Land use shapes riverine nutrient and sediment concentrations on Moorea, French Polynesia

Land use shapes riverine nutrient and sediment concentrations on Moorea, French Polynesia | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Land use shapes riverine nutrient and sediment concentrations on Moorea, French Polynesia Kyle Neumann, Christian John, Terava Atger, Tauira Punu, Jordan A. Hollarsmith, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6247948/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 31 Jul, 2025 Read the published version in Scientific Reports → Version 1 posted 11 You are reading this latest preprint version Abstract Human activities drive changes in freshwater ecosystems by altering biogeochemical cycles. On high volcanic tropical islands, human activities can be compartmentalized by steep terrain that delineates watershed boundaries. Patterns of human activities, such as land use, affect adjacent stream ecosystems through runoff of sediment and nutrients, which varies seasonally in the tropics as a result of seasonal rainfall. Here, we sought to reveal human impacts on the nutrient and sediment regimes of tropical rivers by tracking patterns of river chemistry across a series of watersheds on Moorea, French Polynesia, between 2018 and 2019. Repeated sampling of rivers across a gradient of human activities revealed that water chemistry varied seasonally and with respect to rainfall and land use. In particular, dissolved inorganic nitrogen was more concentrated in rivers of watersheds with higher rates of land clearing. Additionally, total suspended solids and phosphate were higher when recent rainfall was high. Our results show that human activities can have a substantial impact on the amounts of nutrients and sediment that tropical rivers transport, which on tropical islands could facilitate movement of materials from land to sea as precipitation increases with intensifying climate change. Biological sciences/Ecology/Biogeochemistry Biological sciences/Ecology/Freshwater ecology Earth and environmental sciences/Hydrology Earth and environmental sciences/Environmental sciences/Environmental chemistry Earth and environmental sciences/Environmental sciences/Environmental impact Global change Hydrology Land use Precipitation River chemistry Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction Human activity has altered approximately 75% of the Earth’s surface in the last millennium 1 . Global land-use models indicate a 0.8 million km 2 loss in forest cover since 1960 and a corresponding increase of 0.9-1.0 million km 2 in agricultural land 1 . Hydrologic models estimate that ~ 50% of observed increases in river run-off globally between 1900 and 2000 were the result of changes in land use as deforestation reduced water retention capacity of landscapes. This makes land-use change at least as impactful as climatic change for altering runoff patterns 2 . In addition to the increases in water transport, human driven land use change has resulted in an increase in nutrient enrichment and sediment loading in rivers 3 – 5 . Deforestation in mountainous tropical regions can have an outsized impact on riverine sediment loading and discharge due to their steep slopes, highly erodible soils, and high precipitation 6 , 7 . Many tropical rivers are experiencing increased sediment loading as a result of land clearing for new development and agriculture 8 . The removal of forest vegetation exposes soil to erosion from precipitation and wind while also reducing the soil holding capacity of the cleared area by reducing vegetation root depth and root mass 9 . Elevated sediment concentrations in rivers increase turbidity and reduce light availability and can alter the geomorphology of riverbeds 10 . One study of Kolombangara, a high tropical island in the Solomon Islands, estimated that if no land management practices were put in place, a 10% increase in logging activity would result in an increase in sediment runoff of approximately 1500% island-wide, equating to an annual transport of approximately 3,000 tons of sediment from land through rivers to the ocean 11 . If forests are replaced by agriculture or population centers in these regions, nutrient loading in rivers would likely increase substantially. Conversion of native forest to agriculture utilizing high-nitrogen fertilizer could result in a 25-fold increase in nitrogen export from the system, and urbanization at the same scale could result in a nearly 50-fold increase in nitrogen export as compared to native forest 9 . Understanding the role of land use in nutrient and sediment runoff is essential to understanding alterations in aquatic ecosystems as well as effective planning, management, and restoration of altered watersheds. Land use change has resulted in as much as a 20-fold increase in nitrogen (N) and phosphorus (P) concentrations of many rivers worldwide as compared to pre-industrial levels 9 . Nitrogen and phosphorus enter aquatic ecosystems as a result of atmospheric deposition as well as excess fertilizer application on agricultural plots, industrial and domestic activities such as detergent use, and untreated or incompletely treated human and animal waste 9 , 12 , 13 . As either N or P tend to be the limiting nutrient for primary production in freshwater systems, elevated concentrations of these nutrients can have a number of deleterious effects including: the proliferation of algae and cyanobacteria resulting in increased levels of harmful algal toxins, reduction in dissolved oxygen, and alterations to flora and fauna communities 10 , 14 , 15 . In addition to altering the absolute levels of N and P, anthropogenic nutrient loading in freshwater systems can alter the stoichiometric ratios of dissolved N and P, which can shift trophic interactions and biogeochemical cycling in freshwater ecosystems 15 . Watersheds of high volcanic islands are particularly vulnerable to impacts of land clearing. In tropical areas, these high islands often have high annual rainfall and periodic large storms that can quickly displace fertilizer from agricultural fields. The steep slopes and highly erodible soils on these islands also mean that rain events can result in large amounts of terrigenous sediments running into freshwater and coastal systems 16 – 18 . In addition to the potential ecological impacts of runoff, people living on these islands are often dependent on rivers for clean water for drinking and bathing. Thus, freshwater bodies with increased nutrient and sediment pollution can reduce human health and well-being 19 – 21 through exposure to runoff as well as algal and bacterial blooms. Tropical islands are experiencing rapid growth in their populations, tourism, industry, logging, and large-scale agriculture resulting in the clearing of forest for farms and developments 16 , 22 , 23 . As a result, it is important to understand the impacts of active development on the nutrient and sediment concentrations in rivers on tropical islands. Water quality standards for nutrient and sediment concentrations have been established for some tropical regions e.g. 24 , but have not been established or adopted universally across the Tropics. Thresholds for water quality standards are generally based on impacts of river chemistry for both human health 16 and ecosystem function 25 , so regularly exceeding water quality thresholds poses broader concerns for human and environmental outcomes. Here, we focus on rivers of Moorea, French Polynesia (Fig. 1 ), to examine how nutrient and sediment concentrations relate to human alterations of the landscape. Moorea is a high volcanic island with a rapidly changing landscape due to increased tourism, housing, and industrial development as well as increasing agriculture 26 , 27 . Many people on Moorea have raised issues regarding the state of rivers on the island, citing observations of turbid water, algal and bacterial blooms that result in rashes, and reduced abundance of a freshwater shrimp species that is an important traditional food ( Atger and Punu pers. obs. ). Further, Moorea is home to near-shore coral reef ecosystems, and both resident observations and recent academic work suggests that terrestrially-derived nutrients have a negative impact on near-shore reefs 27 , 28 , which could compromise the ecosystem services (e.g. fisheries and tourism) they provide. However, to date no study has assessed nutrient and sediment loading in rivers on Moorea as it relates to land use. In response to this community and scientific need, we tested the following hypotheses about watersheds in Moorea: 1. Watersheds with higher proportions of cleared land will have higher sediment loading in rivers. 2. Watersheds with higher proportions of cleared land and higher populations will have high concentrations of nitrogen and phosphorus. 3. Water nutrient and sediment concentrations will increase following recent precipitation. 4. Watersheds with the highest percentages of land use and higher populations will exceed nutrient and sediment thresholds established for human and environmental protection in similar systems. To test these hypotheses, we collected water samples from 16 streams on Moorea across rainy and dry seasons in two years. Water samples were analyzed for nitrate (NO3-), nitrite (NO2-), and ammonium (NH4+), the three nitrogen species that comprise dissolved inorganic nitrogen (DIN), as well as orthophosphate (PO43-) and total suspended solids (TSS). We integrated stream chemistry data with precipitation, watershed size, land use, and census data, and used principal components analysis (PCA) and linear modeling to evaluate how local factors influenced the nutrient and sediment dynamics of streams across the island. Finally, we compared nutrient and sediment loading in streams of Moorea to similar systems in other tropical islands using previously-established water quality thresholds to assess potential concerns regarding human health and alteration to ecosystem function. Results Patterns in Rainfall Precipitation in Moorea was seasonal but occurred during every month of the study (Fig. 2 ). Within our sampling dates, February of 2018 had the highest cumulative monthly rainfall with 866 mm of precipitation on the North shore, 659 mm on the East shore, and 772 mm on the West shore. The rainy season in 2019 had lower monthly precipitation than in 2018 with a peak of 234 mm in February on the West shore, and 465 and 274 mm in March on the North shore and East shore, respectively. August was the driest month of both years and precipitation totaled 103 mm on the North shore, 56 mm on the East shore, and 39 mm on the West shore in 2018; and 86 mm of on the North shore, 23 mm on the East shore, and 28 mm on the West shore in 2019. Watershed and Seasonal Differences in Water Chemistry Water chemistry varied considerably by nutrient, season, and watershed across the sampling period (Fig. 3 , Supplementary materials S1). DIN concentrations ranged from a minimum of 0.05 µM measured in a sample collected from Papetoai during the dry season, to 37.7 µM measured in a sample collected in Paopao during the rainy season. DIN differed significantly across watersheds (p < 0.001) and seasons (p = 0.03; Fig. 3 a, Supplementary materials S2). We found no significant differences between seasons within each watershed in the post-hoc Tukey test, though the significant interaction between watershed and season in the overall model indicates that rainy season DIN concentrations were higher than dry season in some watersheds (e.g., Papetoai and Maharepa) while the opposite pattern was observed in others (e.g., Ha’apiti and Haumi). The watershed with the highest mean DIN concentration was Paopao (18.20 µM), and the lowest mean concentration was found in Atiha (3.67 µM). The nitrogen species comprising the majority of DIN was nitrate, followed by ammonium and finally nitrite. Nitrate concentrations differed significantly by watershed (p < 0.001), and there was a significant interaction between watershed and season (p < 0.001); like DIN, the post-hoc Tukey test revealed no significant differences in nitrate concentrations between seasons within watersheds (Fig. 3 b). Ammonium concentrations differed significantly by watershed (p < 0.001), while the interaction with season was not significant. Teavaro and Haumi had higher ammonium levels than the other watersheds (Fig. 3 c; the mean ± s.e. ammonium concentration in these watersheds was 1.64 ± 0.31 µM and 1.29 ± 0.15 µM, respectively, whereas across watersheds the mean ammonium concentration was 0.63 ± 0.12 µM). Nitrite concentrations differed significantly by watershed ( p < 0.001) but not by season although there was a significant interaction of watershed and season ( p = 0.003). The rainy season nitrite concentrations were significantly lower than the dry season concentrations in Haumi (difference ± s.e. = 0.38 ± 0.15 µM, p < 0.001). Like ammonium, Teavaro and Haumi had higher nitrite values relative to the other watersheds (Fig. 3 d; nitrite concentration in these watersheds was 0.76 ± 0.16 µM and 0.73 ± 0.18 µM, respectively, whereas across watersheds the mean ammonium concentration was 0.21 ± 0.06 µM). Phosphate concentrations ranged from a minimum of 0.19 µM from a sample collected in Teavaro in the dry season to a maximum of 7.58 µM in a sample collected from Paopao 3 in the rainy season. Phosphate differed significantly among watersheds ( p < 0.001) and season (p = 0.03) and had a significant interaction of watershed and season ( p = 0.007) (Fig. 3 e). The highest mean phosphate concentration from rivers that flowed year-round was observed in Paopao 3 (2.70 µM) while the lowest was in Paopao 1 (0.98 µM). Phosphate concentrations were the only nutrient species significantly associated with TSS concentrations (β = 0.369, p < 0.001, and r 2 = 0.33). The N:P ratio ranged between a minimum of 0.03 in Papetoai, and a maximum of 54 in Paopao 1. The mean N:P ratio was highest in Paopao 1 (23.6) and lowest in Maharepa (2.25). N:P differed significantly across watersheds ( p < 0.001) and had a significant interaction with watershed and season ( p < 0.001), with Paopao 1 and Opunohu 2 having significantly higher N:P in dry seasons than the rainy seasons (Fig. 3 f). Total suspended solid concentrations ranged from below the detection limit in some samples from Atiha, Ha’apiti, and Maharepa to a maximum of 902 mg/L found in a sample collected in Paopao 1 during the rainy season. Mean TSS was higher during the rainy season than the dry season in 10 of the 12 watersheds for which TSS data were collected in both rainy and dry seasons (i.e. all sites except Pihaena, Paopao, and Paopao 3). Differences in TSS among seasons were significant ( p = 0.002) but the interaction between watershed and season was not significant ( p > 0.05) (Fig. 3 ). Although differences in TSS among watersheds were not significant after accounting for season, variability in TSS concentration across watersheds was impressive: The lowest mean TSS concentration was recorded in Ma’atea at 2.5 mg/L while the highest was recorded in Paopao 1 at 137 mg/L. The next highest mean TSS concentration was in Paopao 2 at 117 mg/L, followed by Papetoai with 46 mg/L, less than half the concentration of the two largest Paopao sub-watersheds. Land-use and population across Moorea Our analysis of land use focused on the percent of cleared land in each focal watershed using the consensus land cover classification from a series of Worldview-3 images collected in 2018 (Fig. 1 ) 29 . Island-wide model accuracy was 98% based on a spatially random sample of validation points. Using a hierarchical sampling scheme to balance across training classes, classification accuracy was 82%. Most of the classification discrepancies came from errors of omission of exposed soil, which was often co-occurring with intensive monoculture and captured by that class, and errors of omission of small buildings which were lost as noise to their surrounding vegetation buffer. In both cases, these discrepancies fell within the broader category of “cleared land” and so were determined to be acceptable. The watershed with the highest proportion of cleared land in 2018 was Paopao 2 (23%), followed by Opunohu 1 (10.4%) while the lowest was Ha’apiti (0.8%) (Table 1 ). The human population on Moorea was 17,463 residents in 2017, with the largest population in the Paopao basin (1281 residents) and the smallest in Pihaena (44 residents) (Table 1 ). Population was not significantly related to watershed size or percentage of the watershed that was classified as cleared land ( p > 0.05 but β > 0 in all cases), but these null associations were driven by the large area and broad-scale land clearing in the Opunohu basin (which has a very low population density): among all watersheds except those in the Opunohu basin, population was positively associated with watershed size and percentage of watershed that was cleared (R 2 = 0.60 and 0.32, and p = 0.002 and 0.04, respectively). Table 1 Watershed parameters including orientation of shore, watershed area, population size, Strahler stream order of the sampled river, stream length from source to sampling location, proportion of the watershed that is clear of forest (including agriculture and development). Sub-watersheds comprising main Opunohu and Paopao basins indicated with italics. Watershed Shore Watershed area (km 2 ) Stream order Stream length (km) Population Cleared (%) Atiha W 4.3 3 5 239 3.4 Ha'apiti W 3.3 3 3.1 372 0.8 Haumi E 2.9 2 3.4 422 2 Ma'atea E 4.8 3 4.9 518 3.5 Maharepa N 3 2 3.5 698 8.8 Opunohu N 15.2 4 5.4 159 9.5 Opunohu 1 N 9.2 4 5.2 149 10.4 Opunohu 2 N 5.6 3 4.2 10 5.7 Paopao N 8.8 4 4.5 1281 15.6 Paopao 1 N 2.5 4 3.9 280 3.5 Paopao 2 N 5 3 3.3 721 22.7 Paopao 3 N 0.4 2 1.9 280 9.2 Papetoai N 5 3 4.5 798 5.3 Pihaena N 1.6 2 2.8 44 7.4 Teavaro E 1.3 2 2 367 1.3 Vaianae W 4.7 3 4.1 233 3 PCA of watershed development and river chemistry Sample points across the watersheds and seasons were distributed across the ordination space in the PCA (Fig. 4 ). The first PC axis explained 47.6% of the variability in river chemistry, was positively associated with PO 4 and TSS (PC1 loadings of 0.55 and 0.29, respectively), and was negatively correlated with N:P and DIN (PC1 loadings of -0.62 and − 0.47, respectively). The second PC axis explained 35.5% of the variability and was positively correlated with all of the watershed chemistry variables (PC2 loadings of 0.41, 0.68, 0.53, and 0.28, for PO43-, TSS, DIN, and N:P, respectively) (Table 2 ). The third and fourth PC axes explained 10.4% and 6.6% of the variability, respectively. Of the environmental variables we considered (watershed identity, recent precipitation, % cleared, watershed area, and population), watershed identity, precipitation and % cleared were significantly related to the first two axes of the ordination space ( p = 0.001, 0.001, and 0.045, respectively): watershed identity most strongly explained the spread of the data in the ordination space (r2 = 0.48), followed by recent precipitation (r 2 = 0.25), and percent of watershed cleared in 2018 (r2 = 0.04). Whereas recent precipitation explained variation in parameter space associated with phosphate and total suspended solids, cleared land was more strongly associated with dissolved inorganic nitrogen. Table 2 PCA axis loadings of phosphate, dissolved inorganic nitrogen, N:P, and total suspended solids concentrations. Parameter PC1 PC2 PC3 PC4 DIN (µM) -0.47 0.53 -0.55 0.44 PO 4 3− (µM) 0.55 0.41 -0.5 -0.52 N:P ratio -0.62 0.28 0.26 -0.68 TSS (mg/L) 0.29 0.68 0.62 0.25 Linear models of mean and maximum nutrient and sediment concentrations Across watersheds, land use and population were important factors affecting mean seasonal total suspended solids in water samples. A linear model including the season, cleared land percentage, and population of each watershed revealed that mean river TSS was highest during the rainy season, and in watersheds with more land clearing and higher populations (F 3,23 = 4.24, R2 = 0.36, p = 0.02). These factors were not significant (p > 0.05) predictors in a watershed-level linear model of dissolved inorganic nitrogen. Using the complete (daily) dataset, a linear mixed effects model including the cleared land percentage, population, and precipitation as fixed effects and site as a random effect showed that both TSS and DIN were positively associated with recent precipitation (p = 0.004 and 0.04, and conditional R 2 = 0.10 and 0.49, respectively). The seasonal role of precipitation in the system led to increasingly high river TSS and DIN concentrations in the most cleared watersheds (Fig. 5 ). Mixed effects models also revealed seasonal dynamics for river phosphate, with higher phosphate concentrations following