Dissolved organic matter from tropical peatlands impacts shelf sea light availability on coral reefs in the Singapore Strait, Southeast Asia

Shelf seas provide valuable ecosystem services, but their productivity and ecological functioning depend critically on sunlight transmitted through the water column. Anthropogenic reductions in underwater light availability are thus recognized as a serious threat to coastal habitats. The flux of strongly light-absorbing coloured dissolved organic matter (CDOM) from land to sea may have increased world-wide, but how this has altered the availability and spectral quality of light in shelf seas remains poorly known. Here, we present time-series data from the Sunda Shelf in Southeast Asia, where the monsoon-driven reversal in ocean currents supplies water enriched in CDOM from tropical peatlands for part of the year, resulting in 5–10-fold seasonal variation in light absorption by CDOM. We show that this terrigenous CDOM can dominate underwater light absorption at wavelengths up to 500 nm, and shift in the underwater irradiance spectrum towards longer wavelengths. The seasonal presence of terrigenous CDOM also causes the depth of 10% light penetration to shoal by 1–5 m, or 10–45%. We further estimate that on average 0.6 m, or 25%, of this terrigenous CDOM-mediated shoaling might be attributable to the enhanced loss of dissolved organic matter caused by peatland disturbance. We show that the seasonal change in the light environment is correlated with photo-acclimation by phytoplankton, and infer that terrigenous CDOM likely contributes to limiting the depth distribution of photosynthetic corals. Our results thus reveal an ecologically important but largely overlooked impact of human modifications to carbon fluxes that is likely becoming increasingly important in coastal seas.

Terrigenous CDOM thus has the potential to spread extensively across shelf seas, and 7 increases in tDOC flux might therefore affect the light environment over large areas of 8 coastal ocean. 9 That terrigenous, as opposed to marine, CDOM can significantly affect the underwater light However, although the importance of terrigenous CDOM for the light environment of coastal 1 waters has clearly been recognised, the relative contributions of terrigenous and marine 2 CDOM to light attenuation in shelf seas have usually not been partitioned quantitatively. 3 Consequently, although large-scale anthropogenic changes in land-ocean tDOC fluxes have 4 the potential to alter the light environment of shelf seas by altering the terrigenous CDOM 5 concentration, our knowledge of such impacts is still very limited. Based on long-term trends 6 in salinity and correlations between salinity and CDOM in Norwegian fjords, Aksnes et al. 7 (2009) inferred that increased terrigenous CDOM had resulted in "coastal browning", which 8 may have contributed to mesopelagic regime shifts from fish (visual predators) to jellyfish 9 (tactile predators). Similarly, an increase in non-autotrophic particulate organic carbon in 10 southern Norway was interpreted as indicating an increase in terrigenous CDOM, and this 11 was hypothesised to have contributed to the collapse of coastal kelp forests (Frigstad et al. 12 2013). An increase in terrigenous CDOM was also invoked as a possible mechanism driving 13 decreased water clarity in the North Sea (Dupont & Aksnes 2013), which may have caused a 14 shift in the timing of the spring phytoplankton bloom by up to three weeks (Opdal et al. 15 2019). Ecosystem modelling has additionally demonstrated that increased terrigenous CDOM 16 absorption can lead to a shallower distribution of phytoplankton and a shallower nutricline, 17 resembling symptoms of eutrophication (Urtizberea et al. 2013 This contrasts with our better understanding of the ecological impacts of freshwater "lake been reported contain far higher concentrations of terrigenous CDOM than do shelf seas, so 5 much so that filtered lake water can have a noticeably yellow-brown colouration (Solomon et 6 al. 2015). Whether CDOM-mediated browning can affect the ecological functioning of shelf 7 seas to the same degree as lakes is still unclear. 8 Here, we use biogeochemical and optical time-series data from the Singapore Strait in the 9 Sunda Shelf Sea in Southeast Asia to estimate seasonal changes in the proportion of marine 10 and terrigenous CDOM, and then to quantify how terrigenous CDOM impacts underwater

