Late 20th century mangrove encroachment in the coastal Australian monsoon tropics parallels the regional increase in woody biomass
Post on 22-Apr-2023
0 Views
Preview:
Transcript
Late 20th century mangrove encroachment in the coastal Australian monsoon tropics parallels the regional increase in woody biomass
1 2 3
4
5 6 7 8 9
10
11 12 13 14 15
16 17 18 19 20
21
22
23
Grant J Williamson (Corresponding author) School of Plant Science, University of Tasmania, Private Bag 55, Hobart 7001, Australia. Email: grant.williamson@utas.edu.au Phone: +61 3 6226 1944 Fax: +61 3 6226 2698
Guy S Boggs Tropical Spatial Sciences Group, Charles Darwin University, Darwin NT 0909, Australia. Email: guy.boggs@cdu.edu.au
David MJS Bowman School of Plant Science, University of Tasmania, Private Bag 55, Hobart 7001, Australia. Email: david.bowman@utas.edu.au
Date of manuscript draft: 15 September 2009
1
ABSTRACT 24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
In Kakadu National Park, a World Heritage property in the Australian monsoon
tropics 250 km to the east of Darwin, a number of recent studies have shown that
woody encroachment (expansion of woody communities) and densification
(increased biomass in woody communities) has occurred in the last 40 years. The
cause of this increase in woody biomass is poorly understood, but possibly associated
with the control of invasive Asian water buffalo, trend to higher rainfall, and
increased frequency of fires. Mangroves provide an important context to understand
these landscape changes given that they are unaffected by fire or feral water buffalo.
We examine change in mangrove distribution in a series of coastal tropical swamps
fringing Darwin, Northern Territory, Australia over a 30-year period using a series of
7 aerial photographs spanning 23 years from 1974 and a 2004 high-resolution satellite
image. In late 1974 Darwin was impacted by an intense tropical cyclone. Vegetation
at 3,000 randomly placed points was manually classified, and a multinomial logistic
model was used to asses the impact of landscape position (coastal, intertidal and
upper-tidal) and swamp on mangrove change between 1974 and 2004. Over the study
period there was instability and slight mangrove loss at the coast, stability in the
intertidal zone and mangrove gain in the upper-tidal zone, with an overall increase in
mangrove presence of 16.2% above the pre-cyclone distribution. A swamp that was
impacted by drainage works for mosquito control and the construction of a sewage
treatment plant, showed a greater mangrove increase than the two unmodified
swamps. The mangrove expansion is consistent with woody encroachment observed
in nearby but ecologically distinct systems. Plausible causes for this change include
changed local hydrology, changes in sea level and elevated atmospheric CO2
concentrations.
KEY WORDS
Mangroves, aerial photography, climate change, coastal ecology, landscape change,
vegetation dynamics, woody vegetation
2
INTRODUCTION 53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
In many landscapes throughout the world there is a trend toward encroachment of
woody vegetation into adjoining treeless areas and increased cover of woody
vegetation in wooded communities, sometimes called densification. These processes
have been studied in great detail in Kakadu National Park, a World Heritage property
in the Australian monsoon tropics 250 km to the east of Darwin, using sequences of
orthorectified aerial photography. These studies have revealed that since the 1960s,
monsoon forest patches are expanding into tropical savanna (Banfai et al. 2007;
Banfai and Bowman 2007), tropical savannas are increasing their canopy cover
(Lehmann et al. 2008) and encroachment of fringing woody vegetation is occurring on
treeless freshwater floodplains (Bowman et al. 2008). There remains uncertainty and
debate about the cause of this expansion given that a number of changes have
occurred to potential drivers over the last 40 years. For example, feral Asian water
buffalo underwent a population irruption in the 1970s, which was not brought under
control until the 1990s (Petty et al. 2007). Fire regimes have changed from a
preponderance of fire in the late dry season in the late 1950s to sustained burning
throughout the late dry season (Bowman et al. 2007). Correlative studies have failed
to link the landscape scale woody expansion to either proxies of fire activity or
buffalo density, although these studies show that at the local scale these disturbances
can influence both encroachment and densification (Banfai and Bowman 2007;
Bowman et al. 2008; Lehmann et al. 2008). Collectively, researchers point to the
importance of the recent trend for increasing rainfall, or more controversially,
increased atmospheric CO2, in driving encroachment and densification in Kakadu
National Park.
3
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
In this context, studies of mangrove communities provide an important perspective on
these processes given these systems are unaffected by landscape fire or Asian water
buffalo impacts. Nonetheless, mangroves are highly dynamic systems that respond to
slight changes in sediment accumulation, salinity, water depth and recurrence of tidal
inundation (Bridgewater and Cresswell 1999). For example, mangroves bordering salt
flats can show localized dieback or expansion in response to slight changes in the
hydrology of the system (Duke et al. 1998).
Aerial photography has been shown to be useful for assessing vegetation change over
medium-term time scales (Fensham and Fairfax 2002) and has been used to study
mangrove change (Jones et al. 2004; McTainsh et al. 1986; Saintilan and Wilton
2001). For example, a recent study conducted in Florida with aerial photographs
spanning 60 years has found a dramatic shift in mangrove vegetation up to 3.3km
inland, that appears to be associated with a combination of sea level rise and local
hydrological effects (Ross et al. 2000).