recent precipitation (p < 0.001, conditional r2 = 0.33). Water quality thresholds Relaxed and stringent water quality thresholds were selected for both sediment and nutrient concentrations based on previous work in the Tropics (5 mg/L and 50 mg/L for TSS; and 0.1 mg/L and 0.18 mg/L for DIN) 16 , 25 , 30 , 31 . Water samples from all watersheds, except for Ma’atea, exceeded the lower TSS water quality threshold of 5 mg/L at least once. On average, watersheds on Moorea exceeded this threshold in 39% of sampling events (Table 3 ). The maximum exceedance percentage was found in Pihaena at 78% of the time. Eight of the watersheds included in this study exceeded the upper TSS threshold of 50 mg/L at least once. Across Moorea, this threshold was exceeded on average 8% of the time. Six watersheds exceeded Hawaiian water quality standards with values of 50 mg/L at least 10% of the time (Haumi: 11%, Opunohu 1: 10%, Paopao 1 and 2: 14%, Papetoai: 12%, and Pihaena: 22%). All watersheds except Atiha and Vaianae exceeded both the lower and the upper DIN thresholds (Table 3 ). On average, stream samples across Moorea exceeded 0.1 mg/L DIN 46% of the time and 0.18 mg/L 20% of the time. Opunohu 2, Paopao 1, and Paopao 2 exceeded the lower standard in every sample. Although the frequency by which rivers exceeded thresholds was not significantly associated with the proportion of the watershed cleared of forest, the slope of the relationship between threshold exceedance frequencies and watershed clearing was almost always positive (i.e. β > 0, p > 0.05, except for the 0.18 mg/L DIN threshold). Table 3 Exceedance percentages by watershed of low and high TSS and DIN water quality thresholds. Bold values represent exceedance frequencies of concern (> 10%) . Watershed > 5 mg/L TSS > 50 mg/L TSS > 0.1 mg/L DIN > 0.18 mg/L DIN Atiha 8 0 0 0 Ha'apiti 50 8 24 0 Haumi 22 11 67 33 Ma'atea 0 0 11 0 Maharepa 42 0 12 6 Opunohu 1 50 10 7 0 Opunohu 2 30 0 100 77 Paopao 1 43 14 100 71 Paopao 2 43 14 100 14 Papetoai 62 12 88 12 Pihaena 78 22 23 13 Teavaro 50 0 73 27 Vaianae 33 8 0 0 Overall 39 8 46 20 Discussion Through a combination of island-wide water sampling and remote sensing analyses, we identified associations between land use at the watershed scale and the nutrient and sediment concentrations of rivers on Moorea. Dissolved inorganic nitrogen (DIN), phosphate, total suspended solids (TSS), and the nitrogen to phosphorus (N:P) ratio differed across rivers and between the rainy and dry seasons. Further, DIN concentrations and TSS were positively related to the percentage of land in a watershed that was cleared of forest and that relationship was mediated by seasonal rainfall. Phosphate concentrations were not related to the metrics of land use change we considered, but were strongly associated with recent rainfall, and are likely the result of weathering of the island’s phosphate rich geology. Importantly, DIN and TSS concentrations regularly exceeded thresholds established in similar systems to identify watersheds that may pose a danger to human health and that of aquatic biota. Nutrient enrichment and its implications In the PCA and linear models, we observed a strong positive relationship between land-clearing and high DIN concentrations that was also influenced by recent precipitation. Similar relationships between DIN and land-clearing combined with human population were found in studies of rivers in American Samoa 30 and Guam 31 . These results indicate that these two watershed parameters – cleared land and population – may be sufficient to make approximate predictions of river DIN concentrations in steep tropical islands. This could be a valuable tool to remotely approximate DIN concentrations in unstudied watersheds on steep-islands, using only satellite-derived data on cleared land and census data on human populations. It is important to note that a more data complex runoff model would be required to calculate flux of DIN from watersheds 32 . However, estimates of DIN concentrations can be valuable in assessing potential human or environmental health concerns 30 , 31 and regulatory thresholds are commonly based on concentrations rather than fluxes 24 . We observed substantially higher DIN concentrations in rivers on Moorea than previous studies on tropical islands. The highest observed DIN sample in our study, 2.09 mg/L (37.72 µM), was an order of magnitude higher than the maximum measured in Guam (0.20 mg/L) 30 , and much higher than the maximum in American Samoa (0.75 mg/L) 31 . Another study in American Samoa recorded a mean of 8.5 µM DIN in a heavily disturbed watershed 22 which is close to the overall mean observed in our study (0.55 mg/L, or 9.05 µM), including data from all watersheds from minimally to highly disturbed. As a result of this difference, DIN concentrations in almost all watersheds on Moorea exceeded the water quality threshold of 0.1 mg/L established by these studies. DIN values in Moorea also frequently exceeded the 0.18 mg/L threshold from the Hawaii water quality standards, despite values being comparable to watersheds in Hawaii with similar land use 33 , 34 . Our PCA analysis and linear mixed models indicate that DIN runoff on Moorea is likely primarily a result of agricultural activity as the percent cleared land classification in our satellite image analysis is largely comprised of agricultural land. One of the primary commercial crops grown on Moorea is pineapples 17 . Pineapple farming requires year-round applications of nitrogen fertilizer and has contributed to nitrogen-loading in rivers across the tropics 35 – 37 . The nitrogen in these fertilizers is usually in the form of urea (CH4N 2 O) which transforms to ammonium and then nitrate via nitrification in the soil due to biological activity 38 . Nitrate was the most abundant form of nitrogen we observed. Nitrate is highly water soluble and thus is easily transported from terrestrial to aquatic environments via surface runoff. Nitrogen-enriched runoff is likely exacerbated on Moorea by the lack of runoff control and the lack of requirements to keep riparian buffers between fields and rivers 39 . Previous work on Moorea measuring nitrogen isotopes in nearshore macroalgae observed elevated δ15N in algae tissue sampled near population centers indicating that sewage leaching directly into the nearshore environment via submarine groundwater discharge was likely a large contributor to nitrogen pollution 28 , 40 , 41 . Direct measurements of DIN concentrations in groundwater and submarine groundwater seeps in these studies recorded DIN levels as high as 45 µM 41 . Such groundwater seeps could be at the root of high DIN concentrations reported in American Samoa, described above. It is likely that groundwater carrying DIN is also entering streams through hyporheic exchange in the lower reaches of the watershed due to the porous nature of the volcanic geology 42 and high head pressure from the island’s short, steep slopes 43 . This exchange may be enhanced in the lower reaches of the rivers as they have been heavily channelized and fortified for flood control ( KN personal observation ) which has enhanced stream bed erosion and deepening of river channels 13 resulting in an increase in head near the terminus where our samples were collected. In both the PCA and linear regression analysis, riverine phosphate on Moorea was decoupled from both land clearing and population, suggesting that agriculture and urbanization were not directly responsible for the differences in phosphate concentrations we observed around the island. Phosphate concentrations ranged from a 0.19 µM sample collected in Teavaro to a 7.58µM sample collected in Paopao. Phosphate concentrations differed significantly among watersheds ( p < 0.001) and season (p = 0.01) and had a significant interaction of watershed and season ( p = 0.007). The range of phosphate concentrations recorded in this study are comparable to those found in a study of watersheds on Oahu, Hawaii, across base and stormflow and including watersheds considered to be forested, agricultural and urban 33 . Anthropogenic sources of phosphate often come in the form of agricultural fertilizers, however pineapples need little to no phosphorus fertilizer where soils have a naturally high phosphorus content as they likely do on Moorea 44 . Soils on the island are derived from erosion of volcanic basalts which tend to be rich in phosphorus, and the availability of phosphorus in the soil of a given area is additionally dependent on numerous factors including age, weathering, and localized precipitation 45 . Phosphate concentrations in rivers are often tied to sediment concentrations as phosphate can desorb from sediment particles in transit 46 ; this pattern was evident in our dataset as well. As there are at least eight different volcanic soils on Moorea 47 , differences in riverine phosphate between watersheds could be related to erosion, vegetative utilization, or leaching of soils with different phosphate levels depending on which soil types happen to be exposed in a given watershed. Groundwater and submarine groundwater discharge on Moorea is also elevated in phosphate relative to surface water. This is likely due to weathering of phosphate rich basaltic rock in aquifers 41 . It is likely that phosphate also enters streams on Moorea via hyporheic exchange. Differences in phosphate in rivers on Moorea may also be the result of differences in bedrock phosphate concentrations, underground mobilization rates, and/or differences in groundwater-surface water hyporheic exchange rates. As anthropogenic sources of DIN increase on the island due to expanded agriculture and urbanization, it is likely that the N:P ratio will also increase as nitrogen enrichment outpaces phosphate mobilization. Drivers and consequences of riverine sediments Both the PCA and regression analyses illustrated that TSS concentrations were significantly associated with the percentage of cleared land in a given watershed and recent precipitation, suggesting that both land use and climatic drivers contribute to the mobilization of sediment. Similar statistical modeling approaches have been used to predict TSS in other mountainous watersheds 48 . These models do not predict discharge or sediment flux, but like our DIN model they could be a useful framework for a first order assessment of potential human health or environmental impact concerns in similar regions. TSS thresholds and regulations are often based on concentrations, not flux 24 . In locations where data are lacking, such as many high islands in the South Pacific, this simple linear modeling approach may provide valuable insights where there is insufficient data to inform a more complicated runoff model like the Soil and Water Assessment Tool (SWAT), Revised Universal Soil Loss Equation (RUSLE), Modified Morgan–Morgan–Finney (MMF) model, or others that are commonly paired with remote sensing analysis to assess relationships between land use and TSS 49 – 51 . TSS concentrations in all studied watersheds on Moorea, except for Ma’atea, exceeded the water quality threshold of 5 mg/L TSS for safe drinking, bathing and cleaning 25 , 35 . TSS values from six watersheds also exceeded the 50 mg/L threshold representing potentially lethal levels for fish species 25 , for an overall exceedance rate of 8%. We also observed high variability in TSS levels between samples, including some outliers (e.g. >900 mg/L compared to mean of 26 mg/L in Paopao) that illustrate how sediment is mobilized in these flashy systems and how difficult it can be to capture differences between watersheds or seasons when much of the sediment that moves from land to river can mobilize during a few major rain events 52 . The range of TSS values observed across watersheds experiencing varying degrees of land clearing and urbanization are within the ranges observed in a study across an urbanization gradient in Hawaii 53 and a study of a “pristine” rainforest watershed in Fiji that includes samples taken during a cyclone 54 . TSS values observed in the most disturbed watersheds of our study (Paopao and Opunohu) were considerably lower than those observed in a study of urban and developing watersheds on Tahiti which recorded values as high as 33.3 g/L TSS in a watershed that was actively undergoing sizeable earthworks projects 55 . Much of the existing land suitable for development in Moorea has been utilized, so new residential or industrial development requires considerable earth moving efforts, including forest clearing and terracing to create flat buildable plots. We did not observe use of soil retention measures (e.g. vegetative cover, mulch, silt fences, geotextiles) that have been used in similar steep tropical systems to reduce erosion 56 , 57 . Furthermore, a majority of the roads in the interior of the island are not paved, tend to follow river channels, and lack any soil retention devices (KN pers. observation) that have been successfully demonstrated in comparable systems 58 . As a result of their demonstrated ability to reduce erosion, implementation of erosion control measures such as those listed above are required by law on construction projects in Hawaii 24 . Land clearing associated with agriculture is also likely a large contributor to TSS runoff as pineapple farming is known to contribute to high erosion rates 59 . In a recent study which scored farming practices in French Polynesia based on soil preservation, pineapple farming scored the lowest for its contributions to soil degradation 17 . Across French Polynesia and in many other tropical climates pineapples are often grown on moderate to steep slopes making it difficult to employ erosion control measures such as cover crop or mulching between rows, which have been shown to be effective in reducing soil erosion in other agricultural systems 17 , 35 . Further, to maximize yield from a farmed plot, many farms in Moorea extend right up to riverbanks with no riparian zone or other buffer to slow the movement of sediment from the fields to the rivers ( pers. observation ). All of these conditions create high erosion potential, and contribute to the strong relationship between land clearing and TSS in rivers we observed in Moorea. Conclusion This study provides the first island-wide assessment of riverine nutrient and sediment concentrations on Moorea, French Polynesia, and explores the relationships between river chemistry, weather, land use, and populations. Watersheds on Moorea, as well as other small high tropical islands worldwide, are developing rapidly with growing populations, tourism, logging and large-scale agriculture 7 , 60 . Our study puts Moorea in context of these other efforts to study land use and its impacts on rivers. The relationships we observed between land use and river nutrient chemistry on Moorea are comparable to those observed in other developed and developing islands around the world 16 , 22 , 23 . In some cases, Moorea may be more susceptible to land-use initiated sediment erosion and runoff as indicated by our comparison with other studies with respect to TSS threshold exceedance. Perhaps most urgently, this study highlights watersheds on Moorea which present risks for human health and/or concerns regarding ecosystem processes based on nutrient and sediment thresholds. Finally, this study also serves as a point of comparison against future changes on Moorea as land clearing and population growth continue on the island. As Moorea, and other high tropical islands, continue to develop, impacts of watershed change are likely to increase with potentially profound consequences for human and ecosystem health. Methods Study Location Moorea is a volcanic island located in the Society Islands Archipelago of French Polynesia (17° 29′ S, 149° 50′ W). As the island eroded and subsided, a barrier reef formed separating the open ocean from a 53 km2 lagoon 61 . The island has a relatively small area of 134 km2, but features diverse topography. Since it was formed by a volcanic eruption approximately 1.6 million years ago 62 , Moorea has collapsed and eroded into numerous distinct watersheds separated by steep, sharp ridgelines with a highest elevation of 1207 m at Mount Tohiea 62 (Fig. 1 ). A majority of the soils (98.7%) on the island formed from weathering and erosion of volcanic parent material with the remainder being coralline in origin 47 . Of the eight soils of volcanic origin on Moorea, five originate from weathered basalt 47 . Moorea has been inhabited by humans since around 200 CE when it was settled by the Polynesians 63 , who brought with them a variety of nonnative plants and animals for cultivation. A 2015 study of vegetation on Moorea classified 32% of the island as urbanized and cultivated lands, 17% as novel habitat dominated by invasive species, 45% as hybrid habitats featuring a mix of introduced and native species, and 6% as native habitat which is primarily located on extremely steep slopes and at high elevation 62 . As of the 2017 census, Moorea was home to 17,357 permanent residents 64 . The largest watersheds on Moorea are two adjacent valleys on the north shore, Opunohu and Paopao (Table 1 ), formed by a collapse of the volcano’s caldera 62 . Opunohu is home to a small population (159 people as of 2017), but hosts a large portion of the island’s commercial agriculture which is principally comprised of pineapple plantations and cattle pasture. Using the Strahler method, the Opunohu River is the only 4th order stream on the island 65 . As an example of the steepness of watersheds on Moorea, more than 90% of the Opunohu valley is comprised of slopes greater than 10% 66 . Neighboring Paopao is home to the largest population center (1281 people as of 2017) and small mixed agricultural fields, and also features a 4th order stream. Its predominant sub-watersheds feature 2nd -4th order streams and span a range of areas. The smallest standalone watershed included in this study, Teavaro, is located on the eastern shore of the island and is only 1.3 km2, and home to 367 people. Teavaro and the rest of the rivers in this study are second and third order streams. The variety of watersheds in close proximity featuring a range of areas, land uses and populations makes Moorea a compelling location to investigate interactions between human activity and riverine nutrients and sediments. In order to study rivers and land use on Moorea, we first engaged the local community to receive feedback regarding key questions that would be of interest and value to the people of Moorea. The goals, objectives, and research locations for this study were shaped by meetings with community members organized by the Te Pu Atiti’a Polynesian Cultural Center located directly adjacent to the University of California Gump Research Station, where our research was based. These conversations precipitated a new community science organization on Moorea called Ati Vai (‘Water Clan’ in Tahitian). Members of Ati Vai contributed to the study design, provided local knowledge and site context, and aided with field data collection throughout this study. The combination of modern scientific tools and techniques with traditional and local knowledge of the island and its rivers was paramount to the success of this research. Focal watershed selection In the first year, eight watersheds were chosen to represent a range of watershed sizes, land uses and population densities, and are located on all three shores of the island. These watersheds were: Atiha, Ha’apiti, Maharepa, Opunohu, Paopao, Pihaena, Teavaro, and Vaiane. The Opunohu watershed was evaluated at the level of its two primary sub-watersheds, which have vastly different human populations (Fig. 1 , Table 1 ). Three additional watersheds - Haumi, Ma’atea, and Papetoai – as well as the primary sub-watersheds of the Paopao watershed (defined by 2nd − 4th order streams), were added in the second year of sampling to increase sampling across the ranges of watershed size and human impacts (Fig. 1 , Table 1 ). All of the rivers included in this study are perennial, with the exception of Pihaena which has low or no flow during the dry season. Water sample collection We collected water samples during two rainy and two dry seasons between January 2018 and September 2019 (rainy: January - March 2018 and February - March 2019, dry: August - September 2018 and 2019). During the study periods, water samples were collected as close to the mouths of the rivers as possible, but above brackish mixing zones. Where natural or manmade impediments to access existed, we made every effort to find a suitable sampling location as close to the river mouth as possible. During the above periods, samples were collected at least once a week with attention paid to capturing samples across