Study area 21
The Singapore Strait is located in the central Sunda Shelf Sea, close to the peatlands on 22 Sumatra (Fig. 1). The ocean currents on the Sunda Shelf reverse direction seasonally (van were sampled occasionally to constrain spatial variability (Fig. 1). Conductivity-temperature-3 depth profiles were measured with a Valeport FastCTD at 16 Hz to 12-15 m depending on 4 current and bottom depth; stratification was not observed. Water was collected adjacent to the 5 reefs at 5 m depth using a Niskin bottle. Samples for CDOM and DOC analysis were filtered 6 on board through 0.2 µm Supor polyethersulfone filters (all tubing and filter housings were 7 cleaned with 1 M HCl, then assembled with the filters and pre-rinsed with 200 mL of 8 ultrapure water (18.2 MW cm -1 ) and with sample water;) and stored in 40-mL amber 9 borosilicate EPA vials (pre-baked at 450°C for 4 h) with Teflon-coated septa. Water for 10 chlorophyll-a and particulate absorption was stored in acid-washed HDPE bottles in the dark 11 and filtered (25-mm Whatman GF/F) 3-6 hours later in the laboratory. 12 Backscattering was measured at 412, 440, 488, 510, 532, 595, 650, 676, and 715 nm using a 13 Wetlabs BB9 lowered to 1 m depth; 60 consecutive measurements were taken at 1 Hz and 14 averaged. The data were processed according to manufacturer instructions: the raw 15 measurement was converted to the total volume scattering coefficient by subtracting the dark 16 offset (measured before each field trip) and multiplying by a calibration scaling factor, and 17 corrected for non-water absorption as measured by a TriOS OSCAR instrument. Volume 18 scattering from pure seawater was subtracted and the particulate scattering converted to 19 particulate backscattering coefficients following Boss and Pegau (2001), which were then fit 20 with a power law at 1-nm resolution over the wavelength range of photosynthetically active 21 radiation (PAR, 400-700 nm). 22 The present analysis focuses on the period of December 2018 to August 2020, during which 23 we acquired 77 measurements of CDOM, DOC, and salinity, 60 measurements of particulate 24 absorption, and 36 measurements of backscattering.

CDOM and particulate absorption measurements 2
CDOM samples were stored at +4°C back on land and analysed within 24 h of collection. 3 Samples were brought to room temperature and absorbance measured from 250-800 nm at 1-4 nm resolution in 10-cm pathlength quartz cuvettes on a Thermo Evolution300 dual-beam 5 spectrophotometer against ultrapure water as a reference. Data were baseline-corrected 6 according to Green and Blough (1994), smoothed using a loess function, and converted to 7 Napierian absorption coefficients, using the R package hyperSpec (Beleites & Sergo 2018). 8 Here, we express the concentration of CDOM as the CDOM absorption coefficient at 440 9 nm, aCDOM(440), with units of m -1 . We also calculated the CDOM spectral slope from 275-10 295 nm, S275-295, and the spectral slope ratio (SR, the ratio of the 275-295 nm slope to the 11 350-400 nm slope) following Helms et al. (2008). Both S275-295 and SR have been shown to 12 correlate with DOM molecular weight (Helms et al. 2008) and are widely used as markers of The theoretical mixing curves were plotted together with the measured data on scatter plots 2 of S275-295 and SR against aCDOM(440), following Stedmon and Markager (2001). 3 Samples for particulate absorption (500-1000 mL) were vacuum-filtered onto 25-mm 4 diameter Whatman GF/F filters and stored in liquid nitrogen in tissue embedding cassettes 5 (Kartell Labware) wrapped in aluminium foil. Samples were thawed to room temperature, 6 moistened by briefly placing them on a sponge soaked in filtered seawater, and absorbance 7 measured from 300-800 nm with filters held inside an integrating sphere using a centre-8 mount sample holder on a PerkinElmer Lambda 950 spectrophotometer, as recommended by 9 Stramski et al. (2015). Multiple blank filters were measured throughout each batch of 10 analysis. Filters were then depigmented (as assessed by the complete disappearance of the 11 chlorophyll-a absorption peak at 668 nm) with 5 ml of 0.1% sodium hypochlorite in ultrapure 12 water with 60 g l -1 sodium sulphate for 15 min (Ferrari & Tassan 1999), rinsed with 5 ml 13 ultrapure water, and remeasured. All blank and sample absorbance spectra were first 14 corrected for baseline drift by subtracting the mean absorbance from 801-851 nm from the 15 rest of the spectrum, and then blank-corrected by subtracting the mean baseline drift-16 corrected blank spectrum from all sample absorbance spectra. These corrected sample 17 absorbances were then corrected for pathlength amplification according to Stramski et al. 18 (2015): 19 where As is the pathlength-corrected sample absorbance, and Af is the blank-and baseline-21 corrected absorbance of each filter. Corrected absorbances were then converted to Napierian 22 absorption coefficients by accounting for the area of sample on each filter (the filtered area of 23 each filter had a radius of 11.5 mm) and the sample volume filtered. Phytoplankton 24 absorption (aphyto) was calculated by subtracting the depigmented absorption spectrum (i.e., the non-algal particulate absorption, aNAP) from the total particulate absorption spectrum. 1 From the aphyto spectra, we further calculated the phytoplankton absorption spectral slope Seasonal average spectral absorption budgets were calculated as the fractional contribution of 10 each absorbing constituent (CDOM, non-algal particles, phytoplankton, and water) to the 11 total absorption at each wavelength, and then averaging these data seasonally. where A254 is the CDOM absorbance at 254 nm per metre, and the DOC concentration is in 1 mg C l -1 (note that the absorbance is obtained by dividing the Napierian absorption 2 coefficient by 2.303). SUVA254 consequently has units of l mg -1 m -1 , and is a measure of 3 DOM aromaticity (Traina et al. 1990, Weishaar et al. 2003). Similar to S275-295 and SR, 4

