Accepted Manuscript Temporal changes in physical, chemical and biological sediment parameters in a tropical estuary after mangrove deforestation Marianne Ellegaard, Nguyen Ngoc Tuong Giang, Thorbjørn Joest Andersen, Anders Michelsen, Nguyen Ngoc Lam, Doan Nhu Hai, Erik Kristensen, Kaarina Weckström, Tong Phuoc Hoang Son, Lars Chresten Lund-Hansen PII: S0272-7714(14)00054-7 DOI: 10.1016/j.ecss.2014.03.007 Reference: YECSS 4391 To appear in: Estuarine, Coastal and Shelf Science Received Date: 11 May 2013 Revised Date: 13 January 2014 Accepted Date: 8 March 2014 Please cite this article as: Ellegaard, M., Tuong Giang, N.N., Andersen, T.J., Michelsen, A., Lam, N.N., Hai, D.N., Kristensen, E., Weckström, K., Hoang Son, T.P., Lund-Hansen, L.C., Temporal changes in physical, chemical and biological sediment parameters in a tropical estuary after mangrove deforestation, Estuarine, Coastal and Shelf Science (2014), doi: 10.1016/j.ecss.2014.03.007. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. ACCEPTED MANUSCRIPT 1 Temporal changes in physical, chemical and biological sediment 2 parameters in a tropical estuary after mangrove deforestation 3 Marianne Ellegaardad*, Nguyen Ngoc Tuong Giangbd, Thorbjørn Joest Andersenc, Anders 5 Michelsend, Nguyen Ngoc Lamb, Doan Nhu Haib, Erik Kristensene, Kaarina Weckströmf, Tong Phuoc 6 Hoang Sonb, Lars Chresten Lund-Hanseng RI PT 4 7 *Corresponding author; [email protected], Thorvaldsensvej 40, Department of Plant and 9 Environmental Sciences, University of Copenhagen, DK-1871 Frederiksberg C, Denmark, 10 M AN US C 8 telephone +4535320024 11 12 a Department 13 b Institute 14 c Department 15 Denmark 16 d Department of Biology , University of Copenhagen, Denmark 17 e Department of Biology, University of Southern Denmark, Denmark 18 f Department of Marine Geology and Glaciology, Geological Survey of Denmark and Greenland, 19 Denmark 20 g Aquatic of Plant and Environmental Sciences, University of Copenhagen Denmark of Oceanography, Vietnam Academy of Science and Technology, Nha Trang, Viet Nam TE D of Geosciences and Natural Resource Management, University of Copenhagen, 21 EP Biology, Department of Bioscience, Aarhus University, Denmark Email addresses: 23 Nguyen Ngoc Tuong Giang: [email protected]; Thorbjørn Joest Andersen: [email protected]; 24 Anders Michelsen: [email protected]; Nguyen Ngoc Lam: [email protected]; Doan Nhu Hai: 25 [email protected]; Erik Kristensen: [email protected]; Kaarina Weckström: [email protected]; 26 Tong Phuoc Hoang Son: [email protected], Lars Chresten Lund-Hansen: lund- 27 [email protected] AC C 22 28 29 Key words: Diatoms, palaeoecology, stable isotopes, grain size, sedimentation, Viet Nam 30 1 ACCEPTED MANUSCRIPT Abstract 32 Dated sediment cores taken near the head and mouth of a tropical estuary, Nha-Phu/Binh Cang, in 33 south central Viet Nam were analyzed for changes over time in physical, chemical and biological 34 proxies potentially influenced by removal of the mangrove forest lining the estuary. A time-series 35 of satellite images was obtained, which showed that the depletion of the mangrove forest at the 36 head of the estuary was relatively recent. Most of the area was converted into aquaculture ponds, 37 mainly in the late 1990´s. The sediment record showed a clear increase in sedimentation rate at 38 the head of the estuary at the time of mangrove deforestation and a change in diatom assemblages 39 in the core from the mouth of the estuary indicating an increase in the water column turbidity of 40 the entire estuary at the time of the mangrove deforestation. The proportion of fine-grained 41 sediment and the δ13C signal both increased with distance from the head of the estuary while the 42 carbon content decreased. The nitrogen content and the δ15N signal were more or less constant 43 throughout the estuary. The proportion of fine-grained material and the chemical proxies were 44 more or less stable over time in the core from the mouth while they varied synchronously over 45 time in the core from the head of the estuary. The sediment proxies combined show that 46 mangrove deforestation had large effects on the estuary with regard to both the physical and 47 chemical environment with implications for the biological functioning. 48 1. Introduction 49 The area occupied globally by mangrove forests has declined by more than 35%, and most of this 50 loss has occurred over the past 30 years, primarily due to human disturbance (Agardy et al., 2005; 51 Bouillon et al., 2008). Loss of mangrove forests is caused by a variety of human activities, 52 including clearing for lumber and firewood, development of rice paddies and establishment of 53 aquaculture ponds (Valiela et al., 2001; Wilkie and Fortuna, 2003). In many places, mangrove 54 removal occurs without regard for the role of mangrove forests in the functioning of adjacent 55 coastal ecosystems (e.g. Alongi, 2008). Besides the function as e.g. valuable nursery and feeding 56 grounds for a variety of invertebrate and fish species (Hong & San 1993), mangrove forests are 57 key sites for important biogeochemical transformations as well as exchange of organic and 58 inorganic carbon between land, atmosphere and ocean (Dittmar et al. 2006; Kristensen et al. 59 2008) . This is of particular concern as mangrove forests have the potential to mitigate the effects 60 of future climate changes (Alongi, 2008; Donato et al., 2011; Mcleod et al., 2011), both by 61 sequestering carbon, protecting neighboring estuaries from terrestrial runoff and sheltering AC C EP TE D M AN US C RI PT 31 2 ACCEPTED MANUSCRIPT coastlines from storm damage, (Dahdouh-Guebas et al., 2005; Granek and Ruttenberg, 2008). Viet 63 Nam is very dependent on the functioning of its coastal ecosystems because of its approx. 3260 km 64 long coastline with large areas of lowland including large delta regions in the south (Mekong 65 Delta) and north (Red River Delta). There are over 250 large and small estuaries along the 66 Vietnamese coast and most of these were previously lined with dense mangrove forests. However, 67 much of this mangrove area has now been logged (Hong & San 1993). 