river stages from low-flow to storm-flow. We collected storm-flow samples during or shortly after rain events to ensure that we captured the nutrient and sediment regimes during times of high flow. River samples were collected at approximately 60% water depth (e.g. at 40 cm above the bottom in a 1 m deep river) at the center of flow. Samples were collected in acid washed 1-liter HDPE Cubitainers®, transported on ice to the lab at Gump Station, and kept refrigerated until processed. Samples were processed upon return to the lab within no more than 24 hours. Each sample was divided and filtered for nutrients or total suspended solids using a multi-channel peristaltic filtration system. Nutrient analyses Nutrient sub-samples were filtered through 47 mm 0.15 µm PES filter discs and then divided for separate analytical methods. NH4 + analyses were conducted immediately after filtration using the OPA method 67 on a Turner Trilogy fluorometer fitted with an ammonium detection module (Minimum Detection Limit (MDL): 0.05 µM). The remaining samples were frozen at -20°C and transported to the Centre de Recherches Insulaires et Observatoire de l'Environnement (CRIOBE) on Moorea where they were analyzed for PO43- (MDL: 3 nM), NO3- (MDL: 2 nM), and NO2- (MDL: 2 nM) on an AA3 Auto-analyzer (SEAL Analytical, e.g. [ 68 ]). Total DIN per sample represents the sum of NO2-, NO3-, and NH4+. The molar N:P ratio was calculated for each sample by dividing DIN by phosphate. Total suspended solids (TSS) analysis TSS samples were filtered using pre-dried and pre-weighed 47 mm GF/F (0.70 µm) filter discs. Filters were weighed on a Mettler-Toledo XS104 balance with 0.1 mg precision. Samples were filtered to 200 mL, or until the filters were completely clogged preventing more water to pass through, whichever came first. If filters clogged before filtering 200mL, we recorded the total volume filtered. After filtration, the GF/F filters were placed in a drying oven at 95°C for a minimum of 24 hours at which time they were re-weighed. TSS was calculated using Eq. 1: \(\:TSS=\:\frac{\left(post-filtering\:and\:drying\:weight\right)-(pre-filtering\:weight)}{volume\:filtered}\) [1] Land cover classification The Worldview-3 (WV3) Satellite (Maxar Technologies) collects repeat imagery with the 8-band WV110 sensor every 4.5 days with a ground sampling distance of 1.38m. We used a series of WV3 images collected during four days in 2018 that had little-to-no cloud cover. The imagery was aligned, georeferenced, orthorectified, and mosaicked using the Ortho Mapping workspace in ArcGIS Pro v3.1.2. Training data for land cover classification were generated using data from the Direction des Affaires Foncières de la Polynésie Française (the Directorate of Land Affairs, French Polynesia), and included the classes Water, Forest, Monoculture (intensive agriculture), Buffer/Agriculture (permaculture, orchards, and other vegetated, non-forested land), Buildings, Paved, Dirt and Sand. A U-Net pixel classifier was trained on a ResNet-50 architecture with 10% of training samples withheld for validation for each of the unique dates of imagery collection, yielding an accuracy of 0.86 ± 0.04 (mean ± std. dev. across date-wise models). Individual layers were then compiled into a “consensus” land cover map that used the modal classification for each pixel (Fig. 1 , [***DOI pending]). This approach reduced noise related to solar and topographic variability in pixel classification outputs from individual date layers, and overcame gaps in the dataset introduced by cloud cover. The consensus land cover map had a 98% island-wide classification accuracy. Watersheds were delineated using the Hydrology tool in the Spatial analyst toolbox in ArcGIS Pro v3.1.2. Using a digital elevation model (DEM) from the Shuttle Radar Topography Mission Version 3 (30 m resolution) 69 , we identified the watersheds that feed into our river sampling locations using the Flow Accumulation and Flow Direction Tools in ArcGIS. Points along the rivers were identified towards the mouth of rivers of interest using the Snap Pour Point tool. Watersheds of interest for the project were then delineated using the Watershed tool. Precipitation and Census Data Daily precipitation data for Moorea for 2018 and 2019 were obtained for 5 meterological stations on the island through a license agreement with Meteo France. Census data for 2017 was obtained via a license agreement between the Gump Research Station and the Institut de Statistiques de la Polynésie Française. The census data were originally collected in 102 districts on the island, often with multiple districts located inside of a watershed. In order to determine the population for each watershed, we aggregated census data based on spatial intersections between census districts and focal watersheds as described above. TSS and DIN Concentration Thresholds Low (more relaxed) and high (more stringent) water quality thresholds for both DIN and TSS concentrations were selected from a review of literature from similar systems including Hawaii, Guam, the Solomon Islands and American Samoa. The low TSS threshold (5 mg/L) represents concerns for human consumption and bathing 16 , 25 . The high TSS threshold (50 mg/L) is based on water quality standards established in Hawaii 24 and is based on the levels at which exposure to suspended sediment become lethal to a majority of fish species 25 . According to the Hawaiian standards, TSS in streams should not exceed 50 mg/L for more than 10% of the time in the rainy season. The lower DIN concentration threshold of 0.1 mg/L was selected based on studies of runoff impacts to nearshore reefs in American Samoa 30 and Guam 31 . An exceedance of this threshold in streams for more than 20% of the time was associated with diminished diversity and size of coral communities in both regions. The higher threshold of 0.18 mg/L DIN was based on Hawaiian water quality standards which state that streams should not exceed this threshold more than 10% of the time in the rainy season 24 . We calculated the percentage of time that samples collected from each watershed on Moorea exceeded these thresholds using Eq. 2: \(\:\%\:exceedance=\:\frac{\#\:samples\:from\:watershed\:that\:exceed\:threshold}{total\:\#\:samples\:from\:watershed}*100\) [2] Statistical Analyses We used linear models to analyze differences in river nutrients and TSS by watershed, season, and their interaction. When a significant watershed by season interaction existed, or to investigate differences between sites, we used post-hoc Tukey tests with adjusted alpha (p-value) by the number of tests. Data were 1 + log transformed prior to analysis to improve model fit, based on inspection of q-q plots and residuals using the ‘qqplot’, ’qqnorm’, and ‘resid’ functions in R. Only watersheds that had samples from the wet and dry seasons were included in the linear models, so Pihaena was excluded because it only flowed in the wet season. To assess the significance of watershed variables (area, population, and percent cleared land) in explaining river chemistry in the wet and dry season, we used principal components analyses (PCA). PCA is a form of ordination analysis based on Euclidean distance that is well suited to complex datasets with covarying variables. Differences across watersheds and seasons were assessed based on a correlation matrix of dissolved inorganic nitrogen, phosphate, N:P, and total suspended solids levels at a given sampling event. All data were standardized (mean = 0, SD = 1) prior to analysis. River chemistry data were 1 + log transformed before analysis. We used the ‘envfit’ function in the R package ‘vegan’ to test whether watershed characteristics, namely area, population, percent cleared land in 2018, watershed, and season could explain the variation in river chemistry 70 . Envfit finds vectors or factor averages of environmental variables, analogous to fitting a linear model, in ordination space. To further explore relationships between watershed variables and river conditions, we performed a series of linear models testing whether cleared land percentage and population predict nutrient and sediment concentrations. Models were fit using the mean concentrations of nutrients and sediment for each watershed across all seasons and years. Additional models were fit using maximum observed nutrient and sediment values to elucidate relationships between land use and high flow runoff events. We also analyzed the relationship between population from the census data and the area of cleared land from the land cover classification using linear models. Finally, linear mixed-effects models were used to detect impacts of precipitation events on river chemistry by including watershed as a random intercept and including a continuous measure of recent precipitation as an independent variable. All analyses were conducted in R 71 . Declarations Funding for this research was provided by the NSF GRFP to KN, NSF Career grant OCE—1547952 to DEB, NSF grant OCE-1637396 to the Moorea Coral Reef LTER, and the Zegar Family Foundation, The Schmidt Family Foundation, and The Worster Summer Research Fellowship. Funding for education and outreach related to this work was provided by the ASLO Global Outreach Initiative. Author Contribution Conceptualization: K.N. with input from T.A., T.P., and D.E.B. Data curation: K.N. and C.J.Formal analysis: K.N., C.J., and J.A.H.Funding acquisition: K.N. and D.E.B.Investigation: K.N.Methodology: K.N., T.A., T.P., and D.E.B. Project administration: K.N. and D.E.B.Visualization: K.N., C.J., and J.A.H.Supervision: D.E.B.Writing – original draft: K.N.Writing – review & editing: All authors Acknowledgement We thank the community of Moorea for hosting this research on Tahitian land and waters. Credit to Benoit Espiau at CRIOBE for his meticulous nutrient analyses. Particular mention to Corinne Fuchs, Kyla Pierce, Maya Gorgas and the members of ‘Ati Vai for their assistance collecting, processing and analyzing samples. Thank you to the staff of the University of California Richard B. Gump Research Station for all of the work solving logistical, administrative and mechanical issues. We could not have done this work without you. With respect to the spelling of Moorea, we followed the Raapoto transcription system, but also recognize other community members follow the Te Fare Vanā’a transcription system where the island name is spelled with an ’eta (Mo’orea).Funding for this research was provided by the NSF GRFP to KN, NSF Career grant OCE—1547952 to DEB, NSF grant OCE-1637396 to the Moorea Coral Reef LTER, and the Zegar Family Foundation, The Schmidt Family Foundation, and The Worster Summer Research Fellowship. Funding for education and outreach related to this work was provided by the ASLO Global Outreach Initiative. Data Availability Data from this study are stored on EDI (***DOI available upon manuscript acceptance), and code to reproduce analyses are on github (***Link available upon manuscript acceptance). Please contact the corresponding author for data requests ( [email protected] ). Additional information The authors declare no competing interests. References Winkler, K., Fuchs, R., Rounsevell, M. & Herold, M. Global land use changes are four times greater than previously estimated. Nat. Commun. 12 , 2501 (2021). Piao, S. et al. Changes in climate and land use have a larger direct impact than rising CO2 on global river runoff trends. Proceedings of the National Academy of Sciences 104, 15242–15247 (2007). Green, P. A. et al. Pre-industrial and contemporary fluxes of nitrogen through rivers: a global assessment based on typology. Biogeochemistry 68 , 71–105 (2004). Kroon, F. J. et al. River loads of suspended solids, nitrogen, phosphorus and herbicides delivered to the Great Barrier Reef lagoon. Mar. Pollut. Bull. 65 , 167–181 (2012). Hoffmann, T. et al. Human impact on fluvial regimes and sediment flux during the Holocene: Review and future research agenda. Glob. Planet Change . 72 , 87–98 (2010). Labrière, N., Locatelli, B., Laumonier, Y., Freycon, V. & Bernoux, M. Soil erosion in the humid tropics: A systematic quantitative review. Agriculture Ecosystems Environment . 203 , 127–139 (2015). Tanaka, Y., Minggat, E. & Roseli, W. The impact of tropical land-use change on downstream riverine and estuarine water properties and biogeochemical cycles: a review. Ecol. Processes . 10 , 40 (2021). Bartley, R. et al. Relating sediment impacts on coral reefs to watershed sources, processes and management: A review. Sci. Total Environ. 468–469 , 1138–1153 (2014). Downing, J. A. et al. The impact of accelerating land-use change on the N-Cycle of tropical aquatic ecosystems: Current conditions and projected changes. Biogeochemistry 46 , 109–148 (1999). Allan, J. D. Landscapes and Riverscapes: The Influence of Land Use on Stream Ecosystems. Annu. Rev. Ecol. Evol. Syst. 35 , 257–284 (2004). Wenger, A. S. et al. Best-practice forestry management delivers diminishing returns for coral reefs with increased land-clearing. J. Appl. Ecol. 57 , 2381–2392 (2020). Correll, D. L. The Role of Phosphorus in the Eutrophication of Receiving Waters: A Review. J. Environ. Qual. 27 , 261–266 (1998). Riverine Ecosystem Management: Science for Governing Towards a Sustainable Future . (Springer International Publishing, Cham, (2018). 10.1007/978-3-319-73250-3 Canning, A. D. & Death, R. G. The influence of nutrient enrichment on riverine food web function and stability. Ecol. Evol. 11 , 942–954 (2021). Wurtsbaugh, W. A., Paerl, H. W. & Dodds, W. K. Nutrients, eutrophication and harmful algal blooms along the freshwater to marine continuum. WIREs Water . 6 , e1373 (2019). Albert, S. et al. Water quality challenges associated with industrial logging of a karst landscape: Guadalcanal, Solomon Islands. Mar. Pollut. Bull. 169 , 112506 (2021). Surchat, M. et al. Soil and Pest Management in French Polynesian Farming Systems and Drivers and Barriers for Implementation of Practices Based on Agroecological Principles. Frontiers Sustainable Food Systems 5 , (2021). Koshiba, S. et al. Palau’s taro fields and mangroves protect the coral reefs by trapping eroded fine sediment. Wetlands Ecol. Manage. 21 , 157–164 (2013). Guernier, V. et al. Leptospira diversity in animals and humans in Tahiti, French Polynesia. PLoS Negl. Trop. Dis. 11 , e0005676 (2017). Viau, E. J. & Boehm, A. B. Quantitative PCR-based detection of pathogenic Leptospira in Hawai’ian coastal streams. J. Water Health . 9 , 637–646 (2011). Strauch, A. M. Relative change in stream discharge from a tropical watershed improves predictions of fecal bacteria in near-shore environments. Hydrol. Sci. J. 62 , 1381–1393 (2017). Comeros-Raynal, M. T. et al. Catchment to sea connection: Impacts of terrestrial run-off on benthic ecosystems in American Samoa. Mar. Pollut. Bull. 169 , 112530 (2021). Dadhich, A. P. & Nadaoka, K. Analysis of Terrestrial Discharge from Agricultural Watersheds and Its Impact on Nearshore and Offshore Reefs in Fiji. coas 28, 1225–1235 (2012). Hawaii Department of Health. Amendment and Compilation of Chap. 11–54 Hawaii Administrative Rules. (2021). Wenger, A. S. et al. Predicting the impact of logging activities on soil erosion and water quality in steep, forested tropical islands. Environ. Res. Lett. 13 , 044035 (2018). Boutillier, S. & Duane, T. Land Use Planning to Promote Marine Conservation of Coral Reef Ecosystems in Moorea, French Polynesia . 28 (2006). https://escholarship.org/content/qt10f3q5p4/qt10f3q5p4.pdf Loiseau, C., Thiault, L., Devillers, R. & Claudet, J. Cumulative impact assessments highlight the benefits of integrating land-based management with marine spatial planning. Sci. Total Environ. 787 , 147339 (2021). Adam, T. C. et al. Landscape-scale patterns of nutrient enrichment in a coral reef ecosystem: implications for coral to algae phase shifts. Ecol. Appl. 31 , e2227 (2021). Moorea Coral Reef, L. T. E. R., Adam, C., Neumann, K., Maritorena, S. & Burkepile, D. E. MCR LTER: Coral Reef: A 2018 land cover map of Moorea, French Polynesia. Environmental Data Initiative https://doi.org/DOI PENDING (2024). Houk, P. et al. Nutrient thresholds to protect water quality and coral reefs. Mar. Pollut. Bull. 159 , 111451 (2020). Houk, P. et al. Nutrient thresholds to protect water quality, coral reefs, and nearshore fisheries. Mar. Pollut. Bull. 184 , 114144 (2022). Aguilera, R. & Melack, J. M. Relationships Among Nutrient and Sediment Fluxes, Hydrological Variability, Fire, and Land Cover in Coastal California Catchments. J. Geophys. Research: Biogeosciences . 123 , 2568–2589 (2018). Hoover, D. J. & Mackenzie, F. T. Fluvial Fluxes of Water, Suspended Particulate Matter, and Nutrients and Potential Impacts on Tropical Coastal Water Biogeochemistry: Oahu, Hawai‘i. Aquat. Geochem. 15 , 547–570 (2009). Hoover, D. Fluvial nitrogen and phosphorus in Hawaii: Storm runoff, land use, and impacts on coastal waters - ProQuest. (University of Hawai’i at Manoa, (2002). Ryder, M. & Fares, A. Evaluating Cover Crops (Sudex, Sunn Hemp, Oats) for Use as Vegetative Filters to Control Sediment and Nutrient Loading From Agricultural Runoff in a Hawaiian Watershed. JAWRA J. Am. Water Resour. Association . 44 , 640–653 (2008). Guinto, D. F. & Inciong, M. M. Soil quality, management practices and sustainability of pineapple farms in Cavite, Philippines: Part 1. Soil quality. J. South. Pac. Agric. 16 , 30–41 (2012). Millbank, H. & Nothard, B. Understanding the Economics of Horticultural Management Practices and Systems for Improving Water Quality Runoff in the Great Barrier Reef Catchment Areas . 62 (2023). https://era.dpi.qld.gov.au/id/eprint/9271/ Norton, J. & Ouyang, Y. Controls and Adaptive Management of Nitrification in Agricultural Soils. Front Microbiol 10 , (2019). Feld, C. K. et al. Chapter Three - From Natural to Degraded Rivers and Back Again: A Test of Restoration Ecology Theory and Practice. Adv. Ecol. Res. 44 , 119–209 (2011). Knee, K. L., Crook, E. D., Hench, J. L., Leichter, J. J. & Paytan, A. Assessment of Submarine Groundwater Discharge (SGD) as a Source of Dissolved Radium and Nutrients to Moorea (French Polynesia) Coastal Waters. Estuaries Coasts . 39 , 1651–1668 (2016). Haßler, K. et al. Provenance of nutrients in submarine fresh groundwater discharge on Tahiti and Moorea, French Polynesia. Appl. Geochem. 100 , 181–189 (2019). Boano, F. et al. Hyporheic flow and transport processes: Mechanisms, models, and biogeochemical implications. Rev. Geophys. 52 , 603–679 (2014). Shuler, C. K. et al. Understanding surface water–groundwater interaction, submarine groundwater discharge, and associated nutrient loading in a small tropical island watershed. J. Hydrol. 585 , 124342 (2020). Bartholomew, D. P., Rohrbach, K. G. & Evans, D. O. Pineapple Cultivation in Hawaii . 1–8 (2002). http://hdl.handle.net/10125/2444 Rechberger, M. V., Zehetner, F. & Gerzabek, M. H. Phosphate sorption-desorption properties in volcanic topsoils along a chronosequence and a climatic gradient on the Galápagos Islands. J. Plant Nutr. Soil Sci. 184 , 479–491 (2021). de Campos, M., Antonangelo, J. A., van der Zee, S. E., A. T., M. & Alleoni, L. R. F. Degree of phosphate saturation in highly weathered tropical soils. Agric. Water Manage. 206 , 135–146 (2018). Jamet, R. Les sols de Moorea et des îles Sous-le-Vent [cartographic material]: archipel de la Société, Polynésie Française / par Rémi Jamet. Les sols de Moorea et des îles Sous-… Items | National Library of New Zealand | National Library of New Zealand https://natlib.govt.nz/records/20850230 (2000). Ahearn, D. S. et al. Land use and land cover influence on water quality in the last free-flowing river draining the western Sierra Nevada, California. J. Hydrol. 313 , 234–247 (2005). Arnold, J. G. et al. Model Use, Calibration, and Validation. Trans. ASABE . 55 , 1491–1508 (2012). Renard, K. G., Foster, G. R., Yoder, D. C. & McCool, D. K. RUSLE revisited: Status, questions, answers, and the future. J. Soil Water Conserv. 49 , 213–220 (1994). Das, S., Deb, P., Bora, P. K. & Katre, P. Comparison of RUSLE and MMF Soil Loss Models and Evaluation of Catchment Scale Best Management Practices for a Mountainous Watershed in India. Sustainability 13 , 232 (2021). Wymore, A. S., Leon, M. C., Shanley, J. B. & McDowell, W. H. Hysteretic Response of Solutes and Turbidity at the Event Scale Across Forested Tropical Montane Watersheds. Front Earth Sci 7 , (2019). Augustijn, D., Fares, A. & Tran, D. N. Temporal and Spatial Variations in Total Suspended and Dis- solved Solids in the Upper Part of Manoa Stream, Hawaii. Journal Sustainable Watershed Science Management . 1 , 1–9 (2011). Ram, A. R. & Terry, J. P. Stream turbidity responses to storm events in a pristine rainforest watershed on the Coral Coast of southern Fiji. Int. J. Sedim. Res. 31 , 279–290 (2016). Wotling, G. & Bouvier, C. Impact of urbanization on suspended sediment and organic matter fluxes from small catchments in Tahiti. Hydrol. Process. 