Chlorophyll-a 8
Samples for chlorophyll-a (200-1000 ml) were filtered onto 25 mm diameter Whatman GF/F 9 filters, wrapped in aluminium foil, flash-frozen in liquid nitrogen, and stored at -80°C until 10 analysis within 3 months. Filters were then extracted in 90% acetone at 4°C in the dark 11 overnight, briefly centrifuged to remove particles, and fluorescence measured on a Horiba 12 Fluoromax4 at excitation 436 nm and emission 680 nm with slit widths of 5 nm 13 (Welschmeyer 1994). Fluorescence was acquired as the fluorescence signal normalised to the 14 lamp reference measurement to account for variation in lamp intensity (using the Fluoromax4 15 S1c/R1c acquisition mode), and calibration was performed with a spinach chlorophyll-a 16 standard (Sigma-Aldrich, C5753-1MG). 17 18

Calculating light attenuation spectra 19
Underwater light attenuation can be described by the diffuse attenuation coefficient of 20 downwelling irradiance, Kd, which varies spectrally: 21 where Ed(z,l) is the downwelling irradiance at depth z and wavelength l, Ed(0,l) is the 23 downwelling irradiance at wavelength l at the surface, and Kd(l) is the diffuse attenuation 24 coefficient at wavelength l. We used our spectral measurements of absorption by CDOM and particles, and backscattering by particles, to calculate spectra of Kd over the wavelength range 1 400-700 nm according to Lee et al. (2005): 2 where q is the solar zenith angle, a is total absorption, and bb is total backscattering.

4
Absorption and backscattering spectra of pure seawater were taken from Pope and Fry (1997) 5 and Smith and Baker (1981), respectively. We used the solar zenith angle at solar noon on 6 each date (i.e., the time of day when the sun is at its highest point), such that the result 7 reflects the maximum light penetration for each date. Solar zenith angles and solar noon 8 times were calculated using the R packages GeoLight (Lisovski & Hahn 2012) and suncalc. 9 This calculation was originally developed to estimate Kd between the surface and the depth to 10 which 10% of surface PAR penetrates (Z10%) and was therefore denoted