68 RI PT 62 The Nha Phu-Binh Cang estuary (NP-BC) is located in south-central Viet Nam. The head of this 70 estuary was formerly covered by dense forests with dominance of the mangrove genera 71 Rhizophora, Avicennia, and Sonneratia (Nguyen et al., 2006). However, this area and the mouths of 72 two rivers entering the estuary are now dominated by aquaculture ponds and the mangrove 73 forests have mostly disappeared. The exact timing of the large-scale reduction of the mangrove 74 forest was previously unknown and the associated effects were recognized only from anecdotal 75 evidence. M AN US C 69 76 Although mangrove forests have great influence on the estuaries they line, relatively little is 78 known about the consequences of mangrove deforestation for species distribution and ecosystem 79 functioning of the estuaries. This study aimed at describing changes over time in physical, 80 chemical and biological sediment parameters associated with mangrove deforestation in the NP- 81 BC estuary by exploring the sedimentary archive at stations located at the head and mouth of the 82 estuary. As mangrove forests often act as filters for e.g. nutrients and sediment from river run-off 83 (Prasad and Ramanathan, 2008; Twilley and Rivera-Monroy, 2009), we would expect to see 84 changes in e.g. terrestrial input, turbidity and nutrients. To determine such long-term 85 developments, the parameters explored included temporal changes in sedimentation rate, grain 86 size distribution, carbon and nitrogen content as well as stable isotope (δ13C ,δ15N) signatures and 87 diatom community composition. The age-depth profiles of these proxies were related to the timing 88 of the mangrove forest disappearance and other anthropogenic activities in the area, such as 89 constructions on the rivers entering the head of the estuary. AC C EP TE D 77 90 91 2. Materials and methods 92 2.1 Site description 3 ACCEPTED MANUSCRIPT The Nha Phu-Binh Cang (NP-BC) estuary is located in south-central Viet Nam, about 15 km north 94 of Nha Trang city (Fig. 1). It is about 20 km long and 5-6 km wide. The head of the estuary consists 95 of a shallow area with a maximum depth of 2 whereas the mouth of the estuary at station 3 is ca. 96 12 m deep. The depth increases to more than 30 m within 10 km of the head of the estuary. Two 97 larger rivers with a drainage area of about 1,200 km2 discharge into the estuary. A base-flow of 60 98 m3 s-1 was measured in one of these rivers, the Dinh River, during the rainy season (Lund-Hansen 99 et al., 2013). The residence time of water in the estuary is about 5-6 days (Lund-Hansen et al., 2010). 101 M AN US C 100 RI PT 93 2.2 Satellite images of mangrove distribution 103 Maps of the head of NP-BC were obtained and analyzed to determine the timing of mangrove 104 conversion into shrimp ponds. The maps were based on satellite imagery obtained from the 105 Landsat MSS (1973 and 1983), MOS-1 (1989, 1993, and 1996), and Landsat ETM (1999) with 106 spatial resolutions of 60, 50, and 30 m respectively. Four spectral bands were used from the 107 Landsat MSS, and MOS-1, and 6 bands from the Landsat ETM. The GIS software package ENVI 4.4 108 was used for data processing and generation of thematic maps. A time-series of selected years 109 between 1973 and 1999 covering the head of the estuary is used to demonstrate the changes. D 102 TE 110 2.3 Sampling 112 Three sampling stations, station 5 (12.4264 N 109.1826 E), station 77 (12.4117 N 109.2089 E) and 113 station 3 (12.3500 N 109.2500 E), were established at the head and the mouth of the estuary. They 114 are henceforth termed the head of the estuary (station 5 in the inner and 77 in the outer of the 115 shallow part of the estuary) and the mouth of the estuary (station 3). A Kayak corer with a 116 diameter of 83 mm was used to collect sediment cores in November 2009 (station 5 and 3) and 117 November 2010 (station 77). One core from each station was sliced at 1-cm intervals. The samples 118 were kept in the dark at 5⁰C until freeze-dried. Subsamples from all three cores were processed 119 for 210Pb dating, while depth profiles of grain size analysis, C and N elemental concentrations and 120 stable isotope signatures were only achieved for core 5 and 3. These latter parameters are only 121 included from the surface of core 77. Furthermore, subsamples from core 3 at the mouth were 122 processed for diatom analyses. AC C EP 111 123 4 ACCEPTED MANUSCRIPT 2.4 Sediment dating 125 The freeze-dried subsamples were analyzed for 210Pb, 137Cs and 226Ra by gamma-spectrometry at 126 the Gamma Dating Center at the Department of Geography and Geology, University of Copenhagen. 127 The measurements were carried out on a Canberra low-background Ge-detector. 210Pb was 128 measured from its gamma-peak at 46,5keV, 226Ra from the granddaughter 214Pb (peaks at 295 and 129 352keV) and 137Cs from its peak at 661 keV. RI PT 124 130 2.5 Grain size determination 132 Grain size analyses were conducted at the Centre for Geo-genetics of Copenhagen University. Wet 133 subsamples were suspended in about 10 ml of distilled water in plastic containers. These samples 134 were sonicated for about half an hour to break clusters and aggregates before analysis for grain 135 size using Malvern laser Master-sizer 2000, which measures material from 0.02 to 2000 μm. Each 136 sample was automatically measured 3 times to get a mean grain size distribution. Some samples 137 showed peaks at 500-1000 µm. These fractions were excluded from the analyses as they were 138 considered to be aggregates or fragments of organisms. M AN US C 131 139 2.6 Stable C and N isotopes 141 Shell fragments and living biomass were manually removed from wet sediment before treating 1 g 142 subsamples with 5 ml of 1M HCl to dissolve particulate inorganic carbon (PIC). After freeze-drying 143 the sediment was finely ground to a homogenous powder and subsequently dry weights of 25-30 144 mg were wrapped in 5x9 mm tin capsules before combustion and EA_IRMS (elemental analysis- 145 isotope ratios mass spectrometry) for determination of C and N elemental concentrations and 146 isotopic values. The results are expressed as δ values in per mil units (‰) following the equation: 147 δ = (Rsample/Rstandard – 1) × 1000 (Sulzman 2008) where R = 13C/12C or 15N/14N. Tests showed that 148 there was no substantial effect of acid treatment on N concentration or δ15N. Non-treated and 149 treated cores showed N concentrations of 0.18% and 0.16% and δ15N values of 5.01‰ and 150 5.47‰, respectively. AC C EP TE D 140 151 152 2.7 Diatom analysis 153 Subsamples from the core from the mouth of the estuary were prepared for diatom analysis as 154 described by Renberg (1990) and later modified by Battarbee et al. (2001). Sediment samples 5 ACCEPTED MANUSCRIPT were oxidized with 35 % H2O2 to remove organic matter. After oxidation, a few drops of 10% HCl 156 were added to remove the remaining H2O2 and any carbonates, and finally 10 % NH4 was used to 157 remove minerogenic matter before making permanent slides. A volume of 400 μl final sample 158 solution was carefully pipetted on a coverslip, left to evaporate and permanent slides were made 159 using Naphrax for fixation. Diatoms were identified using an Olympus BH2-Japan light microscope 160 with 60x and 100x phase contrast objectives. On average 300, and at least 100, diatom frustules 161 were counted per sample. Stratigraphic profiles of the diatom distribution, and of the other 162 proxies, were created using the program C2 (Juggins, 2007). M AN US C 163 RI PT 155 3. Results 165 3.1 Timing of mangrove disappearance 166 The time-series of satellite images shows a progressively diminishing area of mangrove forests at 167 the head of the estuary starting in 1973, progressing through the 1980s and accelerating in the 168 early 1990s (Fig. 2). There was a distinct initial decline in the area covered by mangrove forest 169 around 1983, while the major part of the mangrove forest disappeared during a period of 5 years 170 in the mid to late 1990s. At the same time 80% of the area presently covered by aquaculture 171 ponds appeared (Fig. 3). D 164 TE 172 3.2 Chronology in the cores 174 The sediment from the three examined stations have very low contents of 137Cs (below 6 Bq kg-1), 175 which renders this isotope not valuable for dating purposes in the present case. Similarly low 176 activities were found by Frigniani et al (2007) in Tam Giang – Cau Hai lagoon in central Vietnam. 