Our aim is to identify changes in mangrove distribution and examine correlates of this
change in the period 1974 to 2004 in three contiguous coastal swamps in the Northern
Territory of Australia. All three swamps were severely impacted by Cyclone Tracy, a
category 4 tropical cyclone that occurred in December 1974 and destroyed the city of
Darwin and damaged the surrounding native vegetation (Stocker 1976). Leanyer
Swamp has been impacted by urbanisation in its catchment and the construction of
drains in the early 1980s in order to increase the drainage of stormwater and to
prevent to pooling of water from high tides, a preferred habitat of saltwater
mosquitoes such as Aedes vigilax (Medical Entomology Section 1983). The other
swamps have been minimally impacted by urbanization. There is some evidence that
4
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
nearby mangrove systems, for example, in the Mary River region (Mulrennan &
Woodroffe 1998) have responded to sea level change over the last 50 years.
Therefore, swamps provide an opportunity to chart mangrove recovery following
cyclone damage, assess the impacts of urbanization and contextualise background
changes associated potentially associated with global change including increased
rainfall, sea level rise and elevated CO2.
To examine changes in mangrove vegetation we use a sequence of high resolution
remotely sensed imagery (7 aerial photographs spanning 23 years that were
georeferenced to a QuickBird satellite image acquired in 2004).
Study Area
The study system comprises three contiguous swamps, Leanyer, Holmes Jungle and
Micket, developed around two tidal creeks, Buffalo Creek and Micket Creek, to the
east of Darwin in the Northern Territory of Australia (12°22´S, 130°55´E, Fig. 1).
This region experiences a monsoonal climate, with 90% of the average 1660 mm
annual rainfall occurring during the wet season from November to April, and a mean
daily maximum temperature of 32C that varies little throughout the year. The study
covers an area of approximately 30 km2 with a catchment area of 35 km2. The area
has a low elevation gradient and a maximum tidal range of close to 8 m, with the tide
reaching over 4 km from the coast along the tidal creeks at high tide. Vegetation
along each swamp is dominated by a gradient in salinity and water availability,
ranging from freshwater reed swamps, to brackish reeds and sedges, to mangrove
forest, with grassland and salt flats more distant from the tidal creeks. Comparable
5
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
vegetation patterns, in nearby Howard’s Swamp, has been described by Wilson and
Bowman (1987).
MATERIALS AND METHODS
Aerial photographs taken at low tide and during the austral winter dry season (May to
August), covering the study area for the years 1974, 1976, 1979, 1983, 1985, 1990,
and 1997, were obtained from the Northern Territory Department of Planning and
Infrastructure. A high-resolution Quickbird Satellite image, also captured at low tide,
was used for the 2004 coverage. The photographs obtained included colour, black
and white, and colour-infrared prints for various years. The photographs were
registered to the 2004 satellite image in ArcMap 9.1, using a 3rd-order polynomial
transform, with RMS error values of around five metres. The 2004 satellite image
was manually digitised and classified into polygons representing a set of eight
vegetation or landscape units at a scale of 1:3000 following ground reconnaissance.
The creek network was digitised using the 2004 satellite image. The aerial
photographs were all taken on the low tide cycle thus enabling manual mapping of the
coastal line as evidenced by boundary between sand and mud flats.
Elevation data in the form of a digital contour map was obtained from the Northern
Territory Government, but it was found that the 1 m intervals were too coarse to
adequately represent the slope of the salt flat and low-lying mangrove areas. A field
survey was conducted using a differential GPS system at 120 randomly placed points
in the low-lying salt pan and brackish swamp areas, to fill in gaps in the contour map
to a vertical resolution of under 10cm. A digital elevation model with a cell size of
6
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
100 m was produced from the combined data using ordinary kriging with a spherical
semi-variogram model using the Spatial Analysis extension in ArcGIS 9.1.
A total of 3,000 sample points were randomly placed across the vegetation map and a
number of habitat attributes were derived from the 2004 vegetation map and
preceding sequences of aerial photography. For each aerial photographic coverage the
points were manually classified into one of the 8 vegetation/landscape units by
examining the vegetation type inside circles of 10 m diameter at each point. For the
purposes of the analysis a generic mangrove formation was scored as present if > 50%
cover of the mangrove vegetation unit was observed at a point. Data for the elevation,
distance to creek, and distance to coast were spatially joined to the point samples.
Following field surveys on the ground and by helicopter during different times in the
tidal cycle, and following advice from the Medical Entomology Branch of the
Northern Territory Government who conduct frequent assessments of tidal extent for
the purposes of mosquito control, the swamps were a priori classified into three
landscape zones (coastal, intertidal, and upper tidal) based on distance from coast and
vegetation type and the sample points were assigned to these three landscape position
zones. While we recognize a continuous gradient in tidal and saline influence exists,
these three zones tend to represent distinct vegetation associations and tidal regimes.
The sample points were recoded as gain, loss or stable mangrove based on the
difference in mangrove presence at the point between 1974 and 2004, and a
multinomial logistic model, performed in R 2.8.1 (R Development Core Team 2008),
was used to determine the effect of swamp and landscape position on mangrove
change.
7
170
171
172
Darwin tidal gauge and rainfall data used in interpretation was obtained from
Australian National Tidal Centre and the Australian Bureau of Meteorology
respectively.