16 , 1745–1756 (2002). Sutherland, R. A. Rolled erosion control systems for hillslope surface protection: a critical review, synthesis and analysis of available data. I. Background and formative years. Land Degradation Development . 9 , 465–486 (1998). Sutherland, R. A. & Ziegler, A. D. Hillslope runoff and erosion as affected by rolled erosion control systems: a field study. Hydrol. Process. 20 , 2839–2855 (2006). Wemple, B. C. et al. Ecohydrological disturbances associated with roads: Current knowledge, research needs, and management concerns with reference to the tropics. Ecohydrology 11 , e1881 (2018). Martinez, C., Menjívar, J. C. & Saavedra, R. Soils erosion in pineapple (Ananas comosus L. Merr) producing areas. Revista de Ciencias Agrícolas . 39 , 142–154 (2022). Carlson, R. R., Foo, S. A. & Asner, G. P. Land Use Impacts on Coral Reef Health: A Ridge-to-Reef Perspective. Frontiers Mar. Science 6 , (2019). Collin, A. et al. High resolution topobathymetry using a Pleiades-1 triplet: Moorea Island in 3D. Remote Sens. Environ. 208 , 109–119 (2018). Meyer, J. Y., Pouteau, R., Spotswood, E., Taputuarai, R. & Fourdrigniez, M. The importance of novel and hybrid habitats for plant conservation on islands: a case study from Moorea (South Pacific). Biodivers. Conserv. 24 , 83–101 (2015). Kirch, P. V. On the Road of the Winds: An Archaeological History of the Pacific Islands before European Contact, Revised and Expanded Edition (University of California Press, 2017). 10.1525/9780520968899 Institut de la. statistique de la Polynésie Française. https://www.ispf.pf/ Strahler, A. Hypsometric (Area-Altitude) Analysis of Erosional Topography. GSA Bull. 63 , 1117–1142 (1952). Binet, T. & Gonnot, C. Etude de l’érosion sur des plantations d’ananas à Moorea. (2005). Taylor, B. W. et al. Improving the fluorometric ammonium method: matrix effects, background fluorescence, and standard additions. J. North. Am. Benthological Soc. 26 , 167–177 (2007). Rodier, M. et al. Diversity and toxic potential of algal bloom-forming species from Takaroa lagoon (Tuamotu, French Polynesia): a field and mesocosm study. Aquat. Microb. Ecol. 83 , 15–34 (2019). Farr, T. G. et al. The Shuttle Radar Topography Mission. Reviews Geophysics 45 , (2007). Oksanen, J. et al. vegan: Community Ecology Package. (2025). R Core Team. R: A Language and Environment for Statistical Computing. (2022). Additional Declarations No competing interests reported. Supplementary Files NeumannetalSupplementv7.docx Cite Share Download PDF Status: Published Journal Publication published 31 Jul, 2025 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 16 Jun, 2025 Reviews received at journal 13 Jun, 2025 Reviews received at journal 10 Jun, 2025 Reviewers agreed at journal 26 May, 2025 Reviewers agreed at journal 26 May, 2025 Reviewers agreed at journal 26 May, 2025 Reviewers invited by journal 24 May, 2025 Editor assigned by journal 24 May, 2025 Editor invited by journal 24 Apr, 2025 Submission checks completed at journal 05 Apr, 2025 First submitted to journal 05 Apr, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6247948","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":438884006,"identity":"9897e340-6d06-46c2-9b61-79d931350ae7","order_by":0,"name":"Kyle Neumann","email":"","orcid":"","institution":"University of California, Santa Barbara","correspondingAuthor":false,"prefix":"","firstName":"Kyle","middleName":"","lastName":"Neumann","suffix":""},{"id":438884007,"identity":"86c4ed58-e85c-43d4-855e-995cc0a06636","order_by":1,"name":"Christian John","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAArUlEQVRIiWNgGAWjYFCCA2wMDBVAWgKIeYjXcoY0LQxsDIxtpGjhbzx87HHhvMPy8rMbGB+8bSNCi8SBY+nGM7cdNtxw5wCz4VxitBgwnDGT5t12OMFAIoFNmpd4LXMOJ8jPSGD/TYKWhsMJDDcS2JiJ0gL0S5o0z7F0oF8ONkvOOUeEFv4Zh49J89RYA0Os+eCHN2VEaAFaA2MxNhCjHmQNsQpHwSgYBaNg5AIACUU0yUh6rIYAAAAASUVORK5CYII=","orcid":"","institution":"University of California, Santa Barbara","correspondingAuthor":true,"prefix":"","firstName":"Christian","middleName":"","lastName":"John","suffix":""},{"id":438884008,"identity":"3aca6b7e-af2a-483d-93dd-cfede036c109","order_by":2,"name":"Terava Atger","email":"","orcid":"","institution":"University of California Gump Research Station","correspondingAuthor":false,"prefix":"","firstName":"Terava","middleName":"","lastName":"Atger","suffix":""},{"id":438884009,"identity":"fa19e104-2bca-4ebd-b8e8-ea1af7cff288","order_by":3,"name":"Tauira Punu","email":"","orcid":"","institution":"University of California Gump Research Station","correspondingAuthor":false,"prefix":"","firstName":"Tauira","middleName":"","lastName":"Punu","suffix":""},{"id":438884010,"identity":"2baa83a8-f267-4ffd-8868-2d47c3f6a68a","order_by":4,"name":"Jordan A. Hollarsmith","email":"","orcid":"","institution":"NOAA","correspondingAuthor":false,"prefix":"","firstName":"Jordan","middleName":"A.","lastName":"Hollarsmith","suffix":""},{"id":438884011,"identity":"0f1144e3-1178-4a43-838c-0e05bfe7273a","order_by":5,"name":"Deron E. Burkepile","email":"","orcid":"","institution":"University of California, Santa Barbara","correspondingAuthor":false,"prefix":"","firstName":"Deron","middleName":"E.","lastName":"Burkepile","suffix":""}],"badges":[],"createdAt":"2025-03-17 23:08:09","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6247948/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6247948/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-025-13425-1","type":"published","date":"2025-07-31T16:13:20+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":80992283,"identity":"a9dd7ec6-20bd-40d0-b4d4-2b5a25a86094","added_by":"auto","created_at":"2025-04-21 03:40:21","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":856738,"visible":true,"origin":"","legend":"\u003cp\u003eMoorea, French Polynesia with stream channels and focal watersheds superimposed on hillshaded land cover (forested areas shown in dark green, water in blue, intensive agriculture in orange, and other cleared land in pale green; see methods for classification details).\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6247948/v1/fcb484afe89d011ff444f3e5.jpeg"},{"id":80992282,"identity":"2d1cb50b-4045-4254-8ef1-877900c74c52","added_by":"auto","created_at":"2025-04-21 03:40:21","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":124321,"visible":true,"origin":"","legend":"\u003cp\u003eMonthly rainfall by shore orientation (2018-2019). Rainy season sampling was in January – March 2018 and February – March 2019 (blue background). Dry season sampling occurred during August – September in both 2018 and 2019 (beige background). No rainfall data were reported at the east shore meteorological station (Afareaitu 2) during January and February, 2019.\u003c/p\u003e","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6247948/v1/0ee4779870379e4bf369c196.jpeg"},{"id":80992295,"identity":"78889dab-f2ac-4f34-b3a3-b9b79713c12f","added_by":"auto","created_at":"2025-04-21 03:40:21","extension":"jpeg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":276542,"visible":true,"origin":"","legend":"\u003cp\u003eNitrogen concentrations as dissolved inorganic nitrogen (DIN, a), nitrate (NO\u003csub\u003e3\u003c/sub\u003e, b), ammonium (NH\u003csub\u003e4\u003c/sub\u003e, c), Nitrite (NO\u003csub\u003e2\u003c/sub\u003e, d), Phosphate (PO\u003csub\u003e4\u003c/sub\u003e, e), Nitrogen:Phosphorus ratio (N:P, f), and total suspended solids (TSS, g) across sites and seasons (dry season indicated by an open box and rainy season as a striped box). Boxes indicate the first and third quartiles; the vertical bar shows the median; and the vertical lines extend to the upper and lower 1.5 times the interquartile range. Outliers are shown as dots.\u0026nbsp;\u003c/p\u003e","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6247948/v1/0ac15b0807b8fa854a904745.jpeg"},{"id":80992774,"identity":"bf999457-51d6-45d3-87d6-417f625ee30e","added_by":"auto","created_at":"2025-04-21 03:48:21","extension":"jpeg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":201560,"visible":true,"origin":"","legend":"\u003cp\u003ePrincipal components analysis (PCA) results of stream chemistry (DIN, PO4, TSS, and N:P) across watersheds. Stream chemistry loadings are shown in grey. On the first PC axis, DIN and N:P loaded negatively, while TSS and PO\u003csub\u003e4\u003c/sub\u003e loaded positively. On the second PC axis, all four elements loaded positively. Each point represents one sampling event. Black arrows show correlations between stream chemistry and environmental variables; the length of each arrow represents the relative strength of the correlation, and significant associations between environmental variables and principal component space are indicated in the labels (for each symbol p is less than its associated value: . 0.1, * 0.05, ** 0.01, *** 0.001).\u003c/p\u003e","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6247948/v1/c07f62875612533763839fd9.jpeg"},{"id":80992286,"identity":"4e3c2291-80e6-4d50-8428-8ad2d06d3b12","added_by":"auto","created_at":"2025-04-21 03:40:21","extension":"jpeg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":88123,"visible":true,"origin":"","legend":"\u003cp\u003eRiver chemistry in Moorea associated with patterns of land use and rainfall seasonality. Dissolved Inorganic Nitrogen (DIN; a) and Total Suspended Solids (TSS; b) increase more at higher levels of watershed land clearing during the rainy season (blue) compared to the dry season (tan).\u003c/p\u003e","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6247948/v1/5fccc475fdd8c282f230c8f7.jpeg"},{"id":88268282,"identity":"208764bc-78c7-4e08-81f0-1cfbe831b885","added_by":"auto","created_at":"2025-08-04 16:50:41","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2768844,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6247948/v1/7e9005ce-0672-4cb4-a24a-2f7f58765a46.pdf"},{"id":80992773,"identity":"362ed913-2636-4745-9ac6-0efa2d1d8a41","added_by":"auto","created_at":"2025-04-21 03:48:21","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":246935,"visible":true,"origin":"","legend":"","description":"","filename":"NeumannetalSupplementv7.docx","url":"https://assets-eu.researchsquare.com/files/rs-6247948/v1/40b89483edf97e2160005d63.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Land use shapes riverine nutrient and sediment concentrations on Moorea, French Polynesia","fulltext":[{"header":"Introduction","content":"\u003cp\u003eHuman activity has altered approximately 75% of the Earth\u0026rsquo;s surface in the last millennium\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e. Global land-use models indicate a 0.8\u0026nbsp;million km\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e loss in forest cover since 1960 and a corresponding increase of 0.9-1.0\u0026nbsp;million km\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e in agricultural land\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e. Hydrologic models estimate that ~\u0026thinsp;50% of observed increases in river run-off globally between 1900 and 2000 were the result of changes in land use as deforestation reduced water retention capacity of landscapes. This makes land-use change at least as impactful as climatic change for altering runoff patterns\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. In addition to the increases in water transport, human driven land use change has resulted in an increase in nutrient enrichment and sediment loading in rivers\u003csup\u003e\u003cspan additionalcitationids=\"CR4\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eDeforestation in mountainous tropical regions can have an outsized impact on riverine sediment loading and discharge due to their steep slopes, highly erodible soils, and high precipitation\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e. Many tropical rivers are experiencing increased sediment loading as a result of land clearing for new development and agriculture\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. The removal of forest vegetation exposes soil to erosion from precipitation and wind while also reducing the soil holding capacity of the cleared area by reducing vegetation root depth and root mass\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e. Elevated sediment concentrations in rivers increase turbidity and reduce light availability and can alter the geomorphology of riverbeds\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e. One study of Kolombangara, a high tropical island in the Solomon Islands, estimated that if no land management practices were put in place, a 10% increase in logging activity would result in an increase in sediment runoff of approximately 1500% island-wide, equating to an annual transport of approximately 3,000 tons of sediment from land through rivers to the ocean\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e. If forests are replaced by agriculture or population centers in these regions, nutrient loading in rivers would likely increase substantially. Conversion of native forest to agriculture utilizing high-nitrogen fertilizer could result in a 25-fold increase in nitrogen export from the system, and urbanization at the same scale could result in a nearly 50-fold increase in nitrogen export as compared to native forest\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e. Understanding the role of land use in nutrient and sediment runoff is essential to understanding alterations in aquatic ecosystems as well as effective planning, management, and restoration of altered watersheds.\u003c/p\u003e \u003cp\u003eLand use change has resulted in as much as a 20-fold increase in nitrogen (N) and phosphorus (P) concentrations of many rivers worldwide as compared to pre-industrial levels\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e. Nitrogen and phosphorus enter aquatic ecosystems as a result of atmospheric deposition as well as excess fertilizer application on agricultural plots, industrial and domestic activities such as detergent use, and untreated or incompletely treated human and animal waste\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e. As either N or P tend to be the limiting nutrient for primary production in freshwater systems, elevated concentrations of these nutrients can have a number of deleterious effects including: the proliferation of algae and cyanobacteria resulting in increased levels of harmful algal toxins, reduction in dissolved oxygen, and alterations to flora and fauna communities\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e,\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e. In addition to altering the absolute levels of N and P, anthropogenic nutrient loading in freshwater systems can alter the stoichiometric ratios of dissolved N and P, which can shift trophic interactions and biogeochemical cycling in freshwater ecosystems\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eWatersheds of high volcanic islands are particularly vulnerable to impacts of land clearing. In tropical areas, these high islands often have high annual rainfall and periodic large storms that can quickly displace fertilizer from agricultural fields. The steep slopes and highly erodible soils on these islands also mean that rain events can result in large amounts of terrigenous sediments running into freshwater and coastal systems\u003csup\u003e\u003cspan additionalcitationids=\"CR17\" citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. In addition to the potential ecological impacts of runoff, people living on these islands are often dependent on rivers for clean water for drinking and bathing. Thus, freshwater bodies with increased nutrient and sediment pollution can reduce human health and well-being\u003csup\u003e\u003cspan additionalcitationids=\"CR20\" citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e through exposure to runoff as well as algal and bacterial blooms. Tropical islands are experiencing rapid growth in their populations, tourism, industry, logging, and large-scale agriculture resulting in the clearing of forest for farms and developments\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e,\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e,\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e. As a result, it is important to understand the impacts of active development on the nutrient and sediment concentrations in rivers on tropical islands. Water quality standards for nutrient and sediment concentrations have been established for some tropical regions\u003csup\u003ee.g.\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e, but have not been established or adopted universally across the Tropics. Thresholds for water quality standards are generally based on impacts of river chemistry for both human health\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e and ecosystem function\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e, so regularly exceeding water quality thresholds poses broader concerns for human and environmental outcomes.\u003c/p\u003e \u003cp\u003eHere, we focus on rivers of Moorea, French Polynesia (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), to examine how nutrient and sediment concentrations relate to human alterations of the landscape. Moorea is a high volcanic island with a rapidly changing landscape due to increased tourism, housing, and industrial development as well as increasing agriculture\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e,\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e. Many people on Moorea have raised issues regarding the state of rivers on the island, citing observations of turbid water, algal and bacterial blooms that result in rashes, and reduced abundance of a freshwater shrimp species that is an important traditional food (\u003cem\u003eAtger and Punu pers. obs.\u003c/em\u003e). Further, Moorea is home to near-shore coral reef ecosystems, and both resident observations and recent academic work suggests that terrestrially-derived nutrients have a negative impact on near-shore reefs\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e,\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e, which could compromise the ecosystem services (e.g. fisheries and tourism) they provide. However, to date no study has assessed nutrient and sediment loading in rivers on Moorea as it relates to land use. In response to this community and scientific need, we tested the following hypotheses about watersheds in Moorea:\u003c/p\u003e \u003cp\u003e \u003cem\u003e1. Watersheds with higher proportions of cleared land will have higher sediment loading in rivers.\u003c/em\u003e \u003c/p\u003e \u003cp\u003e \u003cem\u003e2. Watersheds with higher proportions of cleared land and higher populations will have high concentrations of nitrogen and phosphorus.\u003c/em\u003e \u003c/p\u003e\u003cp\u003e \u003cem\u003e3. Water nutrient and sediment concentrations will increase following recent precipitation.\u003c/em\u003e \u003c/p\u003e\u003cp\u003e \u003cem\u003e4. Watersheds with the highest percentages of land use and higher populations will exceed nutrient and sediment thresholds established for human and environmental protection in similar systems.\u003c/em\u003e \u003c/p\u003e \u003cp\u003eTo test these hypotheses, we collected water samples from 16 streams on Moorea across rainy and dry seasons in two years. Water samples were analyzed for nitrate (NO3-), nitrite (NO2-), and ammonium (NH4+), the three nitrogen species that comprise dissolved inorganic nitrogen (DIN), as well as orthophosphate (PO43-) and total suspended solids (TSS). We integrated stream chemistry data with precipitation, watershed size, land use, and census data, and used principal components analysis (PCA) and linear modeling to evaluate how local factors influenced the nutrient and sediment dynamics of streams across the island. Finally, we compared nutrient and sediment loading in streams of Moorea to similar systems in other tropical islands using previously-established water quality thresholds to assess potential concerns regarding human health and alteration to ecosystem function.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003ePatterns in Rainfall\u003c/h2\u003e \u003cp\u003ePrecipitation in Moorea was seasonal but occurred during every month of the study (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Within our sampling dates, February of 2018 had the highest cumulative monthly rainfall with 866 mm of precipitation on the North shore, 659 mm on the East shore, and 772 mm on the West shore. The rainy season in 2019 had lower monthly precipitation than in 2018 with a peak of 234 mm in February on the West shore, and 465 and 274 mm in March on the North shore and East shore, respectively. August was the driest month of both years and precipitation totaled 103 mm on the North shore, 56 mm on the East shore, and 39 mm on the West shore in 2018; and 86 mm of on the North shore, 23 mm on the East shore, and 28 mm on the West shore in 2019.