Calculating underwater irradiance spectra and depth of PAR penetration 20
To examine how seasonal variation in absorption and backscattering affect both the spectral 21 quality of irradiance underwater and the depth to which PAR penetrates, we used the Kd 22 spectra together with modelled mid-day solar irradiance for each date to calculate depth 23 profiles of underwater irradiance, the vertical attenuation coefficient of downwelling PAR, 24 Kd(PAR), and Z10%. Our objective with this analysis was not to derive the actual underwater irradiance on each date, which depends especially on cloud cover, but rather to determine the 1 potential effects of the observed variation in absorption and backscattering on the underwater 2 light environment. We therefore modelled the downwelling irradiance spectrum just below 3 the water surface (Ed0 -) for solar noon on each date using the Hydrolight model, assuming 4 identical cloud cover and wind speed for each day (20% and 2 m s -1 , respectively), and used 5 these modelled spectra as inputs for our calculations. This means that seasonal changes in 6 solar zenith angle (and their resulting effects on irradiance) are accounted for, but that our 7 results are otherwise representative of conditions experienced around mid-day on relatively 8 cloud-free days. Variation in cloud cover chiefly alters the total irradiance, but does not affect 9 the shape of the irradiance spectrum very strongly. Our purpose with these calculations was 10 not to estimate exact light doses, but rather to examine how the depth penetration and spectral 11 distribution of underwater light vary over time as a result of our measured changes in 12 absorption and backscattering, for which modelled irradiances are sufficient.
where Z is the depth of the water column. We selected 30 m, which is representative of much 18 of the Singapore Strait surrounding our sampling sites (Chan et al. 2006). 19 Next, to examine how the spectral light quality experienced by benthic organisms is affected, 20 we calculated the underwater irradiance spectrum at fixed depths within the upper 10 m for 21 each date by attenuating the Hydrolight-modelled noon-time Ed0spectra with the calculated 22 Kd spectra according to Eq. (5). 23 Finally, to examine how the overall depth of light penetration varies, we calculated Kd(PAR) 24 and Z10%. To do this, we first attenuated the modelled Ed0spectra with the calculated Kd spectra (Eq. 5) at 0.1 m intervals from the surface down to a depth of 20 m to yield calculated 1 depth profiles of downwelling irradiance (Ed). The calculated Ed spectrum at each depth was 2 then converted from W m -2 nm -1 to the downward flux of photons, Eq, at each wavelength 3 where h is Planck's constant and c is the speed of light in m s -1 . Eq was converted to µmol 6 photons m -2 s -1 and then summed across the wavelength range of 400-700 nm to yield a 7 quantum flux of PAR at each depth. Kd(PAR) was then calculated as the slope of a linear 8 regression of the natural log of quantum PAR flux versus depth, and Z10% was calculated as Consequently, the value of Kd(PAR) calculated by regressing PAR against depth varies 12 depending on the depth to which the regression is performed. Since our objective with this 13 calculation was to quantify Z10%, the regression should ideally be performed down to Z10% 14 rather than to a fixed, arbitrary depth, so we sought to first estimate the approximate depth of 15 Z10% to determine the appropriate depth to which to perform the regression. Using our 19 16 measured radiometer profiles (described in the Supplementary Information), we found that 17 Z10% was closely related to Kd at 520 nm: 18 where Kd(520) is Kd at 520 nm (Fig. S1). We used this initial estimate of Z10% for each station 20 as the depth over which we calculated Kd(PAR) using a PAR versus depth regression as 21 explained above. The final value of Z10% for each station was then calculated from Kd(PAR) 22 as described above. 23 24

Impact of terrigenous CDOM on Z10%
Our time-series of S275-295, SR, and SUVA254 indicated that the variation in CDOM absorption 1 is predominantly the result of conservative mixing between terrigenous CDOM and marine 2 CDOM, as shown in Section 3.1 below. Based on these data, the CDOM during the March-3 April intermonsoon period was predominantly marine, while the CDOM during other periods 4 consisted of a mixture of this background level of marine CDOM and a varying amount of 5 terrigenous CDOM. We therefore quantified the amount of terrigenous CDOM in each 6 sample by subtracting the intermonsoon CDOM spectrum measured on 15 March 2019 7 (which we also used as one endmember in our conservative mixing model; see Section 2.3) 8 from the measured CDOM spectrum in each sample. 9 To quantify the impact of this terrigenous CDOM on the depth of PAR penetration, we 10 recalculated our Kd spectra (Eq. 6) using the 15 March 2019 CDOM spectrum in place of the 11 CDOM spectrum measured for each station. We then recalculated Kd(PAR) and Z10%, as well 12 as Ed(Zmean), as described in Section 2.7. This yielded estimates of what Kd(PAR), Z10%, and 13 Ed(Zmean) would have been at each station in the absence of terrigenous CDOM. 14 To quantify the potential anthropogenic contribution to CDOM-mediated light attenuation, 15 we recalculated Kd again, this time with the terrigenous CDOM absorption reduced by 35% 16 of the observed value. This is based on estimates from Borneo and Sumatra that land-use 17 change has increased the flux of DOC from Southeast Asian peatlands by 54% (Moore et al. absorption, chlorophyll-a concentrations were relatively low (mean ± standard deviation of 3 1.0 ± 0.5 µg l -1 ; Fig. S4) and did not show significant differences between seasons (Kruskal-4 Wallis test, c 2 = 2.9, d.f. = 3, p = 0.40).

