735 
rejected at a 
measurements 
fects of the . 
sy were indeed 
t possible to 
terms of their 
to the distri- 
3 overall rel- 
;ribed by: 
:ale 
3pot heights 
led from the 
ssent at the 
id level of 
Level. The 
juare km. Not 
to the dense 
*ras extrapola 
se vegetation, 
rage flood 
L areas close 
ree metres, 
? area but the 
)t heights 
metres, or 
;h/volume est- 
i with that 
?hey are given 
íes 
(m 5 ) 
) 
I 
I 
) 
8 
8 
8 
8 
both sources 
asurement of 
is clear. It 
of the flood 
the Belize 
floodwater 
ne area of the 
reviding 85% 
irer is likely 
sn can be 
L' scene, by 
irce and dist- 
>f its influe- 
rer floodplain 
it information 
;s are not 
Lwaters and 
the most 
contributions 
>wer floodplain 
source of 
flooding as 
les. The posi- 
ibout + 150 
id slide dist- 
ortion constraints. The source of the floodwaters is 
not easily confirmed, since while that related to the 
Belize river is clearly identified, dense vegetation 
around the Sibun river and the dispersion of the local 
contribution prevent the tracing of all source loca 
tions. The area around the Straight and Almond Hill 
lagoons (approximately miles 9^ to I2-§ on the road) 
is clearly a focal point of inundatery waters that 
spread south from the Belize river. They appear to 
spread as a broad front across the floodplain but 
image processing techniques identified two main cha 
nnels, focal routes for floodwater passage. 
Band 7, undecimated images of the area were examined 
and manipulated by the process of piece-wise stretch 
ing. A histogram of pixel brightness values was obta 
ined and their distribution provided density slice 
boundaries. A series of stretches of increasing 
severity were applied to the extent of the erosion 
of lagoon margins. Two main channel features were seen 
within the broad front of the floodwaters, linear 
features about which the movement from the Belize 
river was concentrated. 
Figures 2 and 3 show the least and most extreme stre 
tches, respectively and may be compared to the distr 
ibutions of figure I and to the more detailed figures 
4, 5 and 6. Figure 3 indicates the probable routes 
taken by the floodwaterj though it must be stated that 
any contribution from the Sibun river is hidden by 
dense vegetation. 
The pattern of these floodwater channels can be seen 
to have a close correlation to the elevations shown 
by Metric Camera and ground survey information, most 
importantly in the area of the Western highway close 
the Straight and Almond Hill lagoons. The stretches 
also show that while the floodwaters are located 
around specific channels on the floodplain, they are 
derived from a broad margin of the Belize river and 
that relatively little flooding- occurs to the north of 
the river. Table 5 gives stretch boundaries. 
Table 5. Boundary brightness values for stretches 
Stretch no. 
Original b.v. 
Redesigned b.v. 
I 
0 
0 
20 
200 
21 
201 
55 
255 
2 
0 
0 
10 
I70 
II 
I7I 
55 
255