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eWatershed and Seasonal Differences in Water Chemistry\u003c/h3\u003e\n\u003cp\u003eWater chemistry varied considerably by nutrient, season, and watershed across the sampling period (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, Supplementary materials S1). DIN concentrations ranged from a minimum of 0.05 \u0026micro;M measured in a sample collected from Papetoai during the dry season, to 37.7 \u0026micro;M measured in a sample collected in Paopao during the rainy season. DIN differed significantly across watersheds (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and seasons (p\u0026thinsp;=\u0026thinsp;0.03; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea, Supplementary materials S2). We found no significant differences between seasons within each watershed in the post-hoc Tukey test, though the significant interaction between watershed and season in the overall model indicates that rainy season DIN concentrations were higher than dry season in some watersheds (e.g., Papetoai and Maharepa) while the opposite pattern was observed in others (e.g., Ha\u0026rsquo;apiti and Haumi). The watershed with the highest mean DIN concentration was Paopao (18.20 \u0026micro;M), and the lowest mean concentration was found in Atiha (3.67 \u0026micro;M).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe nitrogen species comprising the majority of DIN was nitrate, followed by ammonium and finally nitrite. Nitrate concentrations differed significantly by watershed (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), and there was a significant interaction between watershed and season (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001); like DIN, the post-hoc Tukey test revealed no significant differences in nitrate concentrations between seasons within watersheds (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eb). Ammonium concentrations differed significantly by watershed (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), while the interaction with season was not significant. Teavaro and Haumi had higher ammonium levels than the other watersheds (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ec; the mean\u0026thinsp;\u0026plusmn;\u0026thinsp;s.e. ammonium concentration in these watersheds was 1.64\u0026thinsp;\u0026plusmn;\u0026thinsp;0.31 \u0026micro;M and 1.29\u0026thinsp;\u0026plusmn;\u0026thinsp;0.15 \u0026micro;M, respectively, whereas across watersheds the mean ammonium concentration was 0.63\u0026thinsp;\u0026plusmn;\u0026thinsp;0.12 \u0026micro;M). Nitrite concentrations differed significantly by watershed (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) but not by season although there was a significant interaction of watershed and season (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.003). The rainy season nitrite concentrations were significantly lower than the dry season concentrations in Haumi (difference\u0026thinsp;\u0026plusmn;\u0026thinsp;s.e. = 0.38\u0026thinsp;\u0026plusmn;\u0026thinsp;0.15 \u0026micro;M, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Like ammonium, Teavaro and Haumi had higher nitrite values relative to the other watersheds (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ed; nitrite concentration in these watersheds was 0.76\u0026thinsp;\u0026plusmn;\u0026thinsp;0.16 \u0026micro;M and 0.73\u0026thinsp;\u0026plusmn;\u0026thinsp;0.18 \u0026micro;M, respectively, whereas across watersheds the mean ammonium concentration was 0.21\u0026thinsp;\u0026plusmn;\u0026thinsp;0.06 \u0026micro;M).\u003c/p\u003e \u003cp\u003ePhosphate concentrations ranged from a minimum of 0.19 \u0026micro;M from a sample collected in Teavaro in the dry season to a maximum of 7.58 \u0026micro;M in a sample collected from Paopao 3 in the rainy season. Phosphate differed significantly among watersheds (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and season (p\u0026thinsp;=\u0026thinsp;0.03) and had a significant interaction of watershed and season (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.007) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ee). The highest mean phosphate concentration from rivers that flowed year-round was observed in Paopao 3 (2.70 \u0026micro;M) while the lowest was in Paopao 1 (0.98 \u0026micro;M). Phosphate concentrations were the only nutrient species significantly associated with TSS concentrations (β\u0026thinsp;=\u0026thinsp;0.369, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, and r\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.33). The N:P ratio ranged between a minimum of 0.03 in Papetoai, and a maximum of 54 in Paopao 1. The mean N:P ratio was highest in Paopao 1 (23.6) and lowest in Maharepa (2.25). N:P differed significantly across watersheds (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and had a significant interaction with watershed and season (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), with Paopao 1 and Opunohu 2 having significantly higher N:P in dry seasons than the rainy seasons (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ef).\u003c/p\u003e \u003cp\u003eTotal suspended solid concentrations ranged from below the detection limit in some samples from Atiha, Ha\u0026rsquo;apiti, and Maharepa to a maximum of 902 mg/L found in a sample collected in Paopao 1 during the rainy season. Mean TSS was higher during the rainy season than the dry season in 10 of the 12 watersheds for which TSS data were collected in both rainy and dry seasons (i.e. all sites except Pihaena, Paopao, and Paopao 3). Differences in TSS among seasons were significant (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.002) but the interaction between watershed and season was not significant (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.05) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Although differences in TSS among watersheds were not significant after accounting for season, variability in TSS concentration across watersheds was impressive: The lowest mean TSS concentration was recorded in Ma\u0026rsquo;atea at 2.5 mg/L while the highest was recorded in Paopao 1 at 137 mg/L. The next highest mean TSS concentration was in Paopao 2 at 117 mg/L, followed by Papetoai with 46 mg/L, less than half the concentration of the two largest Paopao sub-watersheds.\u003c/p\u003e\n\u003ch3\u003eLand-use and population across Moorea\u003c/h3\u003e\n\u003cp\u003eOur analysis of land use focused on the percent of cleared land in each focal watershed using the consensus land cover classification from a series of Worldview-3 images collected in 2018 (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e)\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e. Island-wide model accuracy was 98% based on a spatially random sample of validation points. Using a hierarchical sampling scheme to balance across training classes, classification accuracy was 82%. Most of the classification discrepancies came from errors of omission of exposed soil, which was often co-occurring with intensive monoculture and captured by that class, and errors of omission of small buildings which were lost as noise to their surrounding vegetation buffer. In both cases, these discrepancies fell within the broader category of \u0026ldquo;cleared land\u0026rdquo; and so were determined to be acceptable. The watershed with the highest proportion of cleared land in 2018 was Paopao 2 (23%), followed by Opunohu 1 (10.4%) while the lowest was Ha\u0026rsquo;apiti (0.8%) (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The human population on Moorea was 17,463 residents in 2017, with the largest population in the Paopao basin (1281 residents) and the smallest in Pihaena (44 residents) (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Population was not significantly related to watershed size or percentage of the watershed that was classified as cleared land (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.05 but β\u0026thinsp;\u0026gt;\u0026thinsp;0 in all cases), but these null associations were driven by the large area and broad-scale land clearing in the Opunohu basin (which has a very low population density): among all watersheds except those in the Opunohu basin, population was positively associated with watershed size and percentage of watershed that was cleared (R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.60 and 0.32, and \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.002 and 0.04, respectively).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eWatershed parameters including orientation of shore, watershed area, population size, Strahler stream order of the sampled river, stream length from source to sampling location, proportion of the watershed that is clear of forest (including agriculture and development). Sub-watersheds comprising main Opunohu and Paopao basins indicated with italics.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eWatershed\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eShore\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eWatershed area (km\u003c/em\u003e\u003csup\u003e\u003cem\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e)\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003eStream order\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cem\u003eStream length (km)\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cem\u003ePopulation\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u003cem\u003eCleared (%)\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAtiha\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eW\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e239\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e3.4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHa'apiti\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eW\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e372\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.8\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHaumi\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e422\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMa'atea\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e518\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e3.5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMaharepa\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e698\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e8.8\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOpunohu\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e15.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e5.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e159\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e9.5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eOpunohu 1\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e9.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e5.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e149\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e10.4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eOpunohu 2\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e5.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e5.7\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePaopao\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e8.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1281\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e15.6\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003ePaopao 1\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e280\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e3.5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003ePaopao 2\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e721\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e22.7\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003ePaopao 3\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e280\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e9.2\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePapetoai\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e798\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e5.3\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePihaena\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e44\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e7.4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTeavaro\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e367\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e1.3\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVaianae\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eW\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e233\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e\n\u003ch3\u003ePCA of watershed development and river chemistry\u003c/h3\u003e\n\u003cp\u003eSample points across the watersheds and seasons were distributed across the ordination space in the PCA (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). The first PC axis explained 47.6% of the variability in river chemistry, was positively associated with PO\u003csub\u003e4\u003c/sub\u003e and TSS (PC1 loadings of 0.55 and 0.29, respectively), and was negatively correlated with N:P and DIN (PC1 loadings of -0.62 and \u0026minus;\u0026thinsp;0.47, respectively). The second PC axis explained 35.5% of the variability and was positively correlated with all of the watershed chemistry variables (PC2 loadings of 0.41, 0.68, 0.53, and 0.28, for PO43-, TSS, DIN, and N:P, respectively) (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). The third and fourth PC axes explained 10.4% and 6.6% of the variability, respectively. Of the environmental variables we considered (watershed identity, recent precipitation, % cleared, watershed area, and population), watershed identity, precipitation and % cleared were significantly related to the first two axes of the ordination space (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.001, 0.001, and 0.045, respectively): watershed identity most strongly explained the spread of the data in the ordination space (r2\u0026thinsp;=\u0026thinsp;0.48), followed by recent precipitation (r\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.25), and percent of watershed cleared in 2018 (r2\u0026thinsp;=\u0026thinsp;0.04). Whereas recent precipitation explained variation in parameter space associated with phosphate and total suspended solids, cleared land was more strongly associated with dissolved inorganic nitrogen.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePCA axis loadings of phosphate, dissolved inorganic nitrogen, N:P, and total suspended solids concentrations.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eParameter\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003ePC1\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003ePC2\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003ePC3\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cem\u003ePC4\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDIN (\u0026micro;M)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.47\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.53\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.55\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.44\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePO\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e3\u0026minus;\u003c/sup\u003e (\u0026micro;M)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.55\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.41\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.52\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN:P ratio\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.62\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.28\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.68\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTSS (mg/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.29\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.62\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.25\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e\n\u003ch3\u003eLinear models of mean and maximum nutrient and sediment concentrations\u003c/h3\u003e\n\u003cp\u003eAcross watersheds, land use and population were important factors affecting mean seasonal total suspended solids in water samples. A linear model including the season, cleared land percentage, and population of each watershed revealed that mean river TSS was highest during the rainy season, and in watersheds with more land clearing and higher populations (F\u003csub\u003e3,23\u003c/sub\u003e = 4.24, R2\u0026thinsp;=\u0026thinsp;0.36, p\u0026thinsp;=\u0026thinsp;0.02). These factors were not significant (p\u0026thinsp;\u0026gt;\u0026thinsp;0.05) predictors in a watershed-level linear model of dissolved inorganic nitrogen. Using the complete (daily) dataset, a linear mixed effects model including the cleared land percentage, population, and precipitation as fixed effects and site as a random effect showed that both TSS and DIN were positively associated with recent precipitation (p\u0026thinsp;=\u0026thinsp;0.004 and 0.04, and conditional R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.10 and 0.49, respectively). The seasonal role of precipitation in the system led to increasingly high river TSS and DIN concentrations in the most cleared watersheds (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). Mixed effects models also revealed seasonal dynamics for river phosphate, with higher phosphate concentrations following recent precipitation (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, conditional r2\u0026thinsp;=\u0026thinsp;0.33).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eWater quality thresholds\u003c/h2\u003e \u003cp\u003eRelaxed and stringent water quality thresholds were selected for both sediment and nutrient concentrations based on previous work in the Tropics (5 mg/L and 50 mg/L for TSS; and 0.1 mg/L and 0.18 mg/L for DIN)\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e,\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e,\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e,\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e. Water samples from all watersheds, except for Ma\u0026rsquo;atea, exceeded the lower TSS water quality threshold of 5 mg/L at least once. On average, watersheds on Moorea exceeded this threshold in 39% of sampling events (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). The maximum exceedance percentage was found in Pihaena at 78% of the time. Eight of the watersheds included in this study exceeded the upper TSS threshold of 50 mg/L at least once. Across Moorea, this threshold was exceeded on average 8% of the time. Six watersheds exceeded Hawaiian water quality standards with values of 50 mg/L at least 10% of the time (Haumi: 11%, Opunohu 1: 10%, Paopao 1 and 2: 14%, Papetoai: 12%, and Pihaena: 22%). All watersheds except Atiha and Vaianae exceeded both the lower and the upper DIN thresholds (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). On average, stream samples across Moorea exceeded 0.1 mg/L DIN 46% of the time and 0.18 mg/L 20% of the time. Opunohu 2, Paopao 1, and Paopao 2 exceeded the lower standard in every sample. Although the frequency by which rivers exceeded thresholds was not significantly associated with the proportion of the watershed cleared of forest, the slope of the relationship between threshold exceedance frequencies and watershed clearing was almost always positive (i.e. β\u0026thinsp;\u0026gt;\u0026thinsp;0, p\u0026thinsp;\u0026gt;\u0026thinsp;0.05, except for the 0.18 mg/L DIN threshold).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eExceedance percentages by watershed of low and high TSS and DIN water quality thresholds. \u003cb\u003eBold values represent exceedance frequencies of concern (\u0026gt;\u0026thinsp;10%)\u003c/b\u003e.