8
RESULTS 173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
A vegetation map and map of point classification into landscape position zones is
presented in figure 1. There was clear zonation in vegetation types from from the
coast inland. In the coastal zone the elevation was around 6 m Australian Height
Datum rising up to around 7 m in the upper tidal zone some 3 km inland (Table 1).
Distance from creek lines increases across this gradient from 100 to 300 m in the
coast and intertidal zones to almost 1 km in the upper-tidal zone (Table 1). The
coastal zone is dominated by mangrove vegetation, the intertidal zone by salt flats,
and the upper-tidal zone has the highest proportion of grass, sedge and reed vegetation
(Table 2). Vegetation patterns are related to salinity gradients and frequency and
depth of tidal inundation. For example, in the coastal zone mangroves (including
Bruguiera parviflora, Rhizophora stylosa and Avicennia marina) are dominant,
occupying areas subject to frequent tidal inundation. Mangroves also extend through
the intertidal zone along tidal creeks although much of this zone is comprised of
sparsely vegetated salt flats, that due to hypersaline conditions and extreme
evaporation rates only support mangrove vegetation close to frequently flushed creeks
The upper-tidal zone comprises a salinity gradient from low mangrove forests
dominated by Ceriops tagal, Lumnitzera racemosa, Avicennia marina and Sonneratia
lanceolata, through brackish Schoenoplectus littoralis reeds to Eleocharis dulcis in
areas subject primarily to freshwater inundation.
Across the 3,000 sample points, there was an overall increase in mangrove presence
of 16.2% between 1974 and 2004, amounting to an increase of approximately 123 ha
from the 1974 coverage of 761 ha. The upper-tidal zone showed a consistent increase
in mangrove cover over the 30 year study period, and little mangrove loss after the
9
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
1974 cyclone. In contrast the coastal zone showed some mangrove loss over the study
period, particularly following the cyclone (Figure 2). The intertidal zone in Micket
and Holmes Jungle swamps showed little change in total mangrove area at the end of
the study period, though an increase in this zone was seen in Leanyer swamp that was
modified by drainage works (Figure 2).
From the beginning to end of the study period, the negligible transitions away from
mangrove vegetation were mostly to open water or salt flat (Table 3). The transition
matrix shows a large proportion of the points transitioning to mangrove vegetation
were originally brackish Schoenoplectus sedgelands characteristic of the upper tidal
zone, and fewer mangrove transitions were from salt flats or open water. In the upper-
tidal zone of Leanyer swamp, in particular, Eleocharis and Schoenoplectus
sedgelands, sample points showed a significant conversion to mangroves as a result of
drain construction for mosquito control (Table 2).
The swamp x landscape position interaction models of mangrove increase and
decrease between the start and end of the study period showed a trend towards greater
conversion to mangrove in the upper tidal zone (Figure 3a) and greater loss of
mangroves at the coastal zone (Figure 3b). There was a statistically significant
difference between swamps (p < 0.05), with Leanyer swamp having particularly high
rates of conversion to mangrove, and low rates of conversion away from mangrove.
Mean sea level and rainfall (Figure 4) in this region have fluctuated during the study
period, and although a number of years stand out as exceptional, no trend can be
discriminated at the available temporal resolution.
10
DISCUSSION 219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
Despite a catastrophic cyclone in 1974 there has been a significant increase in the
coverage in mangrove distribution in the swamps adjacent to Darwin over the 30-year
study period. Immediately following the tropical cyclone, mangrove cover declined,
although there was recovery within 20 years back to the original distributional area of
the mangroves. There was both loss and gain of mangrove cover in the coastal zone
over the study period, consistent with the continuously shifting substrates found
previously in this system (Woodroffe and Grime, 1999). In contrast the upper-tidal
zone mangroves underwent expansion and densification, replacing treeless brackish
vegetation and, to a lesser extent, the hypersaline flats. The contrasting response of
mangroves with different positions along the elevation gradient from the coast to the
inland reflects the geomorphologies of the landscape settings: the coastal zone is
highly dynamic while in the intertidal zone favours more stable vegetation patterns
given the relatively fixed salinity gradients between creeklines and treeless saltflats
and floodplains (Hollins and Ridd 1997), unless impacted by drainage works.
Increased sedimentation as the result of urbanization and changing precipitation
regimes are often cited as drivers of mangrove expansion globally (Schwarz 2003,
Alongi 2005, Jupiter et al. 2007). The significantly different patterns of mangrove
increase in Leanyer swamp appear to be related to hydrological changes caused by
drainage works. This swamp was subject to the construction of a sewage treatment
plant, disrupting the flow of fresh and tidal water, and the construction of a system of
drains for mosquito control (Department of Construction and A.A.Heath 1978),
which allowed more rapid drainage of the swamp after heavy rainfall or high tides
(Medical Entomology Branch 1982). The pattern of mangrove expansion in Leanyer
11
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
around the drains is visible in the satellite imagery (Figure 5), a change that has also
been observed in mosquito control channels in eastern Australia (Breitfuss et al. 2003;
Jones et al. 2004). Harty (2004) notes that urbanization, with the associated increased
nutrient and sediment loads and changes to hydrology through engineering works can
also promote mangrove establishment consistent with our finding for Leanyer swamp.