177 They speculated that the low activity indicated a possible loss of 137Cs attached to fine particles 178 which bypass the estuary and are transported to the ocean. The activity of unsupported 210Pb in 179 cores 5 and 77 from the head of the estuary shows high and only slightly decreasing 210Pb content 180 in the upper 60 and 40 cm of the sediment, respectively. The activity of 210Pb decreased rapidly 181 below these depths with no measurable content of unsupported 210Pb below (Fig. 4).The 182 chronologies for the cores were calculated using a modified CRS-model (constant rate of supply) 183 where the inventory below 60 and 40 cm, respectively, was calculated from a regression of 184 unsupported 210Pb vs. accumulated mass depth. The cores indicate rapid sedimentation since 185 1984 (core 5) and 2000 (core 77), on top of sediment which is at least 100 – 120 years old as AC C EP 173 6 ACCEPTED MANUSCRIPT deduced from the absence of unsupported 210Pb. Calculations of chronologies based on the CIC- 187 method (Constant Initial Concentration, Appleby, 2001) give essentially similar results. It is of 188 note that any sediment mixing, e.g. induced by bioturbation, will tend to increase the apparent 189 accumulation rate in the top of the cores and the calculated sedimentation rates are therefore 190 maximum values. Core 3 from the mouth of the estuary showed a more gradual decline in 191 unsupported 210Pb in the upper 60 cm, but again there is an indication of changing sedimentation 192 at depth, although with considerable variability below 40 cm depth (Fig. 4). The chronology in this 193 core was therefore calculated as outlined above with a regression in only the upper 40 cm. Based 194 on this chronology the upper 40 cm at the mouth has been deposited since 1938 (Fig. 5). Only the 195 data for the dateable parts of each core have been included in all subsequent evaluations. M AN US C RI PT 186 196 3.3 Silt and clay contents 198 There are high levels of fine-grained material at all locations in the estuary, with highest content of 199 silt and clay (<63 µm) at the mouth of the estuary (Fig. 6) and with sand contents generally below 200 10 %. The percentage of fine-grained sediment in the surface increased from around 90% at the 201 head to around 98% at mouth of the estuary. The proportion of fine-grained material at the head 202 of the estuary fluctuated with depth/age between 85% and 91-94%, with lowest levels around 203 1978 and 1998. The proportion of fine-grained material at the mouth is extremely stable at 98% 204 from the surface of the sediment and down to the depth dated around 1973 at station 3 (Fig. 6). 205 Below 1970 (including the un-dateable part of the core), the proportion of fine grained material 206 fluctuated between 92 and 97%. TE EP 207 D 197 3.4 Profiles of carbon (C) and nitrogen (N) concentrations, and stable isotope signatures 209 The carbon content is higher at the head (1.9-2.2%) than at the mouth of the estuary (1.2-1.6%, 210 Fig. 6). The carbon content fluctuated with the lowest level around 1998 at the head, while there 211 was a gradual decrease with depth at the mouth. The nitrogen content is relatively low and 212 similar at both stations (0.09-0.16%) and decreased gradually with depth even below the dateable 213 levels (Fig. 6). Variations in both carbon and nitrogen content coincide with the variations in grain 214 size at the head of the estuary (Fig. 6). The C/N ratio at the head reached values as high as 17 (Fig. 215 6), and generally increased with depth in the sediment. Considerably lower C/N ratios were AC C 208 7 ACCEPTED MANUSCRIPT 216 observed at the mouth of the estuary, where it was stable at around 10 from ca. 1950 to the 217 present, but with a slightly higher level (up to 12) deepest in the sediment. 218 The δ13C signal is lower at the head than at the mouth of the estuary. δ13C at the head varied from - 220 24 to -27‰ with depth in the sediment (Fig. 6). The signal at the mouth remained stable at -22‰ 221 throughout the core, even below the dateable part. The overall levels of δ15N are similar at station 222 5 and 3, but considerably less stable with depth at the head (from 4 to 7‰) than at the mouth 223 (from 5 to 6‰). The fluctuations in δ15N at the head coincided with the fluctuations in 224 concentration of silt and clay (Fig.6). The values for the surface sample from core 77 in the 225 outermost part of the head lagoon were generally between the values for the inner head and the 226 mouth (Fig. 6). M AN US C RI PT 219 227 228 The correlations between the percentage mud/silt and the four chemical parameters are all 229 significant (p ≤ 0.05; Pearsons product moment correlation), confirming the synchronicity of the 230 fluctuations seen in figure 6. 231 3.5 Diatom composition 233 A total of 89 diatom taxa were identified in the core from the mouth of the estuary, with 69 of 234 them identifiable to species level and the rest to genus level (raw data in supplementary material). 235 The proportion of planktonic diatoms fluctuated in the bottom part of the core, stabilized at 10-20 236 % in the middle part of the core (ca. 1960-1985) and increased to levels around 50-80 % in the 237 top of the core (ca. 1988-2007) (Fig. 7). The dominant planktonic species in the upper part 238 included Cyclotella striata with a maximum of 83 %, Paralia sulcata at 69 % and Thalassionema 239 nitzschioides at 48 %. The dominant benthic species in the intermediate layer was Nitzschia 240 valdestriata with a greatest relative abundance of 89%. TE EP AC C 241 D 232 242 4. Discussion 243 4.1 Mangrove deforestation 244 The NP-BC estuary has for decades been influenced by intense anthropogenic activity in the form 245 of intensive fishing (Strehlow, 2006), mangrove deforestation and widespread aquaculture (Table 246 1). The rivers running into the estuary are also influenced by human activities, notably the river 8 ACCEPTED MANUSCRIPT Dinh on which a dam (1983) and two weirs (1968 and 1993) have been constructed (pers. comm. 248 Doan Nhu Hai, Institute of Oceanography). According to Mazda et al (2002), deforestation of 249 mangrove areas in Viet Nam has occurred since the late 19th century and was caused first by 250 development of rice paddies and then in the 20th century by herbicide action during the