5
We found that our measured S275-295 and SR values showed tightly constrained relationships 6 with aCDOM(440) across all seasons (Fig. 3a,b), which closely followed the theoretical mixing 7 model (Eq. 1; grey lines in Fig. 3a,b) between the CDOM spectra measured on 15 March  The large seasonal changes in CDOM absorption altered the average spectral light absorption 16 budget between seasons (Fig. 4). Absorption in the UV range (300-400 nm) was dominated 17 by CDOM in all seasons, but to a greater extent in the SW Monsoon. Between 400-500 nm, 18 CDOM was progressively less dominant, and especially in the intermonsoon seasons the 19 absorption by CDOM at 500 nm was only around 20% of the total absorption (Fig. 4a,c). In 20 the SW Monsoon, however, CDOM contributed ≥50% of the total absorption up to 500 nm, 21 and still contributed 50% of the non-water absorption up to 600 nm (Fig. 4b). During the NE 22 Monsoon, the absorption budget was less CDOM-dominated than in the SW Monsoon, but 23 more than during the intermonsoon seasons (Fig. 4e). In all seasons, absorption by non-algal 24 particles was greater than phytoplankton absorption from 300 nm to roughly 440 nm, then up to 500-550 nm phytoplankton and non-algal particles contributed roughly equally, beyond 1 which phytoplankton increasingly dominated the particulate absorption. In all seasons, water 2 contributed >50% of absorption upwards of 550-570 nm (Fig. 4). 3 Using the modelled surface irradiance for solar noon on each sampling date, we found that 4 the underwater irradiance at fixed depths between 1 and 10 m was shifted to longer 5 wavelengths in the SW Monsoon: the wavelength of peak irradiance ranged from 531-539 6 nm during the intermonsoon and NE Monsoon seasons, but was shifted to between 547-566 7 nm in the SW Monsoon (Fig. 5). Similarly, the ratio of blue to green irradiance (calculated as 8 Ed (440) to Ed(550)) at each depth was significantly lower during the SW Monsoon than other This spectral shift in the underwater irradiance was also evident in the average irradiance at 12 solar noon experienced by phytoplankton under turbulent mixing (Eq. 7), which peaked at 13 567 nm during the SW Monsoon, but at 537-538 nm during the other seasons (Fig. 6). The

Z10% and impacts of terrigenous CDOM
The depth of 10% PAR penetration, Z10%, ranged between 3.7-9.8 m, with an overall average 1 of 7.1 m, and was typically deeper at the more sheltered site (Fig. 7a). Across both sites, Z10% 2 was on average deepest during Intermonsoon 1 (8.6 m) and shallowest during the SW 3 Monsoon (6.6 m); this seasonal difference was statistically significant (Kruskal-Wallis test, to a shoaling of Z10% by 13-45% due to terrigenous CDOM (Fig. 8a,b). Even during the NE 10 Monsoon, when the terrigenous CDOM concentration was lower than during the SW 11 Monsoon, the euphotic zone was shoaled by up to 1.9 m, or 17%. Across all seasons and 12 sites, the percentage shoaling of Z10% was strongly related to the terrigenous CDOM 13 concentration (Fig. 8c), which shows that terrigenous CDOM had a large impact on the depth 14 of underwater light penetration despite variation in the concentrations of suspended 15 sediments and phytoplankton between sites and dates. We also repeated our calculation of the 16 depth-averaged irradiance in a turbulent water column (Eq. 7), but using the Kd spectra 17 calculated without terrigenous CDOM (see Section 2.8). We found that without terrigenous 18 CDOM, the wavelength of peak irradiance was on average nearly identical between seasons 19 (531-534 nm), and that the blue-to-green ratio Ed (440)  To estimate how much of the observed shoaling of Z10% might be the result of anthropogenic 23 disturbance of peatlands, we repeated our calculations of Kd for the SW Monsoon but with 24 the terrigenous CDOM absorption reduced by 35% of the observed value (see Section 2.8).
We found that with only the estimated natural fraction of terrigenous CDOM, Z10% was on 1 average 0.6 m deeper than the actual observed values. This potentially anthropogenic 2 contribution to light attenuation accounted for on average 25% of the observed Z10% shoaling 3 by terrigenous CDOM (Table 1). 4