However, the fact that the two swamps not subject to human influence also showed
strong trends in mangrove in the upper-tidal zone signals some profound changes to
the ecology of these swamps.
At the boundary between temperate and subtropical climates, changing temperatures
and reduced frequency of frosts have been proposed as a mechanism for mangrove
expansion into treeless saltmarsh communities in the United States (Stevens et al.
2006). But changes in temperature are unlikely to impact significantly on mangrove
communities in northern Australia, which experiences a tropical climate with no
frosts. Herbert (2007) suggests that mangrove expansion into saltmarsh habitat at the
landward fringe in the Hunter estuary in New South Wales is due to increased tidal
range resulting from the construction of tidal barriers and channel dredging.
However, such engineering works have not taken place in the region of the Darwin
study, and any alteration in mean sea level or tidal amplitude is presumed to be the
result of global change rather than local anthropogenic impacts.
Elevated atmospheric CO2 concentration has frequently been proposed as a driver of
the expansion of woody species at the expense of non-woody vegetation (Bazazz
1990; Archer et al. 1995; Bond and Midgley 2000; Eamus and Palmer 2007), as
higher CO2 concentrations are expected to increase photosynthetic rate and water use
efficiency more in woody plants than non-woody plants. Mangrove species have
12
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
shown responses in growth to elevated CO2 (Farnsworth et al. 1996) although this is
mediated by salinity. Controlled mangrove growth experiments under elevated CO2
have found little growth enhancement in high-salinity conditions but more growth
enhancement in low-salinity conditions in a less salt-tolerant species (Ball et al.
1997). This effect could lead to expansion into brackish or freshwater areas, as we
have observed in this system. However, controlled environment experiments growing
mangroves and saltmarsh species alone and in competition under elevated CO2 found
that competition reduced the impact of elevated CO2 on mangrove growth suggesting
other factors may be required to explain mangrove expansion (McKee and Rooth,
2008). Without further experimental data we can’t reject the CO2 fertilizer effects as a
plausible contributor to of the mangrove expansion. However, experiments
comparing the growth of the two invading mangrove species at the study site
(Avicennia marina and Sonneratia lanceolata) with and without the presence of
dominant sedge (Schoenoplectus littoralis) under elevated CO2 are required to more
fully evaluate the importance of the putative fertilizer effect in this system. A further
question arises as to the time scale over which any CO2 fertilization effect will be
apparent. Simulations of forest dynamics in response to elevated CO2 in temperate
forests have found significant increases in basal area over a time span of 50-150 years
(Bolker et al. 1995). Free-air carbon enrichment (FACE) experiments have
demonstrated increased carbon flux and photosynthesis in forests in response to CO2
enrichment, an effect that appears generally stronger in trees and C3 species, but
structural changes in woody ecosystems have generally not been observed over the
period of time these studies have been running (Ainsworth and Long 2005). It is
possible rapid-growing, disturbance tolerant mangrove communities may be capable
13
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
of showing a rapid response to CO2 enrichment, and the application of FACE
experiments to these systems would be informative.
The discontinuous landward expansion and seaward contraction is broadly correlated
with corresponding fluctuations of sea level over the study period. Therefore this
study shows general agreement with previous studies qualitatively linking mangrove
distribution changes with trends in sea-level rise (McTainsh et al. 1986; Ross et al.
2000) or that have been based on analysis of aerial photographs (Alongi 2008;
Dowling 1978; Gilman et al. 2007). Like our study, these latter authors found
mangrove invasion of brackish swamps at the limit of the inland tidal extent, and a
coincident loss of seaward mangroves. An obvious explanation for the landward
expansion of mangroves is the increased landward penetration of seawater (Lacerda et
al. 2007). Simulation modelling of mangrove habitat based on projected sea-level in
Florida over the next 100 years showed significant replacement of freshwater marsh
and swamp environments by mangroves (Doyle et al. 2003), and a similar outcome
was predicted for Homebush Bay, New South Wales based on fuzzy set modelling
(Zeng et al. 2007). In our study, direct time-series modelling of mangrove
distribution against sea level or rainfall trends was not possible due to the low number
of available repeat images and fluctuating tidal gauge data over the last decades
(Church et al. 2006). Such variability combined with the sporadic sequence of
imagery makes it difficult to directly link mangrove dynamics to sea level rise,
although saline intrusion, possibly as the result of sea-level rise, is a clear driver of
inland mangrove expansion in nearby areas (Woodroffe and Mulrennan 1993; Bell et
al. 2001; Winn et al. 2006).
14
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
Northern Australia is showing a trend in increasing rainfall, particularly during the
wet season (Smith 2004, Taschetto and England 2009), although this trend is not
evident in the mean annual rainfall data obtained for the years of this study (Figure 4).
Increased rainfall is expected to lead to increased growth rates in mangroves (Ellison
2000). Periods of high rainfall are also expected to contribute to mangrove
establishment in the hypersaline flats due to dilution of the salts in the soil (Field
1995; Duke 1997; Saintilan and Williams 1999; Harty 2004). However, expansion
into hypersaline salt flats is limited at our study site (Table 3). Indeed the major area
of mangrove expansion was into brackish swamps that are inundated by the highest
spring tides in October (at the end of the dry season) before being filled by freshwater
in the wet season. This suggests mangrove expansion into these brackish areas is
driven by increasing rather than decreasing salinity.