Viet Nam 251 war. However, deforestation of the mangrove areas at the head of the NP-BC estuary has been 252 quite recent, beginning ca. 1973 and intensifying in the 1990´s, with the major part of the forests 253 disappearing within a period of only 5 years from ca. 1993-1998 (Figures 2 & 3). RI PT 247 254 4.2 Sediment dynamics 256 The amount of particulate matter input to tropical estuaries depends on various factors. Among 257 these are land use (Carmichael and Valiela, 2005), river discharge (Umezawa et al., 2009) and not 258 least the extent of mangrove vegetation (Furukawa et al., 1997). Much of the particulate matter 259 entering estuaries lined with well-developed mangrove forests are efficiently trapped by the 260 dense root system (Brinkman et al., 2005). If this particle trapping root-system disappears by 261 deforestation, the estuary will receive a large input of suspended particles. Accordingly, the 210Pb 262 profiles in the cores near the head of the NP-BC estuary showed recent rapid sedimentation 263 coinciding approximately with the time of the greatest mangrove deforestation. The approx. 60 cm 264 of sediment above the ~1984 dating horizon in core 5 is estimated to have been deposited within 265 25 years with an average rate of approx. 2.4 cm y-1, whereas the sediment below 60 cm was 266 deposited at least 100 years ago. In core 77, sediment accretion has been 3.6 cm y-1 since 2000, 267 again with underlying sediment deposited at least 100 years ago. Such a rapid change in 268 sedimentation was not evident in core 3 from the mouth of the estuary that showed a much lower 269 and relatively uniform rate of approx. 0.55 cm y-1 since 1938 (Figures 4 & 5). This indicates that 270 the effect of increased sedimentation was diluted or dispersed by tides and ocean currents at the 271 mouth of the estuary. The higher content of fine-grained material near the mouth of the estuary 272 supports the analyses by Bui (1997) indicating that coarser material is deposited rapidly near the 273 head (Fig. 6). AC C EP TE D M AN US C 255 274 275 The sedimentation rates given above are maximum values assuming limited biogenic particle 276 mixing of the sediments. This seems to be a valid assumption towards the mouth where only 277 sparse populations of small macrofaunal species are present (Nguyen et al. 2012; pers. comm. 9 ACCEPTED MANUSCRIPT Kurt Thomas Jensen, Aarhus University). However, numerous holes with a diameter of 2 to 5 cm 279 were observed visually in the muddy sediment bed at the head of the estuary. These may be 280 burrow openings formed by mantis shrimps, which are known to inhabit such sediments, forming 281 burrows to at least 30 cm depth (Atkinson et al., 1997). The presence of such a vigorous 282 bioturbator may cause an overestimation of the sedimentation rate at the head of the estuary, but 283 it is not possible to quantify this effect because no data on particle mixing rates by mantis shrimps 284 are available. Importantly, the indications of recent rapid sedimentation at the head of the estuary 285 are corroborated by questionnaires to local residents in the villages around the estuary, who note 286 a reduction in water depth at the head of the estuary (Strehlow, 2006). In this report it is stated: 287 “…..several water areas around Tam Ich that have been deeper in the past, i.e. 2.5 m, are only 1.5 m 288 deep today…... In front of Tan Dao this phenomenon is even more distinct. The water that used to 289 be more than 2 m deep now only measures 70 cm…...”. Tam Ich is a village close to the river mouth 290 at the southwest part of the head of NP-BC and Tan Dao is a village slightly further along the south 291 shore in the inner part of the estuary (see Fig 1). M AN US C RI PT 278 292 Only few studies have reported sedimentation rates in tropical estuaries, but it appears that the 294 recent rapid sediment accretion at the head of the estuary is considerably larger than typically 295 observed. For example, the sedimentation rates found at the head of the estuary are approx. 5 to 296 10 times higher than the 0.3 to 0.6 cm y-1 found by Frignani et al. (2007) in Tam Giang – Cau Hai 297 lagoon in central Vietnam and 8 to 10 times higher than the rate found in the Las Matas lagoon in 298 the Mexican gulf (Ruiz-Fernández et al., 2012). We infer that deforestation of the mangrove 299 vegetation is the main cause for the rapid sedimentation at the head of estuary, although our rates 300 may be overestimated due to bioturbation. 301 The concurrent temporal changes in the silt/clay content at the head of the estuary with 302 changes in C and N concentrations and δ15N signatures and the negative correlation with 303 δ13C signatures (see Figure 6) indicates that these parameters are all responding to the 304 same forcing factor(s). A similar pattern was found in the Ba Lat Estuary, also in Viet Nam, 305 where TOC and δ13C showed opposite patterns (Tue et al 2011b). In Ba Lat, however, C/N 306 and δ13C covaried, while their relationship is more unclear at the head of the NP-BC 307 estuary. This pattern in Ba Lat was interpreted as being governed by diagenetic processes 308 and a major shift in the source of organic material. Mangrove-derived debris is commonly AC C EP TE D 293 10 ACCEPTED MANUSCRIPT transported into the estuaries lined with mangrove forests. This particulate organic matter 310 feeds the estuarine food webs and will ultimately be deposited in the sediments 311 (Jennerjahn and Ittekkot, 2002; Wolff, 2006) at an extent that varies with the distance from 312 the mangrove vegetation and the amplitude of tidal cycles (Rezende et al., 1990). 313 Mangrove deforestation will therefore not only impact the transport of particulate matter 314 from land to the estuarine ecosystem, but also its composition, leading to higher deposition 315 of silt and lower deposition of mangrove derived organic matter (Alfaro 2010). 316 Accordingly, the observed shifts in elemental composition of NP-BC sediments correlate 317 with the two major periods of mangrove forest disappearance. The first small deforestation 318 occurred around 1983-85 followed by a second and much more extensive deforestation 319 around 1993-95. The timing of these periods also corresponds with the changes in the 320 diatom records after 1975, with an initial increase in planktonic species in at ca. 1980 and 321 later (ca. 1995) a total disappearance of the benthic species Nitzschia valdestriata and 322 appearance of true planktonic taxa. M AN US C RI PT 309 323 The large increase in sedimentation rate at the head of the estuary during the time of 325 mangrove deforestation was not accompanied by an increase in the average sediment grain 326 size. Similarly, no changes in grain size composition were observed following the 327 construction of the dam and weir in Dinh River. TE D 324 328 Sediments in the estuary had organic carbon and nitrogen contents (ca. 1.5% and ca. 330 0.15%, respectively) comparable to that found