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eWatershed\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003e\u0026gt;\u0026thinsp;5 mg/L TSS\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003e\u0026gt;\u0026thinsp;50 mg/L TSS\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003e\u0026gt;\u0026thinsp;0.1 mg/L DIN\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cem\u003e\u0026gt;\u0026thinsp;0.18 mg/L DIN\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAtiha\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHa'apiti\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e50\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e24\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHaumi\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e22\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e11\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e67\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e33\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMa'atea\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e11\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMaharepa\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e42\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e12\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOpunohu 1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e50\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e10\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOpunohu 2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e30\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e100\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e77\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePaopao 1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e43\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e14\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e100\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e71\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePaopao 2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e43\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e14\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e100\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e14\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePapetoai\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e62\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e12\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e88\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e12\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePihaena\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e78\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e22\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e23\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e13\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTeavaro\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e50\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e73\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e27\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVaianae\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e33\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOverall\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e39\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e46\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e20\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThrough a combination of island-wide water sampling and remote sensing analyses, we identified associations between land use at the watershed scale and the nutrient and sediment concentrations of rivers on Moorea. Dissolved inorganic nitrogen (DIN), phosphate, total suspended solids (TSS), and the nitrogen to phosphorus (N:P) ratio differed across rivers and between the rainy and dry seasons. Further, DIN concentrations and TSS were positively related to the percentage of land in a watershed that was cleared of forest and that relationship was mediated by seasonal rainfall. Phosphate concentrations were not related to the metrics of land use change we considered, but were strongly associated with recent rainfall, and are likely the result of weathering of the island\u0026rsquo;s phosphate rich geology. Importantly, DIN and TSS concentrations regularly exceeded thresholds established in similar systems to identify watersheds that may pose a danger to human health and that of aquatic biota.\u003c/p\u003e\n\u003ch3\u003eNutrient enrichment and its implications\u003c/h3\u003e\n\u003cp\u003eIn the PCA and linear models, we observed a strong positive relationship between land-clearing and high DIN concentrations that was also influenced by recent precipitation. Similar relationships between DIN and land-clearing combined with human population were found in studies of rivers in American Samoa\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e and Guam\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e. These results indicate that these two watershed parameters \u0026ndash; cleared land and population \u0026ndash; may be sufficient to make approximate predictions of river DIN concentrations in steep tropical islands. This could be a valuable tool to remotely approximate DIN concentrations in unstudied watersheds on steep-islands, using only satellite-derived data on cleared land and census data on human populations. It is important to note that a more data complex runoff model would be required to calculate flux of DIN from watersheds\u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e. However, estimates of DIN concentrations can be valuable in assessing potential human or environmental health concerns\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e,\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e and regulatory thresholds are commonly based on concentrations rather than fluxes\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eWe observed substantially higher DIN concentrations in rivers on Moorea than previous studies on tropical islands. The highest observed DIN sample in our study, 2.09 mg/L (37.72 \u0026micro;M), was an order of magnitude higher than the maximum measured in Guam (0.20 mg/L)\u003csup\u003e30\u003c/sup\u003e, and much higher than the maximum in American Samoa (0.75 mg/L)\u003csup\u003e31\u003c/sup\u003e. Another study in American Samoa recorded a mean of 8.5 \u0026micro;M DIN in a heavily disturbed watershed\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e which is close to the overall mean observed in our study (0.55 mg/L, or 9.05 \u0026micro;M), including data from all watersheds from minimally to highly disturbed. As a result of this difference, DIN concentrations in almost all watersheds on Moorea exceeded the water quality threshold of 0.1 mg/L established by these studies. DIN values in Moorea also frequently exceeded the 0.18 mg/L threshold from the Hawaii water quality standards, despite values being comparable to watersheds in Hawaii with similar land use\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e,\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eOur PCA analysis and linear mixed models indicate that DIN runoff on Moorea is likely primarily a result of agricultural activity as the percent cleared land classification in our satellite image analysis is largely comprised of agricultural land. One of the primary commercial crops grown on Moorea is pineapples\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. Pineapple farming requires year-round applications of nitrogen fertilizer and has contributed to nitrogen-loading in rivers across the tropics\u003csup\u003e\u003cspan additionalcitationids=\"CR36\" citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e. The nitrogen in these fertilizers is usually in the form of urea (CH4N\u003csub\u003e2\u003c/sub\u003eO) which transforms to ammonium and then nitrate via nitrification in the soil due to biological activity\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e. Nitrate was the most abundant form of nitrogen we observed. Nitrate is highly water soluble and thus is easily transported from terrestrial to aquatic environments via surface runoff. Nitrogen-enriched runoff is likely exacerbated on Moorea by the lack of runoff control and the lack of requirements to keep riparian buffers between fields and rivers\u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003ePrevious work on Moorea measuring nitrogen isotopes in nearshore macroalgae observed elevated δ15N in algae tissue sampled near population centers indicating that sewage leaching directly into the nearshore environment via submarine groundwater discharge was likely a large contributor to nitrogen pollution\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e,\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e,\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e. Direct measurements of DIN concentrations in groundwater and submarine groundwater seeps in these studies recorded DIN levels as high as 45 \u0026micro;M\u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e. Such groundwater seeps could be at the root of high DIN concentrations reported in American Samoa, described above. It is likely that groundwater carrying DIN is also entering streams through hyporheic exchange in the lower reaches of the watershed due to the porous nature of the volcanic geology\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e and high head pressure from the island\u0026rsquo;s short, steep slopes\u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e. This exchange may be enhanced in the lower reaches of the rivers as they have been heavily channelized and fortified for flood control (\u003cem\u003eKN personal observation\u003c/em\u003e) which has enhanced stream bed erosion and deepening of river channels\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e resulting in an increase in head near the terminus where our samples were collected.\u003c/p\u003e \u003cp\u003eIn both the PCA and linear regression analysis, riverine phosphate on Moorea was decoupled from both land clearing and population, suggesting that agriculture and urbanization were not directly responsible for the differences in phosphate concentrations we observed around the island. Phosphate concentrations ranged from a 0.19 \u0026micro;M sample collected in Teavaro to a 7.58\u0026micro;M sample collected in Paopao. Phosphate concentrations differed significantly among watersheds (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and season (p\u0026thinsp;=\u0026thinsp;0.01) and had a significant interaction of watershed and season (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.007). The range of phosphate concentrations recorded in this study are comparable to those found in a study of watersheds on Oahu, Hawaii, across base and stormflow and including watersheds considered to be forested, agricultural and urban\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eAnthropogenic sources of phosphate often come in the form of agricultural fertilizers, however pineapples need little to no phosphorus fertilizer where soils have a naturally high phosphorus content as they likely do on Moorea\u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e. Soils on the island are derived from erosion of volcanic basalts which tend to be rich in phosphorus, and the availability of phosphorus in the soil of a given area is additionally dependent on numerous factors including age, weathering, and localized precipitation\u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e. Phosphate concentrations in rivers are often tied to sediment concentrations as phosphate can desorb from sediment particles in transit\u003csup\u003e\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e; this pattern was evident in our dataset as well. As there are at least eight different volcanic soils on Moorea\u003csup\u003e\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e, differences in riverine phosphate between watersheds could be related to erosion, vegetative utilization, or leaching of soils with different phosphate levels depending on which soil types happen to be exposed in a given watershed. Groundwater and submarine groundwater discharge on Moorea is also elevated in phosphate relative to surface water. This is likely due to weathering of phosphate rich basaltic rock in aquifers\u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e. It is likely that phosphate also enters streams on Moorea via hyporheic exchange. Differences in phosphate in rivers on Moorea may also be the result of differences in bedrock phosphate concentrations, underground mobilization rates, and/or differences in groundwater-surface water hyporheic exchange rates. As anthropogenic sources of DIN increase on the island due to expanded agriculture and urbanization, it is likely that the N:P ratio will also increase as nitrogen enrichment outpaces phosphate mobilization.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eDrivers and consequences of riverine sediments\u003c/h2\u003e \u003cp\u003eBoth the PCA and regression analyses illustrated that TSS concentrations were significantly associated with the percentage of cleared land in a given watershed and recent precipitation, suggesting that both land use and climatic drivers contribute to the mobilization of sediment. Similar statistical modeling approaches have been used to predict TSS in other mountainous watersheds\u003csup\u003e\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e. These models do not predict discharge or sediment flux, but like our DIN model they could be a useful framework for a first order assessment of potential human health or environmental impact concerns in similar regions. TSS thresholds and regulations are often based on concentrations, not flux\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. In locations where data are lacking, such as many high islands in the South Pacific, this simple linear modeling approach may provide valuable insights where there is insufficient data to inform a more complicated runoff model like the Soil and Water Assessment Tool (SWAT), Revised Universal Soil Loss Equation (RUSLE), Modified Morgan\u0026ndash;Morgan\u0026ndash;Finney (MMF) model, or others that are commonly paired with remote sensing analysis to assess relationships between land use and TSS\u003csup\u003e\u003cspan additionalcitationids=\"CR50\" citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eTSS concentrations in all studied watersheds on Moorea, except for Ma\u0026rsquo;atea, exceeded the water quality threshold of 5 mg/L TSS for safe drinking, bathing and cleaning\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e,\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e. TSS values from six watersheds also exceeded the 50 mg/L threshold representing potentially lethal levels for fish species\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e, for an overall exceedance rate of 8%. We also observed high variability in TSS levels between samples, including some outliers (e.g. \u0026gt;900 mg/L compared to mean of 26 mg/L in Paopao) that illustrate how sediment is mobilized in these flashy systems and how difficult it can be to capture differences between watersheds or seasons when much of the sediment that moves from land to river can mobilize during a few major rain events\u003csup\u003e\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e. The range of TSS values observed across watersheds experiencing varying degrees of land clearing and urbanization are within the ranges observed in a study across an urbanization gradient in Hawaii\u003csup\u003e\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u003c/sup\u003e and a study of a \u0026ldquo;pristine\u0026rdquo; rainforest watershed in Fiji that includes samples taken during a cyclone\u003csup\u003e\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e. TSS values observed in the most disturbed watersheds of our study (Paopao and Opunohu) were considerably lower than those observed in a study of urban and developing watersheds on Tahiti which recorded values as high as 33.3 g/L TSS in a watershed that was actively undergoing sizeable earthworks projects\u003csup\u003e\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eMuch of the existing land suitable for development in Moorea has been utilized, so new residential or industrial development requires considerable earth moving efforts, including forest clearing and terracing to create flat buildable plots. We did not observe use of soil retention measures (e.g. vegetative cover, mulch, silt fences, geotextiles) that have been used in similar steep tropical systems to reduce erosion\u003csup\u003e\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e,\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e\u003c/sup\u003e. Furthermore, a majority of the roads in the interior of the island are not paved, tend to follow river channels, and lack any soil retention devices (KN \u003cem\u003epers. observation)\u003c/em\u003e that have been successfully demonstrated in comparable systems\u003csup\u003e\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e\u003c/sup\u003e. As a result of their demonstrated ability to reduce erosion, implementation of erosion control measures such as those listed above are required by law on construction projects in Hawaii\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eLand clearing associated with agriculture is also likely a large contributor to TSS runoff as pineapple farming is known to contribute to high erosion rates\u003csup\u003e\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e\u003c/sup\u003e. In a recent study which scored farming practices in French Polynesia based on soil preservation, pineapple farming scored the lowest for its contributions to soil degradation\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. Across French Polynesia and in many other tropical climates pineapples are often grown on moderate to steep slopes making it difficult to employ erosion control measures such as cover crop or mulching between rows, which have been shown to be effective in reducing soil erosion in other agricultural systems\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e. Further, to maximize yield from a farmed plot, many farms in Moorea extend right up to riverbanks with no riparian zone or other buffer to slow the movement of sediment from the fields to the rivers (\u003cem\u003epers. observation\u003c/em\u003e). All of these conditions create high erosion potential, and contribute to the strong relationship between land clearing and TSS in rivers we observed in Moorea.\u003c/p\u003e \u003c/div\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis study provides the first island-wide assessment of riverine nutrient and sediment concentrations on Moorea, French Polynesia, and explores the relationships between river chemistry, weather, land use, and populations. Watersheds on Moorea, as well as other small high tropical islands worldwide, are developing rapidly with growing populations, tourism, logging and large-scale agriculture\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e,\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e\u003c/sup\u003e. Our study puts Moorea in context of these other efforts to study land use and its impacts on rivers. The relationships we observed between land use and river nutrient chemistry on Moorea are comparable to those observed in other developed and developing islands around the world\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e,\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e,\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e. In some cases, Moorea may be more susceptible to land-use initiated sediment erosion and runoff as indicated by our comparison with other studies with respect to TSS threshold exceedance. Perhaps most urgently, this study highlights watersheds on Moorea which present risks for human health and/or concerns regarding ecosystem processes based on nutrient and sediment thresholds. Finally, this study also serves as a point of comparison against future changes on Moorea as land clearing and population growth continue on the island. As Moorea, and other high tropical islands, continue to develop, impacts of watershed change are likely to increase with potentially profound consequences for human and ecosystem health.