and 2900 ha were periodically flooded. In 
1992, a program of more active water 
management was begun. This included 
leaving impoundments open to lagoonal 
waters for longer periods of time and drawing 
down marsh water levels in the summer. 
Although meant as improvements to 
the area, the impounding of marsh areas had 
serious detrimental impacts on large groups of 
flora and fauna. Programs to restore critical 
marsh for the now extinct dusky seaside 
sparrow (Ammospiza maritima nigrescens) 
were unsuccessful. Vegetation was altered 
through the decrease in salinity of marsh areas 
and reduction of emergent vegetation. 
According to Montague ef al. (1984 and 
1985), salinity changes promoted the increase 
in freshwater vegetation such as cattail (Typha 
cf. domingensis) and shrubs such as Brazilian 
pepper (Shinus terebinthifolius), Carolina 
willow (Salix caroliniana), and wax myrtle 
(Myrica cerifera). 
Fire management, the focus of this 
study, has gone through substantive changes 
over the same period of time. Fire 
suppression was practiced from the time of 
acquisition (1962) until 1975. This resulted in 
the build-up of fire fuels and high risk for 
catastrophic fires. In 1975, a program of 
limited prescribed burning was initiated. 
Following severe fires in 1981, the area came 
under an intensified burning schedule, with 
generally a three-year fire cycle (Lee et al. 
1981, Adrian et al. 1983). 
2. METHODS 
To facilitate fire management, 
MINWR has been demarcated into fire units. 
Using 1979 1:12,000 color infrared 
photography (CIR), fire units that contained 
over 50% marsh composition were identified. 
Seven areas were then selected from these 
110 
high marsh areas with a variety of bum 
frequencies and timings (Table 1). 
Table 1. Years of prescribed burning on study 
fire units. 
  
  
Fire unit Years burned 
T-10-J 1986, 88, 89, 90 
T-10-K 1983, 85, 87, 89, 90 
T-10-L 1986, 88, 91, 92 
T-16 1982, 84, 88, 90 
T-17 1985, 86, 87, 91 
C-15-C 1989 
C-15-D 1989 
  
All of the selected fire units abut the eastern 
shore of the Indian River, and all except T-10- 
K are surrounded by dikes. T-10-K was one 
of the Dusky seaside sparrow recovery areas 
and had its dike system completely removed 
by 1979. 
Aerial photography (1:12,000, CIR) 
was acquired for 1982 (flown in October, 
1982) and 1992 (flown in April, 1992). The 
1992 photography was used to help train the 
vegetation interpreter in identification of major 
vegetation types. Each unit was then 
interpreted and digitized for the two years. 
A vegetation classification scheme that 
focused on shrub types was developed based 
on schemes use by Schmalzer and Hinkle 
(1985) and Smalzer (1995). These classes are 
defined in Table 2. Three of the vegetation 
classes were combined to form the overall 
shrub class. These were cabbage palm with 
shrub (CSH), willow (WIL), and shrub (SHR). 
Difficulties were noted in distinguishing 
between mangroves and shrub types. 
To test interpretation error, 91 
polygons meant to represent the varied 
vegetation types present were identified on the 
1992 aerial photography. These were 
  
  
Vegetati 
STG 
SPJ 
CAT 
CPS 
CPH 
CPS 
WIL 
SHR 
UPL 
WAT 
OTH 
selected 
identific: 
produce 
proporti 
was then 
field che 
E 
jointly 
divided i 
area rati 
polygons 
We bel 
would h: 
Polygon: 
quartiles 
five occ 
quartile