by Tue et al. (2012) for another Vietnamese 331 estuary, but lower than most mangrove lined tropical estuaries (Bouillon and Boschker, 332 2006; Kristensen et al., 2008). The gradual decrease of carbon and nitrogen with depth in 333 the sediment may indicate microbial degradation over time. Such a pattern is often 334 observed particularly in sediments as a rather steep and leveling-off decrease in 335 concentrations with depth, which reflects a decreasing reactivity of the organic matter with 336 age (Canfield et al. 2005, Burdige 2006). This is supported by the apparent preferential 337 removal of nitrogen with depth as indicated from the gradually increased C/N ratios, a 338 pattern that is commonly observed in marine sediments (Canfield et al. 2005). However, 339 the continued decrease observed here also suggests that an overall increase in organic AC C EP 329 11 ACCEPTED MANUSCRIPT deposition may have occurred over time. Trinh et al. (1979) actually found that in 1976 341 organic carbon and nitrogen content in NP-BC surface sediments varied between 1.75 and 342 0.07% and between 0.18 and 0.015 %, respectively. The levels found in the present study 343 are higher than that for both carbon and nitrogen, suggesting that nutrient levels, and thus 344 pelagic primary production, indeed were lower three decades ago and that the decrease in 345 organic matter with depth may indicate changes in organic deposition over time combined 346 with post-depositional biogeochemical modifications. 347 RI PT 340 4.3 Stable isotopes 349 The stable carbon isotope signature, δ13C, is a reliable indicator for the source of organic 350 matter (Sweeney et al, 1978; Andrews et al., 1998). Thus, a δ13C signal close to -5‰ may 351 indicate that the organic material is predominantly of marine origin while a signal near - 352 30‰ refers to organic material of predominantly terrestrial or mangrove (C3 plant) origin. 353 Accordingly, Tue et al (2011a) reported a δ13C value of -28.3‰ for mangrove leaves from a 354 Vietnamese estuary. The generally increasing δ13C signal in the sediments from ca -26‰ at 355 the head to ca. –22‰ at the mouth of the estuary, indicates a change in organic matter 356 input from terrestrial to marine origin towards the mouth of the estuary, which is 357 independent of mangrove deforestation (Figure 6). Tue et al (2011a) similarly reported a 358 gradient in δ13C from -25.4‰ in the upper reaches to -21.2‰ at the mouth of a mangrove- 359 lined estuary. The δ13C signal is almost completely stable throughout the examined 360 sediment depth at the mouth of the estuary, while the signal fluctuates at the head of the 361 estuary, possibly associated with changes in the intensity of mangrove deforestation, as 362 explained in section 4.2. The conclusion by Pham (1981) that 70% of the sediment in NP- 363 BC is terrestrially derived is confirmed by the low δ13C signals found at the head of the 364 estuary in the present study. D TE EP AC C 365 M AN US C 348 366 In contrast to the other sediment parameters in the NP-BC estuary, there are no spatial differences 367 in the δ15N signal. The δ15N signature of mangrove-derived material does not systematically 368 deviate from that of marine production (Bouillon et al. 2008), and is therefore not a valid indicator 369 of the organic source, but is merely used to determine trophic levels in food chains. The δ15N 370 signature of primary producers is controlled by the isotopic value of their N-source, the temporal 12 ACCEPTED MANUSCRIPT and spatial variation in N availability, and changes in their demand (Dawson et al., 2002). 372 However, elevated δ15N values have been recorded from urbanized areas exposed to sewage- 373 derived eutrophication (Cole et al. 2004, Bannon and Roman 2008). Accordingly, no strong sewage 374 influence is apparent at the head of the NP-BC lagoon, as indicated by the similar and relatively 375 low levels of δ15N. Nevertheless, the fluctuating signal with depth at the head of the estuary may, 376 together with the pattern of other parameters, be driven by erratic changes in sewage influence 377 coinciding with the disappearance of the mangrove forest and other anthropogenic activities. RI PT 371 378 4.4 Changes in diatom community 380 Diatoms can be valuable indicators of water quality, since they are sensitive to changes in 381 the environment regarding e.g. nutrient status, turbidity/light, temperature, salinity and 382 pH (Smol & Stoermer 2010). The dominant diatom species in this study are cosmopolitan 383 taxa with broad ecological tolerances commonly found in coastal areas. Hence the indicator 384 value of the individual species is limited. However, there are clear changes in diatom life- 385 forms in the sediment record, which can illustrate past changes at the study site. The most 386 marked shift in the diatom community occurred ca. 1985 from a predominantly benthic 387 community to dominance by planktonic species. Shifts from benthic to planktonic diatom 388 communities are often driven by increased turbidity in the water, leading to reduced light 389 penetration (Cooper & Brush, 1991; Weckström et al., 2007 and references therein) caused 390 by either high phytoplankton biomass (following increased nutrient levels) or by high 391 suspended sediment load in the water column (following mangrove deforestation). The 392 shift in diatom community composition in NP-BC (Figure 7) may be instigated by either of 393 these factors or by a combination. The slight increasing trend in nitrogen concentration 394 through time may have increased the phytoplankton biomass, but it is more likely that the 395 shift is caused mainly by the increased amount of fine-grained sediment load entering the 396 estuary since the mangrove forest removal. Recent studies have shown relatively slow 397 settling velocities of suspended material in the estuary (Andersen et al., in prep) causing a 398 permanent and high turbidity when the sediment loading is high. It is therefore likely that 399 the observed shift from predominantly benthic to planktonic diatoms is primarily driven by 400 the removal of the mangrove forest. The periodic shifts between the benthic Nitzschia 401 valdestriata and the (tycho)planktonic Cyclotella striata and Paralia sulcata in the diatom AC C EP TE D M AN US C 379 13 ACCEPTED MANUSCRIPT record prior to the mangrove deforestation are probably also caused by changes in 403 turbidity due to other unknown perturbations. However, these shifts in diatom community 404 composition in the bottom part of the core differ from the large shift in species composition 405 coinciding with the main mangrove deforestation, when true planktonic diatom taxa 406 become more abundant. RI PT 402 407 4.5 Main conclusions 409 Overall, the sediment core proxies show both a distinct horizontal gradient in the estuary 410 and clear temporal changes, coinciding with the timing of mangrove deforestation, with 411 consequences both for the physical and chemical properties as well as biological structure 412 of the estuary (exemplified by the diatom community). 