\u003c/p\u003e "},{"header":"Methods","content":"\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003cdiv id=\"Sec14\" class=\"Section3\"\u003e \u003ch2\u003eStudy Location\u003c/h2\u003e \u003cp\u003eMoorea is a volcanic island located in the Society Islands Archipelago of French Polynesia (17\u0026deg; 29\u0026prime; S, 149\u0026deg; 50\u0026prime; W). As the island eroded and subsided, a barrier reef formed separating the open ocean from a 53 km2 lagoon\u003csup\u003e\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e\u003c/sup\u003e. The island has a relatively small area of 134 km2, but features diverse topography. Since it was formed by a volcanic eruption approximately 1.6\u0026nbsp;million years ago\u003csup\u003e\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e\u003c/sup\u003e, Moorea has collapsed and eroded into numerous distinct watersheds separated by steep, sharp ridgelines with a highest elevation of 1207 m at Mount Tohiea\u003csup\u003e\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e\u003c/sup\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). A majority of the soils (98.7%) on the island formed from weathering and erosion of volcanic parent material with the remainder being coralline in origin\u003csup\u003e\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e. Of the eight soils of volcanic origin on Moorea, five originate from weathered basalt\u003csup\u003e\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eMoorea has been inhabited by humans since around 200 CE when it was settled by the Polynesians\u003csup\u003e\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e\u003c/sup\u003e, who brought with them a variety of nonnative plants and animals for cultivation. A 2015 study of vegetation on Moorea classified 32% of the island as urbanized and cultivated lands, 17% as novel habitat dominated by invasive species, 45% as hybrid habitats featuring a mix of introduced and native species, and 6% as native habitat which is primarily located on extremely steep slopes and at high elevation\u003csup\u003e\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e\u003c/sup\u003e. As of the 2017 census, Moorea was home to 17,357 permanent residents\u003csup\u003e\u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe largest watersheds on Moorea are two adjacent valleys on the north shore, Opunohu and Paopao (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), formed by a collapse of the volcano\u0026rsquo;s caldera\u003csup\u003e\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e\u003c/sup\u003e. Opunohu is home to a small population (159 people as of 2017), but hosts a large portion of the island\u0026rsquo;s commercial agriculture which is principally comprised of pineapple plantations and cattle pasture. Using the Strahler method, the Opunohu River is the only 4th order stream on the island\u003csup\u003e\u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e\u003c/sup\u003e. As an example of the steepness of watersheds on Moorea, more than 90% of the Opunohu valley is comprised of slopes greater than 10%\u003csup\u003e66\u003c/sup\u003e. Neighboring Paopao is home to the largest population center (1281 people as of 2017) and small mixed agricultural fields, and also features a 4th order stream. Its predominant sub-watersheds feature 2nd -4th order streams and span a range of areas. The smallest standalone watershed included in this study, Teavaro, is located on the eastern shore of the island and is only 1.3 km2, and home to 367 people. Teavaro and the rest of the rivers in this study are second and third order streams. The variety of watersheds in close proximity featuring a range of areas, land uses and populations makes Moorea a compelling location to investigate interactions between human activity and riverine nutrients and sediments.\u003c/p\u003e \u003cp\u003eIn order to study rivers and land use on Moorea, we first engaged the local community to receive feedback regarding key questions that would be of interest and value to the people of Moorea. The goals, objectives, and research locations for this study were shaped by meetings with community members organized by the Te Pu Atiti\u0026rsquo;a Polynesian Cultural Center located directly adjacent to the University of California Gump Research Station, where our research was based. These conversations precipitated a new community science organization on Moorea called Ati Vai (\u0026lsquo;Water Clan\u0026rsquo; in Tahitian). Members of Ati Vai contributed to the study design, provided local knowledge and site context, and aided with field data collection throughout this study. The combination of modern scientific tools and techniques with traditional and local knowledge of the island and its rivers was paramount to the success of this research.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eFocal watershed selection\u003c/h2\u003e \u003cp\u003eIn the first year, eight watersheds were chosen to represent a range of watershed sizes, land uses and population densities, and are located on all three shores of the island. These watersheds were: Atiha, Ha\u0026rsquo;apiti, Maharepa, Opunohu, Paopao, Pihaena, Teavaro, and Vaiane. The Opunohu watershed was evaluated at the level of its two primary sub-watersheds, which have vastly different human populations (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Three additional watersheds - Haumi, Ma\u0026rsquo;atea, and Papetoai \u0026ndash; as well as the primary sub-watersheds of the Paopao watershed (defined by 2nd \u0026minus;\u0026thinsp;4th order streams), were added in the second year of sampling to increase sampling across the ranges of watershed size and human impacts (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). All of the rivers included in this study are perennial, with the exception of Pihaena which has low or no flow during the dry season.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eWater sample collection\u003c/h2\u003e \u003cp\u003eWe collected water samples during two rainy and two dry seasons between January 2018 and September 2019 (rainy: January - March 2018 and February - March 2019, dry: August - September 2018 and 2019). During the study periods, water samples were collected as close to the mouths of the rivers as possible, but above brackish mixing zones. Where natural or manmade impediments to access existed, we made every effort to find a suitable sampling location as close to the river mouth as possible. During the above periods, samples were collected at least once a week with attention paid to capturing samples across river stages from low-flow to storm-flow. We collected storm-flow samples during or shortly after rain events to ensure that we captured the nutrient and sediment regimes during times of high flow.\u003c/p\u003e \u003cp\u003eRiver samples were collected at approximately 60% water depth (e.g. at 40 cm above the bottom in a 1 m deep river) at the center of flow. Samples were collected in acid washed 1-liter HDPE Cubitainers\u0026reg;, transported on ice to the lab at Gump Station, and kept refrigerated until processed. Samples were processed upon return to the lab within no more than 24 hours. Each sample was divided and filtered for nutrients or total suspended solids using a multi-channel peristaltic filtration system.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eNutrient analyses\u003c/h2\u003e \u003cp\u003eNutrient sub-samples were filtered through 47 mm 0.15 \u0026micro;m PES filter discs and then divided for separate analytical methods. NH4\u0026thinsp;+\u0026thinsp;analyses were conducted immediately after filtration using the OPA method\u003csup\u003e\u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e67\u003c/span\u003e\u003c/sup\u003e on a Turner Trilogy fluorometer fitted with an ammonium detection module (Minimum Detection Limit (MDL): 0.05 \u0026micro;M). The remaining samples were frozen at -20\u0026deg;C and transported to the Centre de Recherches Insulaires et Observatoire de l'Environnement (CRIOBE) on Moorea where they were analyzed for PO43- (MDL: 3 nM), NO3- (MDL: 2 nM), and NO2- (MDL: 2 nM) on an AA3 Auto-analyzer (SEAL Analytical, e.g. [\u003csup\u003e\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e\u003c/sup\u003e]). Total DIN per sample represents the sum of NO2-, NO3-, and NH4+. The molar N:P ratio was calculated for each sample by dividing DIN by phosphate.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eTotal suspended solids (TSS) analysis\u003c/h2\u003e \u003cp\u003eTSS samples were filtered using pre-dried and pre-weighed 47 mm GF/F (0.70 \u0026micro;m) filter discs. Filters were weighed on a Mettler-Toledo XS104 balance with 0.1 mg precision. Samples were filtered to 200 mL, or until the filters were completely clogged preventing more water to pass through, whichever came first. If filters clogged before filtering 200mL, we recorded the total volume filtered. After filtration, the GF/F filters were placed in a drying oven at 95\u0026deg;C for a minimum of 24 hours at which time they were re-weighed. TSS was calculated using Eq.\u0026nbsp;1:\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(\\:TSS=\\:\\frac{\\left(post-filtering\\:and\\:drying\\:weight\\right)-(pre-filtering\\:weight)}{volume\\:filtered}\\)\u003c/span\u003e \u003c/span\u003e [1]\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eLand cover classification\u003c/h2\u003e \u003cp\u003eThe Worldview-3 (WV3) Satellite (Maxar Technologies) collects repeat imagery with the 8-band WV110 sensor every 4.5 days with a ground sampling distance of 1.38m. We used a series of WV3 images collected during four days in 2018 that had little-to-no cloud cover. The imagery was aligned, georeferenced, orthorectified, and mosaicked using the Ortho Mapping workspace in ArcGIS Pro v3.1.2. Training data for land cover classification were generated using data from the Direction des Affaires Fonci\u0026egrave;res de la Polyn\u0026eacute;sie Fran\u0026ccedil;aise (the Directorate of Land Affairs, French Polynesia), and included the classes Water, Forest, Monoculture (intensive agriculture), Buffer/Agriculture (permaculture, orchards, and other vegetated, non-forested land), Buildings, Paved, Dirt and Sand. A U-Net pixel classifier was trained on a ResNet-50 architecture with 10% of training samples withheld for validation for each of the unique dates of imagery collection, yielding an accuracy of 0.86\u0026thinsp;\u0026plusmn;\u0026thinsp;0.04 (mean\u0026thinsp;\u0026plusmn;\u0026thinsp;std. dev. across date-wise models). Individual layers were then compiled into a \u0026ldquo;consensus\u0026rdquo; land cover map that used the modal classification for each pixel (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, [***DOI pending]). This approach reduced noise related to solar and topographic variability in pixel classification outputs from individual date layers, and overcame gaps in the dataset introduced by cloud cover. The consensus land cover map had a 98% island-wide classification accuracy.\u003c/p\u003e \u003cp\u003eWatersheds were delineated using the Hydrology tool in the Spatial analyst toolbox in ArcGIS Pro v3.1.2. Using a digital elevation model (DEM) from the Shuttle Radar Topography Mission Version 3 (30 m resolution)\u003csup\u003e\u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e69\u003c/span\u003e\u003c/sup\u003e, we identified the watersheds that feed into our river sampling locations using the Flow Accumulation and Flow Direction Tools in ArcGIS. Points along the rivers were identified towards the mouth of rivers of interest using the Snap Pour Point tool. Watersheds of interest for the project were then delineated using the Watershed tool.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003ePrecipitation and Census Data\u003c/h2\u003e \u003cp\u003eDaily precipitation data for Moorea for 2018 and 2019 were obtained for 5 meterological stations on the island through a license agreement with Meteo France. Census data for 2017 was obtained via a license agreement between the Gump Research Station and the Institut de Statistiques de la Polyn\u0026eacute;sie Fran\u0026ccedil;aise. The census data were originally collected in 102 districts on the island, often with multiple districts located inside of a watershed. In order to determine the population for each watershed, we aggregated census data based on spatial intersections between census districts and focal watersheds as described above.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003eTSS and DIN Concentration Thresholds\u003c/h2\u003e \u003cp\u003eLow (more relaxed) and high (more stringent) water quality thresholds for both DIN and TSS concentrations were selected from a review of literature from similar systems including Hawaii, Guam, the Solomon Islands and American Samoa. The low TSS threshold (5 mg/L) represents concerns for human consumption and bathing\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e,\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. The high TSS threshold (50 mg/L) is based on water quality standards established in Hawaii\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e and is based on the levels at which exposure to suspended sediment become lethal to a majority of fish species\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. According to the Hawaiian standards, TSS in streams should not exceed 50 mg/L for more than 10% of the time in the rainy season. The lower DIN concentration threshold of 0.1 mg/L was selected based on studies of runoff impacts to nearshore reefs in American Samoa\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e and Guam\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e. An exceedance of this threshold in streams for more than 20% of the time was associated with diminished diversity and size of coral communities in both regions. The higher threshold of 0.18 mg/L DIN was based on Hawaiian water quality standards which state that streams should not exceed this threshold more than 10% of the time in the rainy season\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. We calculated the percentage of time that samples collected from each watershed on Moorea exceeded these thresholds using Eq.\u0026nbsp;2:\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(\\:\\%\\:exceedance=\\:\\frac{\\#\\:samples\\:from\\:watershed\\:that\\:exceed\\:threshold}{total\\:\\#\\:samples\\:from\\:watershed}*100\\)\u003c/span\u003e \u003c/span\u003e [2]\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003eStatistical Analyses\u003c/h2\u003e \u003cp\u003eWe used linear models to analyze differences in river nutrients and TSS by watershed, season, and their interaction. When a significant watershed by season interaction existed, or to investigate differences between sites, we used post-hoc Tukey tests with adjusted alpha (p-value) by the number of tests. Data were 1\u0026thinsp;+\u0026thinsp;log transformed prior to analysis to improve model fit, based on inspection of q-q plots and residuals using the \u0026lsquo;qqplot\u0026rsquo;, \u0026rsquo;qqnorm\u0026rsquo;, and \u0026lsquo;resid\u0026rsquo; functions in R. Only watersheds that had samples from the wet and dry seasons were included in the linear models, so Pihaena was excluded because it only flowed in the wet season.\u003c/p\u003e \u003cp\u003eTo assess the significance of watershed variables (area, population, and percent cleared land) in explaining river chemistry in the wet and dry season, we used principal components analyses (PCA). PCA is a form of ordination analysis based on Euclidean distance that is well suited to complex datasets with covarying variables. Differences across watersheds and seasons were assessed based on a correlation matrix of dissolved inorganic nitrogen, phosphate, N:P, and total suspended solids levels at a given sampling event. All data were standardized (mean\u0026thinsp;=\u0026thinsp;0, SD\u0026thinsp;=\u0026thinsp;1) prior to analysis. River chemistry data were 1\u0026thinsp;+\u0026thinsp;log transformed before analysis. We used the \u0026lsquo;envfit\u0026rsquo; function in the R package \u0026lsquo;vegan\u0026rsquo; to test whether watershed characteristics, namely area, population, percent cleared land in 2018, watershed, and season could explain the variation in river chemistry\u003csup\u003e\u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e70\u003c/span\u003e\u003c/sup\u003e. Envfit finds vectors or factor averages of environmental variables, analogous to fitting a linear model, in ordination space.\u003c/p\u003e \u003cp\u003eTo further explore relationships between watershed variables and river conditions, we performed a series of linear models testing whether cleared land percentage and population predict nutrient and sediment concentrations. Models were fit using the mean concentrations of nutrients and sediment for each watershed across all seasons and years. Additional models were fit using maximum observed nutrient and sediment values to elucidate relationships between land use and high flow runoff events. We also analyzed the relationship between population from the census data and the area of cleared land from the land cover classification using linear models. Finally, linear mixed-effects models were used to detect impacts of precipitation events on river chemistry by including watershed as a random intercept and including a continuous measure of recent precipitation as an independent variable.\u003c/p\u003e \u003cp\u003eAll analyses were conducted in R\u003csup\u003e\u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e71\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003ch2\u003eFunding\u003c/h2\u003e \u003cp\u003efor this research was provided by the NSF GRFP to KN, NSF Career grant OCE\u0026mdash;1547952 to DEB, NSF grant OCE-1637396 to the Moorea Coral Reef LTER, and the Zegar Family Foundation, The Schmidt Family Foundation, and The Worster Summer Research Fellowship. Funding for education and outreach related to this work was provided by the ASLO Global Outreach Initiative.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eConceptualization: K.N. with input from T.A., T.P., and D.E.B. Data curation: K.N. and C.J.Formal analysis: K.N., C.J., and J.A.H.Funding acquisition: K.N. and D.E.B.Investigation: K.N.Methodology: K.N., T.A., T.P., and D.E.B. Project administration: K.N. and D.E.B.Visualization: K.N., C.J., and J.A.H.Supervision: D.E.B.Writing \u0026ndash; original draft: K.N.Writing \u0026ndash; review \u0026amp; editing: All authors\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eWe thank the community of Moorea for hosting this research on Tahitian land and waters. Credit to Benoit Espiau at CRIOBE for his meticulous nutrient analyses. Particular mention to Corinne Fuchs, Kyla Pierce, Maya Gorgas and the members of \u0026lsquo;Ati Vai for their assistance collecting, processing and analyzing samples. Thank you to the staff of the University of California Richard B. Gump Research Station for all of the work solving logistical, administrative and mechanical issues. We could not have done this work without you. With respect to the spelling of Moorea, we followed the Raapoto transcription system, but also recognize other community members follow the Te Fare Vanā\u0026rsquo;a transcription system where the island name is spelled with an \u0026rsquo;eta (Mo\u0026rsquo;orea).Funding for this research was provided by the NSF GRFP to KN, NSF Career grant OCE\u0026mdash;1547952 to DEB, NSF grant OCE-1637396 to the Moorea Coral Reef LTER, and the Zegar Family Foundation, The Schmidt Family Foundation, and The Worster Summer Research Fellowship. Funding for education and outreach related to this work was provided by the ASLO Global Outreach Initiative.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eData from this study are stored on EDI (***DOI available upon manuscript acceptance), and code to reproduce analyses are on github (***Link available upon manuscript acceptance). Please contact the corresponding author for data requests ([email protected]).\u003c/p\u003e\n\u003ch2\u003eAdditional information\u003c/h2\u003e \u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e "},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eWinkler, K., Fuchs, R., Rounsevell, M. \u0026amp; Herold, M. Global land use changes are four times greater than previously estimated. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cb\u003e12\u003c/b\u003e, 2501 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePiao, S. et al. Changes in climate and land use have a larger direct impact than rising CO2 on global river runoff trends. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e 104, 15242\u0026ndash;15247 (2007).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGreen, P. A. et al. Pre-industrial and contemporary fluxes of nitrogen through rivers: a global assessment based on typology. \u003cem\u003eBiogeochemistry\u003c/em\u003e \u003cb\u003e68\u003c/b\u003e, 71\u0026ndash;105 (2004).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKroon, F. J. et al. River loads of suspended solids, nitrogen, phosphorus and herbicides delivered to the Great Barrier Reef lagoon. \u003cem\u003eMar. Pollut. Bull.