The expansion of mangroves observed in these coastal swamps is temporally and
spatially consistent with the woody densification and expansion observed in nearby
but ecologically distinct savannah and rainforest ecosystems. Densification in these
systems has often been attributed to changes in fire regime or feral animal populations
(Banfai et al. 2007, Bowman et al. 2008), factors that should not affect the mangroves.
Conversely, mangrove expansion is often attributed to sea-level rise, a factor unable
to influence inland forests. Little is known about the two potential effects of those
environmental drivers shared by both coastal mangrove forests and inland forests:
atmospheric CO2 concentration and changing rainfall patterns. Studies examining
mangrove tree growth across a wider area of northern Australia, so local differences
in substrate types, disturbance history and rainfall can be statistically evaluated would
be a useful future direction for research, as would paired comparison of mangrove and
rainforest tree growth within local climatic zones.
15
Conclusion 339
340
341
342
343
344
345
346
347
348
349
350
351
352
Despite severe damage from a tropical cyclone the coastal swamps examined in this
study all show landward mangrove expansion over a 30-year period, primarily
replacing brackish reed swamps in the upper tidal zone. Expansion rate is particularly
high in the swamp that has undergone hydrological changes as the result of human
engineering works, but given the expansion is also occurring in the other swamps,
direct human interference alone cannot be established as the cause. While the
observed changes are similar to those expected to be seen with sea-level rise, this
cannot be confirmed as the primary driver of change given the fragmentary aerial
photographic record. The mangrove expansion is consistent with densification trends
in other ecologically distinct ecosystems in northern Australia, including savannas and
rainforests, suggesting regional-scale factors are driving woody expansion. Plausible
candidates for this change include changed local hydrology, changes in sea level and
elevated atmospheric CO2 concentrations.
16
17
353
354
355
356
357
358
359
360
361
362
ACKNOWLEDGEMENTS
Funding for this project was provided by the Australian Research Council (ARC)
Linkage Grant (No. LP0667619), Northern Territory Department of Health and
Community Services, Australian Bureau of Meteorology, Northern Territory
Research and Innovation Fund, Australian Department of Defence and Charles
Darwin University. The authors would like to thank the Medical Entomology Branch
of the Northern Territory Health Department, especially Peter Whelan for expertise
and assistance, the Northern Territory Department of Planning and Infrastructure for
assistance with spatial data and equipment, and Lubomir Bisevac and Dimity Boggs
for assistance in the field.
REFERENCES
Ainsworth EA, Long SP (2005) What Have We Learned from 15 Years of Free-Air CO2 Enrichment (FACE)? A Meta-Analytic Review of the Responses of Photosynthesis, Canopy Properties and Plant Production to Rising CO2. New Phyt 165, 351-371. Alongi DM, Pfitzner J, Trott LA, Tirendi F, Dixon P, Klumpp, DW (2005) Rapid sediment accumulation and microbial mineralization in forests of the mangrove Kandelia candel in the Jiulongjiang Estuary, China. Estuarine, Coast Shelf S 63, 605-618. Alongi DM (2008) Mangrove forests: Resilience, protection from tsunamis, and responses to global climate change. Estuar Coast Shelf S 76, 1-13. Archer S, Schimel DS, Holland EA (1995) Mechanisms of Shrubland Expansion: Land Use, Climate or CO2? Climatic Change 29, 91-99. Ball MC, Cochrane MJ, Rawson HM (1997) Growth and water use of the mangroves Rhizophora apiculata and R. stylosa in response to salinity and humidity under ambient and elevated concentrations of atmospheric CO2. Plant Cell Environ 20, 1158-1166. Banfai D, Brook B, Bowman DMJS (2007) Multiscale modelling of the drivers of rainforest boundary dynamics in Kakadu National Park, northern Australia. Divers Distrib 13, 680-691. Banfai D, Bowman, DMJS (2007) Drivers of rain-forest boundary dynamics in Kakadu National Park, northern Australia: a field assessment. J Trop Ecol 23, 73-86. Bazzaz FA (1990) The Response of Natural Ecosystems to the Rising Global CO2 Levels. Annu Rev Ecol Systemat. 21, 167-196. Bell F, Menges CH, Bartolo RE (2001) Assessing the Extent of Saltwater Intrusion in a Tropical Coastal Environment Using Radar and Optical Remote Sensing. Geocarto International 16, 45-52. Bolker BM, Pacala SW, Bazzaz FA, Canham CD, Levin SA (1995) Species diversity and ecosystem response to carbon dioxide fertilization: conclusions from a temperate forest. Global Change Biol 1, 373-381. Bond WJ, Midgley GF (2000) A proposed CO2-controlled mechanism of woody plant invasion in grasslands and savannas. Global Change Biol 6, 865-869 Bowman DMJS, Dingle JK, Johnston FH, Parry D, Foley M (2007) Seasonal patterns in biomass smoke pollution and the mid 20th-century transition from Aboriginal to European fire management in northern Australia. Global Ecol Biogeogr 16, 246–256