413 414 415 416 417 M AN US C 408 1. There are clear signs of increased sedimentation at the head of the estuary during and after the time of mangrove disappearance. 2. The sediment in NP-BC is dominated by fine particles both now and in the past, with temporal variability at the head of the estuary. 3. The variability in grain size distribution over time at the head of the estuary, with concurring shifts in C and N content, as well as δ13C and δ15N may be linked to variability in 419 the intensity of mangrove deforestation. TE 420 D 418 4. Relatively low contents of C and N and low δ13C signals indicate that the estuary is not greatly organically enriched and that the material in the estuary is predominantly of 422 terrestrial (and mangrove) origin, although gradually mixed with more marine material 423 near the mouth. 5. The diatom assemblage shows indications of increased turbidity in the water after ca. 1985, AC C 424 EP 421 425 most likely caused by increased supply of fine grained material to the estuary after the 426 disappearance of the mangrove forest and/or an increase in primary productivity in the 427 estuary, driven by increased N levels. 428 429 5. Acknowledgements 430 431 This study was part of the project ClimeeViet “Climate Change and Estuarine Ecosystem in Viet 432 Nam” (CLIMEEViet) funded by DANIDA –Fellowship Center under project code “P2-08-Vie”. 14 ACCEPTED MANUSCRIPT 6. References 434 Agardy, T., Alder, J., Dayton, P., Curran, S., Kitchingman, A., Wilson, M., Catenazzi, A., Restrepo, J., 435 Birkeland, C., Blaber, S., Saifullah, S., Branch, G., Boersma, D., Nixon, S., Dugan, P., Davidson, N., 436 Vörösmarty, C., 2005. Coastal Systems. UN Millennium Ecosystem Assessment. Chapter 19. pp. 437 513-548. RI PT 433 438 Alfaro, A. C., 2010. Effects of mangrove removal on benthic communities and sediment 440 characteristics at Mangawhai Harbour, northern New Zealand. ICES Journal of Marine Science, 67, 441 1087-1104. M AN US C 439 442 443 Alongi, D.M., 2008. Mangrove forests: resilience, protection from tsunamis, and responses 444 to global climate change. Estuarine, Coastal and Shelf Science 76, 1–13 445 446 Andrews, J.E., Greenaway, A.M., Dennis, P.F., 1998. Combined carbon isotope and C/N ratios as 447 indicators of source and fate of organic matter in a poorly flushed, tropical estuary: Hunts Bay, 448 Kingston Harbour, Jamaica. Estuarine Coastal and Shelf Science 46, 743-756. D 449 Appleby, P.G., 2001. Chronostratigraphic techniques in recent sediments. In: Tracking 451 environmental change using lake sediments. Volume 1: Basin analysis, coring and chronological 452 techniques. W.M. Last & J.P. Smol (Eds.) Kluwer Academic Publishers, the Netherlands. TE 450 EP 453 Atkinson, R. J. A., Frogliae, C., Arneri, E., Antolini, B., 1997. Observations on the Burrows 455 and Burrowing Behaviour of Squilla mantis (L.) (Crustacea: Stomatopoda). P.S.Z.N.: Marine 456 Ecology, 18, 337-359. 457 AC C 454 458 Bannon, R.O., Roman, C.T., 2008. Using stable isotopes to monitor anthropogenic nitrogen inputs 459 to estuaries. Ecological Applications 18, 22-30 460 461 Battarbee, R., W., Jones, V. J., Flower, R. J., Camerons, N.G., Bennion, H., 2001. Diatoms. In: Tracking 462 Environmental Change Using Lake Sediments. J. P. Smol, H. J. B. Birks & W. M.Last (Eds.), Volume 3: 15 ACCEPTED MANUSCRIPT 463 Terrestrial, Algal, and Siliceous Indicators. Kluwer Academic Publishers, Dordrecht, the 464 Netherlands. 465 Bouillon, S., Boschker, H.T.S., 2006. Bacterial carbon sources in coastal sediments: a cross system 467 analysis based on stable isotope data of biomarkers. Biogeosciences 3, 175–185. RI PT 466 468 469 Bouillon, S., Connolly, R.M., Lee, S.Y., 2008. Organic matter exchange and cycling in mangrove 470 ecosystems: Recent insights from stable isotope studies. Journal of Sea Research 59, 44–58. M AN US C 471 472 Brinkman, R., Wattayakorn, G., Wolanski, E., Spagnol, S., Marshall, K., 2005. Storm-driven 473 erosion of fine sediment and its subsequent transport and trapping in fringing mangroves, 474 Sawi Bay, Thailand. Journal of Coastal Research 42, 211-220. 475 476 Bui, H. L. (ed.), 1997. Investigation of the environmental status and marine resources in NhaPhu – 477 BinhCang (KhanhHoa). Institute of Oceanography unpublished research (In Vietnamese), 150 478 pages. D 479 Burdige, D.J., 2006. Geochemistry of marine sediments. Princeton University Press, New Jersey, 481 609 pp. TE 480 482 484 Canfield, D.E., Kristensen, E., Thamdrup, B., 2005. Aquatic Geomicrobiology. Academic Press. EP 483 Carmichael, R.H., Valiela, I., 2005. Coupling of near-bottom seston and surface sediment 486 composition: changes with nutrient enrichment and implications for estuarine food supply and 487 biogeochemical processing. Limnology and Oceanography 50, 97–105. 488 AC C 485 489 Cole, M.L., Valiela, I., Kroeger, K.D., Tomasky, G.L., Cebrian, J., Wigand, C., McKinney, R.A., Grady, 490 S.P., da Silva, M.H.C., 2004. Assessment of a delta N-15 isotopic method to indicate anthropogenic 491 eutrophication in aquatic ecosystems. Journal of Environmental Quality 33, 124-132. 492 493 Cooper SR, Brush GS (1991) Long-term history of Chesapeake Bay anoxia. Science 254:992–996. 16 ACCEPTED MANUSCRIPT 494 Dahdouh-Guebas, F., Jayatissa, L.P., Di Nitto, D., Bosire, J.O., Lo Seen, D., Koedam, N., 2005. How 496 effective were mangroves as a defence against the recent tsunami? Current Biology 15, 1337– 497 1338. 498 RI PT 495 499 Dawson, T.E., Mambelli, S., Plamboeck, A.H., Templer, P.H., Tu, K.P., 2002. Stable isotopes in plant 500 ecology. Annual Reviews in Ecology and Systematics 33, 507–559. 501 Dittmar, T., Hertkorn, N., Kattner, G., Lara, R.J. 2006. Mangroves, a major source of dissolved 503 organic carbon to the oceans. Global Biogeochemical Cycles 20, GB1012. M AN US C 502 504 505 Donato, D. C., Kauffman, J. B., Murdiyarso, D., Kurnianto, S., Stidham, M., Kanninen, M., 2011. 506 Mangroves among the most carbon-rich forests in the tropics. Nature Geoscience Letters. 507 NGEO1123. 508 Frignani, M., Piazza, R., Bellucci, L. G., Cu, N. H., Zangrando, R., Albertazzi, S., Moret, I., Romano, S., 510 Gambaro, A. 2007. Polychlorinated biphenyls in sediments of the Tam Giang-Cau Hai Lagoon, 511 Central Vietnam Chemosphere 67, 1786-1793 TE D 509 512 Furukawa, K., Wolanski, E., Mueller, H., 1997. Currents and sediment transport in mangrove 514 forests. Estuarine, Coastal and Shelf Science 44: 301–310. 515 EP 513 Granek, E.F., Ruttenberg, B.I. 2008. Changes in biotic and abiotic processes following mangrove 517 clearing. Estuarine, Coastal and Shelf Science 80, 555–562. 518 519 520 AC C 516 Hong, P.N., San, H.T. 1993. Mangroves of Vietnam. IUCN Bangkok, Thailand, 1-173. 521 Jennerjahn, T.C., Ittekkot, V. , 2002. Relevance of mangroves for the production and deposition of 522 organic matter along tropical continental margins. Naturwissenschaften 89, 23-30. 