983 – 2012

Tarpon Springs Sanctuary

• Negative trend in manatee density with -0.00008manatees per sq. meter per year

y = -8E-05x + 0.0021R² = 0.4667

-0.0005

0

0.0005

0.001

0.0015

0.002

0.0025

0.003

0.0035M

anat

ee D

ensi

ty (

per

sq

. m)

Year

North Banana Island Linear (North Banana Island)

Manatee densities within the Tarpon Springs sanctuary from 1983 – 2012

Buzzard Island Sanctuary

• Negative trend in manatee density with -0.00003manatees per sq. meter per year

y = -3E-05x + 0.001R² = 0.4477

0

0.0002

0.0004

0.0006

0.0008

0.001

0.0012

0.0014

0.0016

0.0018

0.002

Man

atee

Den

sity

(p

er s

q. m

)

Year

Buzzard Island Linear (Buzzard Island)

Manatee densities within the Buzzard Island sanctuary from 1983 – 2012

Summer Usage of Kings Bay•Manatees not limited by water temps and disperse in search of food and other resources

•King Spring supports a dense cluster of manatees, possibly due to freshwater flow

•Manatees have been using Kings Bay year-round in increasing numbers

• Warm-water refuge in winter may become a cold water source in summer

• Fresh water attracts the manatees

• Females with young calves do not migrate as readily

•Manatee distributions displayed an inconsistent pattern of seasonal use

•Heavier reliance must be placed on boat speed zone regulations

Conclusion and Recommendations

Conclusion•Significant change in manatee abundance among winter seasons

•Strong correlation between high manatee density and artesian springs

•Highest densities at three locations: King Spring, Three Sisters Springs, and Magnolia Springs

• Additional coverage needed to support the overflow of manatees outside of sanctuary boundaries

•Density patterns were not uniform across summer periods

•Differences in quality among sanctuaries

•Cold weather events had a significant effect on manatee density

•These findings are extremely valuable for implementing additional conservation measures in Kings Bay

Future Recommendations

•Additional shelter to protect high density populations

•Additional efforts in both enforcement of policies and public education

•Continuation of aerial manatee surveys

•Further analysis of the effects of human recreation and altered habitat

•Examine the effectiveness of boat speed zones during summer months

Acknowledgements

Dr. Richard Flamm Dr. Curtis Burney Joyce Kleen

Aarin-Conrad Allen Dr. Patrick Hardigan Amber BrelandDr. Edward Keith

Questions?