18
Bowman DMJS, Riley JE, Boggs GS, Lehmann CER, Prior LD (2008) Do feral buffalo (Bubalus bubalis) explain the increase of woody cover in savannas of Kakadu National Park, Australia? J Biogeogr 35, 1976-1988. Breitfuss MJ, Connolly RM & Dale PER (2003) Mangrove distribution and mosquito control: transport of Avicennia marina propagules by mosquito-control runnels in southeast Queensland saltmarshes. Estuar Coast Shelf S 56, 573-579. Bridgewater PB & Cresswell ID (1999) Biogeography of mangrove and saltmarsh vegetation: implications for conservation and management in Australia. Mangroves and Salt Marshes 3, 117-125. Church J, Hunter J, McInnes K & White N (2006) Sea-level rise around the Australian coastline and the changing frequency of extreme sea-level events. Aust Meteorol Mag 55. Department of Construction, Department of the Northern Territory, AA Heath and Partners (1978) 'Leanyer and Marrara swamps, Darwin, environmental study: reduction of mosquito breeding, solid waste disposal, expansion of sewage treatment lagoons.' Darwin. Dowling R (1978) The Mangrove Communities of Moreton Bay. Northern Moreton Bay Symposium, The Royal Society of Queensland, St Lucia, Australia, 54-62. Doyle TW, Girod GF, Books MA (2003) Modelling Mangrove Forest Migration Along the Southwest Coast of Florida Under Climate Change. In Integrated Assessment of Climate Change Impacts on the Gulf Coast Region eds. ZH Ning, RETurner, TDoyle, KK Abdollahi. Gulf Coast Climate Change Assessment Council, Louisiana. Duke N (1997) Mangroves in the Great Barrier Reef World Heritage Area: current status, long-term trends, management implications and research. In State of the Great Barrier Reef World Heritage Area Workshop. Proceedings of a Technical Workshop held in Townsville, Queensland, Australia 27-29 November 1995 eds D Wachenfeld, J Oliver J, K David K. Great Barrier Reef Marine Park Authority. Duke NC, Ball MC, Ellison JC (1998) Factors Influencing Biodiversity and Distributional Gradients in Mangroves. Global Ecol Biogeogr Lett 7, 27-47. Eamus D, Palmer AR (2007) Is Climate Change a Possible Explanation for Woody Thickening in Arid and Semi-Arid Regions? Res Lett Ecol. Article ID 37364, 5 pp, doi:10.1155/2007/37364. Ellison JC (2000) How South Pacific mangroves may respond to predicted climate change and sealevel rise. In Climate Change in the South Pacific: Impacts and Responses in Australia, New Zealand, and Small Islands States. Eds. A Gillespie, W Burns, Kluwer Academic Publishers, Dordrecht, Netherlands.
19
Farnsworth EJ, Ellison AM, Gong WK (1996) Elevated CO2 alters anatomy, physiology, growth, and reproduction of red mangroves (Rhizophora mangle L.) Oecologia 108, 599-609. Fensham RJ & Fairfax RJ (2002) Aerial photography for assessing vegetation change: a review of applications and the relevance of findings for Australian vegetation history. Aust J Bot 50, 415-429. Field CD (1995) Impact of expected climate change on mangroves. Hydrobiologia 295, 75-81. Gilman,E. Ellison JC, Coleman R (2007) Assessment of Mangrove Response to Projected Relative Sea-Level Rise And Recent Historical Reconstruction of Shoreline Position. Environ Monit Assess 124, 105-130. Harty C (2004) Planning Strategies for Mangrove and Saltmarsh Changes in Southeast Australia. Coast Manag 32, 405-415. Herbert C (2007) Mangrove proliferation and saltmarsh loss in the Hunter Estuary. The Whistler 1, 38045. Hollins,S, Ridd PV (1997) Evaporation over a tropical tidal salt flat. Mangroves and Salt Marshes, 1, 95-102. Jones J, Dale PER, Chandica AL, Breitfuss MJ (2004) Changes in the distribution of the grey mangrove Avicennia marina (Forsk.) using large scale aerial color infrared photographs: are the changes related to habitat modification for mosquito control? Estuar Coast Shelf S 61, 45-54. Jupiter SD, Potts DC, Phinn SR, Duke, NC (2007) Natural and anthropogenic changes to mangrove distributions in the Pioneer River Estuary (QLD, Australia). Wetls Ecol Manag 15, 51-62 Lacerda LD, Menezes MOT, Molisani MM (2007) Changes in mangrove extension at the Pacoti River estuary, CE, NE Brazil due to regional environmental changes between 1958 and 2004. Biota Neotropica 7, 67-72. Lehmann CER, Prior LD, Williams RJ, Bowman DMJS (2008) Spatio-temporal trends in tree cover of a tropical mesic savanna are driven by landscape disturbance. J Appl Ecol. 45, 1304-1311. McKee KL, Rooth JE (2008) Where temperate meets tropical: multi-factorial effects of elevated CO2, nitrogen enrichment and competition on a mangrove-salt marsh community. Global Change Biol 14, 971-984. McTainsh G, Iles B, Saffigna P (1986) Spatial and temporal patterns of mangroves at Oyster Point Bay, south east Queensland 1944–1983. Proc Roy Soc Queensl 99, 83-91.