Sources and seasonality of CDOM 7
Our time series data showed strong and correlated seasonal variation in CDOM absorption, 8 markers of CDOM terrestrial origin (S275-295, SR, and SUVA254), and salinity. Moreover, the 9 variability in CDOM across all seasons followed a pattern that is consistent with simple 10 conservative mixing between two CDOM end-members (Fig. 3), where one end-member has 11 high aCDOM and SUVA254 with low S275-295 and SR (SW Monsoon) and the other has low  The fact that chlorophyll-a concentrations did not vary seasonally and were overall relatively 16 low for a coastal environment further indicates that production of autochthonous, marine 17 CDOM is unlikely to show strong seasonal variation (note that benthic communities such as mostly consists of mineral soils rather than peatlands (Fig. 1). 10 11

Impact of terrigenous CDOM on light availability 12
The strong light attenuation, and resulting shallow Z10% depth that we observed in our time and also alters the spectral quality of the available light. This was already evident when just 18 considering the seasonal averages of Z10%, even though the observed Z10% was also clearly 19 affected by the variability in particulate absorption and backscattering. After first estimating 20 the fraction of CDOM that was terrigenous, we could quantify its impact directly by 21 calculating hypothetically how the spectrum of Kd would differ in its absence; this showed 22 that the advection of terrigenous CDOM during both monsoon seasons leads to shoaling of 23 Z10% by tens of percent (Fig. 8). Moreover, this shoaling was accompanied by spectral shifts in underwater irradiance, leading to less blue light and an irradiance peak shifted towards 1 longer wavelengths (Figs. 5,6). also demonstrate that the spectral distribution of PAR is affected. 12 Our results also provide further support for the use of CDOM spectral slope measurements, 13 especially S275-295, to distinguish between marine and terrigenous CDOM in coastal waters 14

Ecological implications 16
Primary production by benthic communities and by phytoplankton requires sufficient light 17 availability, and strong extinction of PAR can therefore limit productivity and restrict the Our estimates of the depth-averaged underwater irradiance in a turbulent water column show 1 that phytoplankton in the Singapore Strait are subject to seasonal changes in intensity and 2 spectral composition of irradiance (Fig. 6). The fact that the phytoplankton absorption 3 spectral slope increased and the aPhyto(490):aPhyto(510) ratio decreased during the SW 4 Monsoon suggests that the phytoplankton were adjusting their pigment composition in 5 response to the changing light environment. Specifically, these data indicate a higher ratio of 6 photoprotective to photosynthetic carotenoid pigments during the intermonsoon, and a lower Our data also show that benthic communities within the upper 10 m are exposed to seasonal 22 changes both in total PAR intensity and in spectral quality of irradiance (Figs. 5,8). The 23 Singapore Strait is home to >100 different scleractinian coral species, but their depth range is It has been suggested that coral communities in shallow low-light environments such as 7 Singapore should be considered as mesophotic coral ecosystems, like those found below 30- shallow and deep mesophotic systems will prove to be ecologically significant, perhaps by 9 controlling lower depth limits and coral community composition, or requiring a greater 10 reliance on heterotrophic versus autotrophic nutrition (Anthony & Fabricius 2000). 11 The decline in coral cover at 6-7 m in the Singapore Strait since the 1980s reported by Guest  Our results indicate that if peatland disturbance has indeed increased tDOC fluxes by as 16 much as currently thought, then the associated reduction in light transmission due to 17 terrigenous CDOM has likely contributed to these benthic cover changes.

Conclusions 20
Our data demonstrate the importance of terrigenous CDOM for the optical properties of 21 peatland-influenced areas of the Sunda Shelf Sea, and show further that the seasonal, 22 monsoon-driven advection of this terrigenous CDOM drives significant variation in the 23 transmission and spectral quality of light underwater. We also observed seasonal variation in 24 phytoplankton absorption spectra that are indicative of changes in photo-acclimation, which suggests that this variation in light attenuation was ecologically relevant. Moreover, our study 1 suggests that land conversion in the tropics has the potential to cause CDOM-mediated 2 coastal browning in biodiverse shelf sea environments, which may have contributed to 3 observed coral cover decline. Overall, our study underscores the importance of examining not 4 only biogeochemical impacts of land-ocean tDOC fluxes, but also the consequences for 5 optical water quality due to the associated terrigenous CDOM. Competing  background; the eastern site ("K") is the exposed site, Kusu, while the western site ("H") is 6 the sheltered site, Hantu. Other stations that were sampled occasionally are shown in small 7 blue dots. typically higher at the more exposed site.