523 17 ACCEPTED MANUSCRIPT 524 Juggings, S., 2007. C2 Data analysis v.1.5.1.Computer program. Department of Geography, 525 University of Newcastle, UK. 526 Kristensen, E., Bouilllon S., Dittmar T., Marchand, C., 2008. Organic carbon dynamics in mangrove 528 ecosystems: A review. Aquatic Botany 89, 201-219. 529 RI PT 527 Lund-Hansen, L.C., Hai D.N., Lam N.N., Nielsen M.H., 2010. Optical properties of a tropical estuary 531 during wet and dry conditions in the Nha Phu Estuary, Khanh Hoa Province, south-east Vietnam. 532 Hydrobiologia 644, 207-216. M AN US C 530 533 534 Lund-Hansen, L.C., Nielsen J., Blüthgen J., Hai D.N., Nielsen, M.H., Lam N.H., 2013. Estuarine 535 morphometry governs optically active substances, Kd(PAR) and beam attenuation: assessment 536 from a tropical ria and a temperate coastal plain estuary. Hydrobiologia, DOI: 10.1007/s10750- 537 013-1457-1 538 Mazda, Y., Magi, M., Nanao, H., Kogo, M., Miyagi, T., Kanazawa, N., Kobashi, D., 2002. Coastal erosion 540 due to long-term human impact on mangrove forests. Wetlands Ecology and Management 10, 1-9. D 539 TE 541 Mcleod, E., Chmura, G.L., Bouillon, S., Salm, R., Björk, M., Duarte, C.M., Lovelock, C.E., Schlesinger, 543 W.H., Silliman, B.R., 2011. A blueprint for blue carbon: toward an improved understanding of the 544 role of vegetated coastal habitats in sequestering CO2. Frontiers in Ecology and the Environment. 545 9, 552-560. AC C 546 EP 542 547 Nguyen N. L., Nguyen T. M. A., Doan N. H., Ho V. T., 2006. Phytoplankton in Nha Phu lagoon, Viet 548 Nam. Selection of Marine Research Work, Inst. of Oceanography, Science and Technique 549 Publishing House, Hanoi, Viet Nam 15, 117-135. In Vietnamese, with English abstract. 550 551 Hanh Kim Nguyen, H. K, Kristensen, E., Lund-Hansen, L. C., 2013. Benthic metabolism and nitrogen 552 transformations affected by fish cage farming in the tropical Nha Phu estuary (Vietnam). Marine 553 and Freshwater Research, 63, 887–897. 554 18 ACCEPTED MANUSCRIPT 555 Pham, V. T., 1981. Surface sediment of coastal waters of PhuKhanh province. Selection of Marine 556 Research Work, Inst. of Oceanography, Science and Technique Publishing House, Hanoi, Viet Nam 557 2, 179-188. In Vietnamese, with English abstract. 558 Prasad, M.B.K., Ramanathan, A.L., 2008. Sedimentary nutrient dynamics in a tropical estuarine 560 mangrove ecosystem. Estuarine Coastal and Shelf Science 80, 60-66. 561 RI PT 559 Renberg, I., 1990. A procedure for preparing large sets of diatom slides from sediment cores. 563 Journal of Paleolimnology 4, 87-90. M AN US C 562 564 565 Rezende, C.E., Lacerda, L.D., Ovalle, A.R.C., Silva, C.A.R., Martinelli, L.A., 1990. Nature of POC 566 transport in a mangrove ecosystem: a carbon stable isotopic study. Estuarine, Coastal and Shelf 567 Science 30, 641–645. 568 Ruiz-Fernandez, A.C., Sprovieri, M., Piazza, R., Frignani, M., Sanchez-Cabeza, J.A., Feo, M.L., Belluci, 570 L.G., Vecchiato, M., Perez-Bernal, L.H., Paez-Osuna, F., 2012. Pb-210-derived history of PAH and 571 PCB accumulation in sediments of a tropical inner lagoon (Las Matas, Gulf of Mexico) near a major 572 oil refinery. Geochimica et Cosmochimica Acta 82, 136-153. TE D 569 573 Smol, J.P & Stoermer, E.F., 2010. The Diatoms: Applications for the Environmental and 575 Earth Sciences. Book 686 pp. Cambridge University Press. 576 EP 574 Strehlow, H. V., 2006. Integrated natural resources management of coastal fisheries. The Case of Nha Phu Lagoon, Vietnam. Unpublished PhD thesis. 580 Sulzman, E. W., 2008 Stable isotope chemistry and measurement: a primer. In: Michener, R., 581 Lajtha, K. (Eds.) Stable isotopes in Ecology and Environmental Science. Second edition. AC C 577 578 579 582 583 Sweeney, R. E., Liu, K. K. & Kaplan, I. R., 1978 Oceanic nitrogen isotopes and their use in 584 determining the sources of sedimentary nitrogen. In: B. W. Robinson (Ed.) Stable isotopes in the 19 ACCEPTED MANUSCRIPT 585 Earth Sciences, pp 9-26. Bulletin 220, Department of Scientific and Industrial Research, 586 Wellington. 587 Trinh, P., Pham, V. T., Nguyen, T. S., Trinh, T. H., Tran, H., Tran, D. T., Nguyen, H. S., 1979. 589 Geomorpholigical and Sedimentological characteristics in BinhCang – NhaPhu Bay. Collection of 590 Marine Research 1, 77-92. In Vietnamese, with English abstract. 591 RI PT 588 Tue, N.T., Hamaoka, H., Sogabe, A., Quy, T.D, Nhuan, M.T., Omori, K., 2011a. Sources of sedimentary 593 organic carbon in mangrove ecosystems from Ba Lat Estuary, Red River, Vietnam. In: Omori, K., 594 Guo, X., Yoshie, N., Fujii, N., Isobe, I.C, Tanabe, S (Eds.) Interdisciplinary studies on environmental 595 chemistry – marine environmental modeling and analysis,pp. 151-157. M AN US C 592 596 597 Tue, N.T., Hamaoka, H., Sogabe, A., Quy, T.D, Nhuan, M.T., Omori, K., 2011b. The application of δ13C 598 and C/N ratios as indicators of organic carbon sources and palaeoenvironmental change of the 599 mangrove ecosystem from Ba Lat Estuary, Red River, Vietnam. Environmental Earth Sciences 600 64,1475-1486. D 601 Tue, N.T., Ngoc, N.T., Quy, T.D., Hamaoka, H.. Nhuan, M.T., Omori, K., 2012. A cross-system analysis 603 of sedimentary organic carbon in the mangrove ecosystems of Xuan Thuy National Park, Vietnam. 604 Journal of Sea Research 67, 69-76 TE 602 EP 605 Twilley, R.R., Rivera-Monroy, V.H., 2009. Ecogeomorphic Models of Nutrient Biogeochemistry for 607 Mangrove Wetlands. In: Perillo G.M.E., Wolanski E., Cahoon D.R., Brinson M.M. (Eds.), Coastal 608 Wetlands: An Integrated Ecosystem Approach. Elsevier. 609 AC C 606 610 Umezawa, Y., Komatsu, T., Yamamuro, M., Koike, I., 2009.Physical and topographic factors affecting 611 suspended particulate matter composition in a shallow tropical estuary. Marine Environmental 612 Research 68, 59–70. 613 614 Valiela, I., Bowen, J.L., York, J.K., 2001. Mangrove forests: one of the world’s threatened major 615 tropical environments. Bioscience 51, 807–815. 20 ACCEPTED MANUSCRIPT 616 617 Weckström K., A. Korhola & J. Weckström 2007. Impacts of eutrophication on diatom life forms and 618 species richness in coastal waters of the Baltic Sea. Ambio 36: 155-160. 