20
Medical Entomology Branch (1982) Ecology of the Major Mosquito Breeding Areas in Darwin. Casuarina, NT, Australia, NT Department of Health and Community Services. Medical Entomology Section (1983) 'A Guide to Mosquito Control in the Top End of the Northern Territory.' Northern Territory Department of Health, Casuarina, NT, Australia. Mulrennan ME, Woodroffe CD (1998) Saltwater intrusion into the coastal plains of the Lower Mary River, Northern Territory, Australia. J Environ Manag 54, 169-188. Petty AM, Werner PA, Lehman CER, Riley JE, Banfai DS, Elliott LP (2007) Savanna responses to feral buffalo in Kakadu National Park, Australia. Ecol Monogr 77 441-463. R Development Core Team (2008). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0, URL http://www.R-project.org. Ross M, Meeder J, Sah J, Ruiz P, Telesnicki G (2000) The Southeast Saline Everglades Revisited: 50 Years of Coastal Vegetation Change. J Veg Sci 11, 101-112. Saintilan N, Williams RJ (1999) Mangrove transgression into saltmarsh environments in south-east Australia. Global Ecol Biogeogr 8, 117-124. Saintilan N, Wilton K (2001) Changes in the distribution of mangroves and saltmarshes in Jervis Bay, Australia. Wetl Ecol Manag 9, 409-420. Schwarz A-M (2003) Spreading mangroves: a New Zealand phenomenon or a global trend? Water Atmos 11, 8-10. Smith I (2004) An assessment of recent trends in Australian rainfall. Aust Meteorol Mag 53, 163–173. Stevens PW, Fox SL, Montague CL (2006) The interplay between mangroves and saltmarshes at the transition between temperate and subtropical climate in Florida. Wet Ecol and Manag 14, 435-444. Stocker, GC (1976) Report on cyclone damage to natural vegetation in the Darwin area after cyclone Tracy, 25 December 1974. Forestry and Timber Bureau, Canberra. Taschetto AS, England MH (2009) An analysis of late twentieth century trends in Australian rainfall. Int J Climatol 29, 791-807. Wilson BA, Bowman DMJS (1987) Fire, storm, flood and drought: The vegetation ecology of Howards Peninsula, Northern Territory, Australia. Aust J Ecol 12, 165-174.
21
Winn K, Saynor M, Eliot M, Elio I (2006) Saltwater Intrusion and Morphological Change at the Mouth of the East Alligator River, Northern Territory. J Coast Res 22, 137-149. Woodroffe CD, Mulrennan ME (1993) Geomorphology of the Lower Mary River Plains, Northern Territory. Australian National University North Australia Research Unit and the Conservation Commission of the Northern Territory, Darwin. Woodroffe C, Grime D (1999) Storm impact and evolution of a mangrove-fringed chenier plain, Shoal Bay, Darwin, Australia. Mar Geol, 159, 303-321. Zeng TQ, Cowell PJ, Hickey D (2007) Predicting climate change impacts on saltmarsh and mangrove distribution: GIS fuzzy set methods. In GIS for the Coastal Zone: A Selection of Papers from CoastGIS 2006 Eds. CD Woodroffe , EM Bruce, M Puotinen & RA Furness, University of Wollongong.
22
Table 1. Mean and standard deviation of environmental variables for three landscape zones in the three swamps. Elevation is based upon Australian Height Datum, creek distance is the distance from nearest tidal creek, shore distance is the nearest distance to the low tide point.
Swamp Coastal Intertidal Upper-tidal Elevation Leanyer 6.95 ± 0.47 6.92 ± 0.48 7.46 ± 0.15(meters) Holmes Jungle 6.51 ± 0.31 6.92 ± 0.21 7.05 ± 0.29 Micket 6.62 ± 0.36 6.92 ± 0.70 7.40 ± 0.43Creek Distance Leanyer 0.32 ± 0.22 0.27 ± 0.25 0.91 ± 0.51(kilometers) Holmes Jungle 0.19 ± 0.14 0.43 ± 0.33 0.66 ± 0.49 Micket 0.18 ± 0.14 0.07 ± 0.06 0.53 ± 0.40Shore Distance Leanyer 0.65 ± 0.37 1.78 ± 0.39 3.58 ± 0.59(kilometers) Holmes Jungle 0.57 ± 0.34 2.04 ± 0.67 4.04 ± 0.74 Micket 0.62 ± 0.37 2.19 ± 0.50 3.48 ± 0.43
24
Table 2. Change in the percentage of all vegetation/landscape units for the three swamps broken down by coastal, intertidal and upper-tidal zones between 1974 and 2004. Salt flat includes all vegetation-free areas inundated by the highest tides. Typha, Eleocharis, and Schoenoplectus are graminoid dominated vegetation types that fall along a gradient of increasing salinity, while the grass/sedge category comprises the remaining treeless vegetation. Woodland comprises all non-mangrove woody vegetation, including savanna and coastal thicket.