619 Wilkie, M.L., Fortuna, S., 2003. Status and trends in mangrove area extent worldwide. Forest 621 Resources Assessment Working Paper No. 63. FAO, Rome. 622 RI PT 620 Wolff, M., 2006. Biomass flow structure and resource potential of two mangrove estuaries: 624 insights from comparative modeling in Costa Rica and Brazil. Revista de Biologia Tropical 54, 625 Suppl 1, 69-86 AC C EP TE D M AN US C 623 21 ACCEPTED MANUSCRIPT 626 627 628 Proxy responses RI PT Event Dating horizon at station 3 Construction of first weir First signs of Mangrove cutting Dating horizon at station 5 Construction of dam Grain size shift at station 3 Initial decline in Mangrove covered area Diatom shift Shifts in proxies at station 5 M AN US C Approx. year 1940 1963 1973 1980 1983 1985 1993 1993-1998 1998-1999 Construction of second weir Maximal mangrove cutting Maximal expansion of shrimp ponds 629 Table 1: Timing of events in the Nha Phu-Binh Cang estuary based on core dating, satellite imagery 630 as well as physical, chemical and biological properties of the sediment. AC C EP TE D 631 22 ACCEPTED MANUSCRIPT 632 Figure 1 633 Map of NP-BC showing the locations of the three sampling stations 5, 77and 3 as well as the 634 approximate location of two villages, Tam Ich and Tan Dao (referred to in the discussion) 635 Figure 2 637 Time series maps of the NP-BC estuary head covering 6 selected years beween1973-2002. The 638 images are based on satellite footage and show the gradual decimation of the area covered by 639 mangrove forests (green areas) and the increase of the area covered by aquaculture ponds 640 (blue/purple area) 641 M AN US C RI PT 636 642 Figure 3 643 The extent of the areas covered by mangrove forests and aquaculture ponds over time. Filled 644 squares show the area of mangrove forests in hectares (left axis) and open squares the area of 645 aquaculture ponds in hectares (right axis). 646 Figure 4 648 The activity of unsupported 210Pb (beq kg-1) over depth in the cores from stations 3, 5 and 77. 649 Left: Station 3 (red), Right Stations 5 (blue; filled circles) and 77 (green; open circles). 650 Note the abrupt shift at ca. 60 and 40 cm in the cores from the head of the estuary (core 5 and 77). TE D 647 651 Figure 5 653 Age-depth model output for the dated cores from the estuary head (cores 5;blue and 77;green) 654 and mouth (core 3; red) based on 210Pb dating, assuming constant rate of supply of unsupported 655 210Pb AC C EP 652 (CRS-model) 656 657 Figure 6 658 Depth profiles of fine-grained material (silt and clay), N concentration, C concentration, C/N ratios, 659 δ13C and δ15N signal at stations 3 (black) and 5 (grey). Surface sample values from station 77 are 660 shown with an *. 661 662 Figure 7 23 ACCEPTED MANUSCRIPT 663 Stratigraphic diagrams of changes in diatom species composition over time. Only dominant 664 species are shown. The last panel shows the total proportion of planktonic diatoms (grey 665 area). The remainder is the proportion of benthic diatoms. All species and raw counts can 666 be downloaded from (supplementary material). RI PT 667 668 669 670 AC C EP TE D M AN US C 671 24 ACCEPTED MANUSCRIPT o 109.2 E AC C EP TE D M AN US C RI PT Figure 1 5 km o 12.4 N AC C EP TE D M AN US C RI PT ACCEPTED MANUSCRIPT M AN US C RI PT ACCEPTED MANUSCRIPT AC C EP TE D Figure 3 EP AC C Core 3 TE D M AN US C RI PT ACCEPTED MANUSCRIPT Cores 5 (blue; filled circles) and 77 (green; open circles) ACCEPTED MANUSCRIPT RI PT Figure 5 Core 77 M AN US C Core 3 AC C EP TE D Core 5 M AN US C RI PT ACCEPTED MANUSCRIPT AC C EP TE D Figure 6 D M AN US C RI PT ACCEPTED MANUSCRIPT AC C EP TE Figure 7 ACCEPTED MANUSCRIPT Supplementary material (Raw counts of diatoms) 0 0 0 1 0 0 0 0 0 0 0 Actinocyclus ochotensis 0 0 1 2 0 0 0 1 0 0 0 Actinocyclus cf. ortonarius 0 2 0 0 0 0 0 0 0 0 0 Actinocyclus sp. 0 0 1 0 0 0 0 0 1 0 0 Auliscus sp. 0 0 1 0 0 0 0 0 1 0 0 Bacteriastrum sp. 12 0 0 0 0 0 0 0 0 0 0 Bacteriastrum varians 1 0 0 0 0 0 0 0 0 0 0 Chaetoceros lorenziana 2 0 0 0 0 0 0 0 0 0 0 Chaetoceros messanensis 0 3 0 0 0 0 0 0 0 0 0 Chaetoceros sp. 10 0 0 0 0 0 0 0 0 0 0 Cocconeis costata 0 0 0 0 0 0 0 2 0 0 0 Coscinodiscus sp. 0 0 0 0 1 0 0 0 0 0 0 Cyclotella meneghiania 2 1 0 1 2 0 1 0 0 0 0 Cyclotella striata 20 59 36 49 30 22 17 52 19 41 3 Cyclotella stylorum 0 0 1 1 0 2 0 0 2 4 0 Hemiaulus sinensis 6 0 0 0 0 0 0 0 0 0 0 Paralia sulcata 20 43 25 46 0 8 18 43 18 46 3 Rhizosolenia bergonii 1 0 0 0 0 0 0 0 2 0 0 Rhizosolenia crassispina 2 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 0 3 0 0 0 0 0 0 0 0 Triceratium scitulum 0 0 0 0 0 0 0 1 0 0 0 Achnanthes cf. parvula 2 0 0 0 0 0 0 0 0 0 0 Achnanthes sp. 0 0 0 1 0 0 0 2 1 0 0 Amphora angutissima 0 0 1 0 0 0 0 0 0 0 0 Amphora laevis 1 0 0 0 0 0 0 0 0 0 0 Amphora graeffeana 1 0 0 0 0 0 0 0 0 0 0 Amphora tp. Immarginata 0 0 0 1 0 0 0 0 0 0 0 Amphora ovalis 1 0 0 0 0 0 0 0 0 0 0 AC C Triceratium favus D TE EP Triceratium dubium M AN US C Actinocyclus normanii RI PT 1,5 5,5 9,5 15,5 17,5 21,5 25,5 29,5 31,5 36,5 47,5 1 1 0 0 0 0 0 2 0 0 0 Campylodiscus undulatus 0 1 1 0 0 0 1 0 0 1 0 Cocconeis scitulum 0 0 1 0 0 0 0 0 0 0 0 Cocconeis sp. 0 1 0 0 0 0 0 0 0 0 0 Delphineis cf. surirella 0 0 0 1 0 0 0 0 0 0 0 Denticula sp. 0 0 0 2 0 0 0 0 0 0 0 Diploneis bombus 8 19 3 1 1 1 2 4 0 2 0 Diploneis cabro 0 0 0 1 0 1 0 0 0 0 0 Diploneis cf. nitescens 0 0 0 0 0 2 0 0 0 0 0 Diploneis coffaeiformis 1 0 3 1 0 0 0 0 0 0 0 Diploneis fusca 0 0 2 0 0 0 0 0 0 0 0 Diploneis smithii 0 2 0 0 2 0 1 1 0 0 0 Diploneis spp. 1 0 0 0 0 0 0 0 0 2 0 Diploneis weissflogii 5 8 4 2 5 0 0 3 0 0 0 Eunotogramma laevis 0 0 1 1 0 0 0 0 0 0 0 marinum 0 0 0 1 0 0 0 0 0 0 0 Gomphonema sp. 2 0 0 1 0 0 0 1 0 0 0 Grammatophora oceanica 0 3 0 0 1 1 0 0 0 0 0 Grammatophora sp. 0 0 0 1 0 0 2 0 6 3 0 Grammatophora undulata 0 1 1 3 0 0 0 0 0 1 0 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0 Lyrella cf. atlantica 0 0 0 0 2 0 0 0 0 0 0 Lyrella cf. spectabilis 0 1 0 0 0 1 0 0 0 0 0 Lyrella sp. 0 0 0 0 1 0 0 1 0 0 0 Mastogloia fimbriata 1 0 0 0 0 0 0 0 0 0 0 Navicula cf. bipustulata 1 0 0 0 0 0 0 0 0 0 0 Navicula spp. 8 0 1 0 0 0 1 2 0 4 0 Navicula tp. obtusangula 1 0 0 0 0 0 0 0 0 0 0 Navicula tp. palpebralis 1 0 0 0 0 0 0 0 0 0 0 AC C Lyrella lyra TE EP Gyrosigma sp. D Eunotogramma cf. M AN US C Amphora sp. RI PT ACCEPTED MANUSCRIPT 0 0 0 1 0 0 0 0 0 0 0 Nitzschia cf. coarctata 4 0 0 0 1 0 0 0 0 0 0 Nitzschia cf. palea 0 0 2 0 0 0 0 0 0 0 0 Nitzschia lorenziana 3 0 0 0 0 0 0 0 0 0 0 Nitzschia panduriformis 4 0 1 0 0 0 1 1 0 1 0 Nitzschia spp. 1 1 1 2 0 0 0 2 0 2 1 Nitzschia tp. angularis 2 1 0 0 0 0 0 0 0 0 0 Nitzschia tp. gregaria 1 0 0 0 0 0 0 0 0 0 0 Nitzschia tp. persuadens 1 0 0 0 0 0 0 0 0 0 0 Nitzschia tp. plioveterana 4 0 0 0 0 0 0 0 0 0 0 Nitzschia tp. subconstricta 4 0 0 0 0 0 0 0 0 0 0 Nitzschia valdestriata 0 0 66 0 211 361 506 0 580 27 395 Opephora sp. 0 0 1 1 0 0 0 0 0 0 0 Petroneis marina 0 2 0 2 0 0 0 0 1 0 0 Plagiogramma puchellum 0 0 0 1 0 0 0 0 0 0 0 1 0 0 1 0 3 0 0 0 1 0 1 2 2 0 3 2 0 3 0 0 0 1 0 0 0 0 0 0 0 0 0 3 0 0 1 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 2 4 1 3 0 0 1 0 0 0 0 Surirella ovalis 0 0 0 0 1 0 0 0 0 0 0 Terpsinoe americana 0 0 0 1 0 0 0 0 0 0 0 8 1 5 4 0 0 3 0 1 0 staurophorum 6 Rhaphoneis amphiceros 0 1 Rhaphoneis sp. 0 Stauroneis sp. 1 AC C Surirella fastuosa EP Stauronella arctica TE Pleurosigma naviculaceum 55 D Plagiogramma Thalassionema frauenfeldii 41 M AN US C Neofragilaria nicobarica RI PT ACCEPTED MANUSCRIPT Thalassionema nitzschioides 47 47 12 9 15 0 9 7 0 10 0 Thalassiosira eccentrica 1 1 1 0 1 0 0 0 0 1 0 Thalassiosira oestrupii 14 19 1 4 3 1 3 4 1 6 0 Thalassiosira spp. 1 1 2 3 2 0 0 1 0 2 0 ACCEPTED MANUSCRIPT Trachyneis aspera 4 1 1 2 0 2 0 0 0 1 0 taxa 47 28 30 33 20 12 15 21 13 20 4 Sum 305 239 177 154 289 405 567 136 634 159 402 AC C EP TE D M AN US C RI PT Number of encountered