Swamp Position Salt Flat % Water % Grass/Sedge % Eleocharis % Typha % Schoenoplectus % Mangrove % Woodland % Holmes Jungle Coast 20.8 → 28.3 0.5 → 0.5 3.7 → 2.7 0.0 → 0.0 0.0 → 0.0 0.0 → 0.0 72.2 → 64.2 2.7 → 4.3 1974 → 2004 Inter 58.4 → 71.2 1.5 → 0.3 14.8 → 6.2 0.0 → 0.0 0.0 → 0.0 0.0 → 0.0 24.9 → 21.4 0.3 → 0.9 Upper 21.2 → 18.0 0.0 → 0.0 41.7 → 39.0 9.9 → 8.7 0.9 → 1.7 5.2 → 2.6 20.3 → 28.2 0.9 → 1.7 Leanyer Coast 32.7 → 28.4 0.5 → 1.1 5.2 → 6.1 0.0 → 0.0 0.0 → 0.0 0.0 → 0.0 54.7 → 58.4 6.9 → 6.1 1974 → 2004 Inter 44.4 → 36.2 0.3 → 0.6 14.3 → 7.7 0.0 → 0.0 0.0 → 0.0 0.0 → 0.0 35.2 → 51.0 5.7 → 4.6 Upper 13.8 → 27.1 0.0 → 0.0 26.9 → 44.8 17.8 → 0.0 0.0 → 0.0 28.4 → 0.0 5.5 → 22.7 7.6 → 5.4 Micket Coast 8.2 → 10.6 8.8 → 15.3 0.0 → 0.0 0.0 → 0.0 0.0 → 0.0 0.0 → 0.0 82.9 → 74.1 0.0 → 0.0 1974 → 2004 Inter 37.5 → 32.9 3.9 → 5.4 3.9 → 4.9 0.0 → 0.0 0.0 → 0.0 0.0 → 0.0 51.9 → 54.2 2.8 → 2.6 Upper 19.8 → 16.7 0.0 → 0.0 30.6 → 30.6 14.4 → 14.9 0.0 → 0.0 14.0 → 6.8 11.7 → 24.8 9.5 → 6.3
25
Table 3. Vegetation transition matrix, showing proportion of points in each category converting to other categories between 1974 and 2004. Vegetation/landscape units are as in Table 2.
2004 Grass/Sedge Mangrove Salt Flat Schoenoplectus Woodland Eleocharis Water Urban Typha
Grass/Sedge 71.0% 4.2% 20.6% 0.0% 3.5% 0.5% 0.0% 0.2% 0.0%Mangrove 1.0% 87.7% 7.2% 0.2% 0.5% 0.0% 3.2% 0.1% 0.0%Salt Flat 5.0% 18.7% 75.4% 0.1% 0.5% 0.0% 0.1% 0.2% 0.0%Schoenoplectus 9.4% 44.1% 28.3% 12.6% 0.8% 4.7% 0.0% 0.0% 0.0%Woodland 21.5% 10.3% 4.7% 0.0% 57.0% 5.6% 0.0% 0.9% 0.0%Eleocharis 27.0% 7.0% 13.9% 4.3% 1.7% 42.6% 0.0% 0.9% 2.6%Water 0.0% 33.3% 10.3% 0.0% 0.0% 0.0% 56.4% 0.0% 0.0%Urban 0.0% 10.0% 0.0% 0.0% 30.0% 0.0% 0.0% 60.0% 0.0%
1974
Typha 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 100.0%
26
130°58'0"E130°57'30"E130°57'0"E130°56'30"E130°56'0"E130°55'30"E130°55'0"E130°54'30"E
12°20'30"S
12°21'0"S
12°21'30"S
12°22'0"S
12°22'30"S
12°23'0"S
12°23'30"S
12°24'0"S
12°24'30"S
VegetationSalt Pan / Bare
WaterWoodland
GrasslandMangrove
Brackish SedgeFreshwater Reed
Leanyer
HolmesJungle
Micket
0 1 20.5 km
LandscapePosition
Costal
Intertidal
Upper-tidal
Darwin
Figure 1. Map of 2004 vegetation communities in the study. The coastal, inter-tidal and upper-tidal zones used in the analysis are shown. The zones were defined in respect of tidal regimes and vegetation types (Table 2).
27
0.0 0.2 0.4 0.6 0.8
0.0 0.2 0.4 0.6 0.8
1975 1980 1985 1990 1995 2000 2005
0.0 0.2 0.4 0.6 0.8
Year
0.0 0.2 0.4 0.6 0.8
0.0 0.2 0.4 0.6 0.8
1975 1980 1985 1990 1995 2000 2005 0.0 0.2 0.4 0.6 0.8
Year
0.0
0.2
0.4
0.6
0.8
0.0
0.2
0.4
0.6
0.8
1975 1980 1985 1990 1995 2000 2005
0.0
0.2
0.4
0.6
0.8
Year
Leanyer Holmes Jungle Micket
Man
grov
e P
ropo
rtion
Coast
Inter
Upper
Figure 2. Proportion of sample points with mangroves over time for the coastal, intertidal and upper-tidal zones across the three swamps for all intervals between 1974 and 2004.
28
0
0.05
0.1
0.15
0.2
0.25
Coast Inter Upper
LeanyerHJMicketTotal
(a)
0
0.05
0.1
0.15
0.2
0.25
Coast Inter Upper
LeanyerHJMicketTotal
(b)
Figure 3. Mean proportion of points showing mangrove increase (a) and decrease (b) between 1974 and 2004. Error bars indicate standard deviation reported by multinomial logistic regression.
29
Figure 4. Preceding 5-year mean sea level and mean annual rainfall for study intervals in Darwin.
3900
3950
4000
4050
4100
1974 1976 1979 1983 1985 1990 1997 2004
Mea
n Se
a Le
vel (
mm
)
1500
1600
1700
1800
1900
2000
1974 1976 1979 1983 1985 1990 1997 2004
Year
Mea
n R
ainf
all (
mm
)
30
top related