\u003c/em\u003e \u003cb\u003e65\u003c/b\u003e, 167\u0026ndash;181 (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHoffmann, T. et al. Human impact on fluvial regimes and sediment flux during the Holocene: Review and future research agenda. \u003cem\u003eGlob. Planet Change\u003c/em\u003e. \u003cb\u003e72\u003c/b\u003e, 87\u0026ndash;98 (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLabri\u0026egrave;re, N., Locatelli, B., Laumonier, Y., Freycon, V. \u0026amp; Bernoux, M. Soil erosion in the humid tropics: A systematic quantitative review. \u003cem\u003eAgriculture Ecosystems Environment\u003c/em\u003e. \u003cb\u003e203\u003c/b\u003e, 127\u0026ndash;139 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTanaka, Y., Minggat, E. \u0026amp; Roseli, W. The impact of tropical land-use change on downstream riverine and estuarine water properties and biogeochemical cycles: a review. \u003cem\u003eEcol. Processes\u003c/em\u003e. \u003cb\u003e10\u003c/b\u003e, 40 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBartley, R. et al. Relating sediment impacts on coral reefs to watershed sources, processes and management: A review. \u003cem\u003eSci. Total Environ.\u003c/em\u003e \u003cb\u003e468\u0026ndash;469\u003c/b\u003e, 1138\u0026ndash;1153 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDowning, J. A. et al. The impact of accelerating land-use change on the N-Cycle of tropical aquatic ecosystems: Current conditions and projected changes. \u003cem\u003eBiogeochemistry\u003c/em\u003e \u003cb\u003e46\u003c/b\u003e, 109\u0026ndash;148 (1999).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAllan, J. D. Landscapes and Riverscapes: The Influence of Land Use on Stream Ecosystems. \u003cem\u003eAnnu. Rev. Ecol. Evol. Syst.\u003c/em\u003e \u003cb\u003e35\u003c/b\u003e, 257\u0026ndash;284 (2004).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWenger, A. S. et al. Best-practice forestry management delivers diminishing returns for coral reefs with increased land-clearing. \u003cem\u003eJ. Appl. Ecol.\u003c/em\u003e \u003cb\u003e57\u003c/b\u003e, 2381\u0026ndash;2392 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCorrell, D. L. The Role of Phosphorus in the Eutrophication of Receiving Waters: A Review. \u003cem\u003eJ. Environ. Qual.\u003c/em\u003e \u003cb\u003e27\u003c/b\u003e, 261\u0026ndash;266 (1998).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e\u003cem\u003eRiverine Ecosystem Management: Science for Governing Towards a Sustainable Future\u003c/em\u003e. (Springer International Publishing, Cham, (2018). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/978-3-319-73250-3\u003c/span\u003e\u003cspan address=\"10.1007/978-3-319-73250-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCanning, A. D. \u0026amp; Death, R. G. The influence of nutrient enrichment on riverine food web function and stability. \u003cem\u003eEcol. Evol.\u003c/em\u003e \u003cb\u003e11\u003c/b\u003e, 942\u0026ndash;954 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWurtsbaugh, W. A., Paerl, H. W. \u0026amp; Dodds, W. K. Nutrients, eutrophication and harmful algal blooms along the freshwater to marine continuum. \u003cem\u003eWIREs Water\u003c/em\u003e. \u003cb\u003e6\u003c/b\u003e, e1373 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlbert, S. et al. Water quality challenges associated with industrial logging of a karst landscape: Guadalcanal, Solomon Islands. \u003cem\u003eMar. Pollut. Bull.\u003c/em\u003e \u003cb\u003e169\u003c/b\u003e, 112506 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSurchat, M. et al. Soil and Pest Management in French Polynesian Farming Systems and Drivers and Barriers for Implementation of Practices Based on Agroecological Principles. \u003cem\u003eFrontiers Sustainable Food Systems\u003c/em\u003e \u003cb\u003e5\u003c/b\u003e, (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKoshiba, S. et al. Palau\u0026rsquo;s taro fields and mangroves protect the coral reefs by trapping eroded fine sediment. \u003cem\u003eWetlands Ecol. Manage.\u003c/em\u003e \u003cb\u003e21\u003c/b\u003e, 157\u0026ndash;164 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuernier, V. et al. Leptospira diversity in animals and humans in Tahiti, French Polynesia. \u003cem\u003ePLoS Negl. Trop. Dis.\u003c/em\u003e \u003cb\u003e11\u003c/b\u003e, e0005676 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eViau, E. J. \u0026amp; Boehm, A. B. Quantitative PCR-based detection of pathogenic Leptospira in Hawai\u0026rsquo;ian coastal streams. \u003cem\u003eJ. Water Health\u003c/em\u003e. \u003cb\u003e9\u003c/b\u003e, 637\u0026ndash;646 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStrauch, A. M. Relative change in stream discharge from a tropical watershed improves predictions of fecal bacteria in near-shore environments. \u003cem\u003eHydrol. Sci. J.\u003c/em\u003e \u003cb\u003e62\u003c/b\u003e, 1381\u0026ndash;1393 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eComeros-Raynal, M. T. et al. Catchment to sea connection: Impacts of terrestrial run-off on benthic ecosystems in American Samoa. \u003cem\u003eMar. Pollut. Bull.\u003c/em\u003e \u003cb\u003e169\u003c/b\u003e, 112530 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDadhich, A. P. \u0026amp; Nadaoka, K. Analysis of Terrestrial Discharge from Agricultural Watersheds and Its Impact on Nearshore and Offshore Reefs in Fiji. \u003cem\u003ecoas\u003c/em\u003e 28, 1225\u0026ndash;1235 (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHawaii Department of Health. Amendment and Compilation of Chap. 11\u0026ndash;54 Hawaii Administrative Rules. (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWenger, A. S. et al. Predicting the impact of logging activities on soil erosion and water quality in steep, forested tropical islands. \u003cem\u003eEnviron. Res. Lett.\u003c/em\u003e \u003cb\u003e13\u003c/b\u003e, 044035 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBoutillier, S. \u0026amp; Duane, T. \u003cem\u003eLand Use Planning to Promote Marine Conservation of Coral Reef Ecosystems in Moorea, French Polynesia\u003c/em\u003e. 28 (2006). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://escholarship.org/content/qt10f3q5p4/qt10f3q5p4.pdf\u003c/span\u003e\u003cspan address=\"https://escholarship.org/content/qt10f3q5p4/qt10f3q5p4.pdf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLoiseau, C., Thiault, L., Devillers, R. \u0026amp; Claudet, J. Cumulative impact assessments highlight the benefits of integrating land-based management with marine spatial planning. \u003cem\u003eSci. Total Environ.\u003c/em\u003e \u003cb\u003e787\u003c/b\u003e, 147339 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAdam, T. C. et al. Landscape-scale patterns of nutrient enrichment in a coral reef ecosystem: implications for coral to algae phase shifts. \u003cem\u003eEcol. Appl.\u003c/em\u003e \u003cb\u003e31\u003c/b\u003e, e2227 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoorea Coral Reef, L. T. E. R., Adam, C., Neumann, K., Maritorena, S. \u0026amp; Burkepile, D. E. MCR LTER: Coral Reef: A 2018 land cover map of Moorea, French Polynesia. Environmental Data Initiative https://doi.org/DOI PENDING (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHouk, P. et al. Nutrient thresholds to protect water quality and coral reefs. \u003cem\u003eMar. Pollut. Bull.\u003c/em\u003e \u003cb\u003e159\u003c/b\u003e, 111451 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHouk, P. et al. Nutrient thresholds to protect water quality, coral reefs, and nearshore fisheries. \u003cem\u003eMar. Pollut. Bull.\u003c/em\u003e \u003cb\u003e184\u003c/b\u003e, 114144 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAguilera, R. \u0026amp; Melack, J. M. Relationships Among Nutrient and Sediment Fluxes, Hydrological Variability, Fire, and Land Cover in Coastal California Catchments. \u003cem\u003eJ. Geophys. Research: Biogeosciences\u003c/em\u003e. \u003cb\u003e123\u003c/b\u003e, 2568\u0026ndash;2589 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHoover, D. J. \u0026amp; Mackenzie, F. T. Fluvial Fluxes of Water, Suspended Particulate Matter, and Nutrients and Potential Impacts on Tropical Coastal Water Biogeochemistry: Oahu, Hawai\u0026lsquo;i. \u003cem\u003eAquat. Geochem.\u003c/em\u003e \u003cb\u003e15\u003c/b\u003e, 547\u0026ndash;570 (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHoover, D. Fluvial nitrogen and phosphorus in Hawaii: Storm runoff, land use, and impacts on coastal waters - ProQuest. (University of Hawai\u0026rsquo;i at Manoa, (2002).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRyder, M. \u0026amp; Fares, A. Evaluating Cover Crops (Sudex, Sunn Hemp, Oats) for Use as Vegetative Filters to Control Sediment and Nutrient Loading From Agricultural Runoff in a Hawaiian Watershed. \u003cem\u003eJAWRA J. Am. Water Resour. Association\u003c/em\u003e. \u003cb\u003e44\u003c/b\u003e, 640\u0026ndash;653 (2008).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuinto, D. F. \u0026amp; Inciong, M. M. Soil quality, management practices and sustainability of pineapple farms in Cavite, Philippines: Part 1. Soil quality. \u003cem\u003eJ. South. Pac. Agric.\u003c/em\u003e \u003cb\u003e16\u003c/b\u003e, 30\u0026ndash;41 (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMillbank, H. \u0026amp; Nothard, B. \u003cem\u003eUnderstanding the Economics of Horticultural Management Practices and Systems for Improving Water Quality Runoff in the Great Barrier Reef Catchment Areas\u003c/em\u003e. 62 (2023). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://era.dpi.qld.gov.au/id/eprint/9271/\u003c/span\u003e\u003cspan address=\"https://era.dpi.qld.gov.au/id/eprint/9271/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNorton, J. \u0026amp; Ouyang, Y. Controls and Adaptive Management of Nitrification in Agricultural Soils. \u003cem\u003eFront Microbiol\u003c/em\u003e \u003cb\u003e10\u003c/b\u003e, (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFeld, C. K. et al. Chapter Three - From Natural to Degraded Rivers and Back Again: A Test of Restoration Ecology Theory and Practice. \u003cem\u003eAdv. Ecol. Res.\u003c/em\u003e \u003cb\u003e44\u003c/b\u003e, 119\u0026ndash;209 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKnee, K. L., Crook, E. D., Hench, J. L., Leichter, J. J. \u0026amp; Paytan, A. Assessment of Submarine Groundwater Discharge (SGD) as a Source of Dissolved Radium and Nutrients to Moorea (French Polynesia) Coastal Waters. \u003cem\u003eEstuaries Coasts\u003c/em\u003e. \u003cb\u003e39\u003c/b\u003e, 1651\u0026ndash;1668 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHa\u0026szlig;ler, K. et al. Provenance of nutrients in submarine fresh groundwater discharge on Tahiti and Moorea, French Polynesia. \u003cem\u003eAppl. Geochem.\u003c/em\u003e \u003cb\u003e100\u003c/b\u003e, 181\u0026ndash;189 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBoano, F. et al. Hyporheic flow and transport processes: Mechanisms, models, and biogeochemical implications. \u003cem\u003eRev. Geophys.\u003c/em\u003e \u003cb\u003e52\u003c/b\u003e, 603\u0026ndash;679 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShuler, C. K. et al. Understanding surface water\u0026ndash;groundwater interaction, submarine groundwater discharge, and associated nutrient loading in a small tropical island watershed. \u003cem\u003eJ. Hydrol.\u003c/em\u003e \u003cb\u003e585\u003c/b\u003e, 124342 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBartholomew, D. P., Rohrbach, K. G. \u0026amp; Evans, D. O. \u003cem\u003ePineapple Cultivation in Hawaii\u003c/em\u003e. 1\u0026ndash;8 (2002). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://hdl.handle.net/10125/2444\u003c/span\u003e\u003cspan address=\"http://hdl.handle.net/10125/2444\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRechberger, M. V., Zehetner, F. \u0026amp; Gerzabek, M. H. Phosphate sorption-desorption properties in volcanic topsoils along a chronosequence and a climatic gradient on the Gal\u0026aacute;pagos Islands. \u003cem\u003eJ. Plant Nutr. Soil Sci.\u003c/em\u003e \u003cb\u003e184\u003c/b\u003e, 479\u0026ndash;491 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ede Campos, M., Antonangelo, J. A., van der Zee, S. E., A. T., M. \u0026amp; Alleoni, L. R. F. Degree of phosphate saturation in highly weathered tropical soils. \u003cem\u003eAgric. Water Manage.\u003c/em\u003e \u003cb\u003e206\u003c/b\u003e, 135\u0026ndash;146 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJamet, R. Les sols de Moorea et des \u0026icirc;les Sous-le-Vent [cartographic material]: archipel de la Soci\u0026eacute;t\u0026eacute;, Polyn\u0026eacute;sie Fran\u0026ccedil;aise / par R\u0026eacute;mi Jamet. \u003cem\u003eLes sols de Moorea et des \u0026icirc;les Sous-\u0026hellip; Items | National Library of New Zealand | National Library of New Zealand\u003c/em\u003e https://natlib.govt.nz/records/20850230 (2000).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAhearn, D. S. et al. Land use and land cover influence on water quality in the last free-flowing river draining the western Sierra Nevada, California. \u003cem\u003eJ. Hydrol.\u003c/em\u003e \u003cb\u003e313\u003c/b\u003e, 234\u0026ndash;247 (2005).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArnold, J. G. et al. Model Use, Calibration, and Validation. \u003cem\u003eTrans. ASABE\u003c/em\u003e. \u003cb\u003e55\u003c/b\u003e, 1491\u0026ndash;1508 (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRenard, K. G., Foster, G. R., Yoder, D. C. \u0026amp; McCool, D. K. RUSLE revisited: Status, questions, answers, and the future. \u003cem\u003eJ. Soil Water Conserv.\u003c/em\u003e \u003cb\u003e49\u003c/b\u003e, 213\u0026ndash;220 (1994).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDas, S., Deb, P., Bora, P. K. \u0026amp; Katre, P. Comparison of RUSLE and MMF Soil Loss Models and Evaluation of Catchment Scale Best Management Practices for a Mountainous Watershed in India. \u003cem\u003eSustainability\u003c/em\u003e \u003cb\u003e13\u003c/b\u003e, 232 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWymore, A. S., Leon, M. C., Shanley, J. B. \u0026amp; McDowell, W. H. Hysteretic Response of Solutes and Turbidity at the Event Scale Across Forested Tropical Montane Watersheds. \u003cem\u003eFront Earth Sci\u003c/em\u003e \u003cb\u003e7\u003c/b\u003e, (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAugustijn, D., Fares, A. \u0026amp; Tran, D. N. Temporal and Spatial Variations in Total Suspended and Dis- solved Solids in the Upper Part of Manoa Stream, Hawaii. \u003cem\u003eJournal Sustainable Watershed Science Management\u003c/em\u003e. \u003cb\u003e1\u003c/b\u003e, 1\u0026ndash;9 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRam, A. R. \u0026amp; Terry, J. P. Stream turbidity responses to storm events in a pristine rainforest watershed on the Coral Coast of southern Fiji. \u003cem\u003eInt. J. Sedim. Res.\u003c/em\u003e \u003cb\u003e31\u003c/b\u003e, 279\u0026ndash;290 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWotling, G. \u0026amp; Bouvier, C. Impact of urbanization on suspended sediment and organic matter fluxes from small catchments in Tahiti. \u003cem\u003eHydrol. Process.\u003c/em\u003e \u003cb\u003e16\u003c/b\u003e, 1745\u0026ndash;1756 (2002).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSutherland, R. A. Rolled erosion control systems for hillslope surface protection: a critical review, synthesis and analysis of available data. I. Background and formative years. \u003cem\u003eLand Degradation Development\u003c/em\u003e. \u003cb\u003e9\u003c/b\u003e, 465\u0026ndash;486 (1998).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSutherland, R. A. \u0026amp; Ziegler, A. D. Hillslope runoff and erosion as affected by rolled erosion control systems: a field study. \u003cem\u003eHydrol. Process.\u003c/em\u003e \u003cb\u003e20\u003c/b\u003e, 2839\u0026ndash;2855 (2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWemple, B. C. et al. Ecohydrological disturbances associated with roads: Current knowledge, research needs, and management concerns with reference to the tropics. \u003cem\u003eEcohydrology\u003c/em\u003e \u003cb\u003e11\u003c/b\u003e, e1881 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMartinez, C., Menj\u0026iacute;var, J. C. \u0026amp; Saavedra, R. Soils erosion in pineapple (Ananas comosus L. Merr) producing areas. \u003cem\u003eRevista de Ciencias Agr\u0026iacute;colas\u003c/em\u003e. \u003cb\u003e39\u003c/b\u003e, 142\u0026ndash;154 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCarlson, R. R., Foo, S. A. \u0026amp; Asner, G. P. Land Use Impacts on Coral Reef Health: A Ridge-to-Reef Perspective. \u003cem\u003eFrontiers Mar. Science\u003c/em\u003e \u003cb\u003e6\u003c/b\u003e, (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCollin, A. et al. High resolution topobathymetry using a Pleiades-1 triplet: Moorea Island in 3D. \u003cem\u003eRemote Sens. Environ.\u003c/em\u003e \u003cb\u003e208\u003c/b\u003e, 109\u0026ndash;119 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMeyer, J. Y., Pouteau, R., Spotswood, E., Taputuarai, R. \u0026amp; Fourdrigniez, M. The importance of novel and hybrid habitats for plant conservation on islands: a case study from Moorea (South Pacific). \u003cem\u003eBiodivers. Conserv.\u003c/em\u003e \u003cb\u003e24\u003c/b\u003e, 83\u0026ndash;101 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKirch, P. V. \u003cem\u003eOn the Road of the Winds: An Archaeological History of the Pacific Islands before European Contact, Revised and Expanded Edition\u003c/em\u003e (University of California Press, 2017). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1525/9780520968899\u003c/span\u003e\u003cspan address=\"10.1525/9780520968899\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eInstitut de la. statistique de la Polyn\u0026eacute;sie Fran\u0026ccedil;aise. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ispf.pf/\u003c/span\u003e\u003cspan address=\"https://www.ispf.pf/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStrahler, A. Hypsometric (Area-Altitude) Analysis of Erosional Topography. \u003cem\u003eGSA Bull.\u003c/em\u003e \u003cb\u003e63\u003c/b\u003e, 1117\u0026ndash;1142 (1952).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBinet, T. \u0026amp; Gonnot, C. Etude de l\u0026rsquo;\u0026eacute;rosion sur des plantations d\u0026rsquo;ananas \u0026agrave; Moorea. (2005).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTaylor, B. W. et al. Improving the fluorometric ammonium method: matrix effects, background fluorescence, and standard additions. \u003cem\u003eJ. North. Am. Benthological Soc.\u003c/em\u003e \u003cb\u003e26\u003c/b\u003e, 167\u0026ndash;177 (2007).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRodier, M. et al. Diversity and toxic potential of algal bloom-forming species from Takaroa lagoon (Tuamotu, French Polynesia): a field and mesocosm study. \u003cem\u003eAquat. Microb. Ecol.\u003c/em\u003e \u003cb\u003e83\u003c/b\u003e, 15\u0026ndash;34 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFarr, T. G. et al. The Shuttle Radar Topography Mission. \u003cem\u003eReviews Geophysics\u003c/em\u003e \u003cb\u003e45\u003c/b\u003e, (2007).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOksanen, J. et al. vegan: Community Ecology Package. (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eR Core Team. R: A Language and Environment for Statistical Computing. (2022).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Global change, Hydrology, Land use, Precipitation, River chemistry","lastPublishedDoi":"10.21203/rs.3.rs-6247948/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6247948/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eHuman activities drive changes in freshwater ecosystems by altering biogeochemical cycles. On high volcanic tropical islands, human activities can be compartmentalized by steep terrain that delineates watershed boundaries. Patterns of human activities, such as land use, affect adjacent stream ecosystems through runoff of sediment and nutrients, which varies seasonally in the tropics as a result of seasonal rainfall. Here, we sought to reveal human impacts on the nutrient and sediment regimes of tropical rivers by tracking patterns of river chemistry across a series of watersheds on Moorea, French Polynesia, between 2018 and 2019. Repeated sampling of rivers across a gradient of human activities revealed that water chemistry varied seasonally and with respect to rainfall and land use. In particular, dissolved inorganic nitrogen was more concentrated in rivers of watersheds with higher rates of land clearing. Additionally, total suspended solids and phosphate were higher when recent rainfall was high. Our results show that human activities can have a substantial impact on the amounts of nutrients and sediment that tropical rivers transport, which on tropical islands could facilitate movement of materials from land to sea as precipitation increases with intensifying climate change.\u003c/p\u003e","manuscriptTitle":"Land use shapes riverine nutrient and sediment concentrations on Moorea, French Polynesia","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-04-21 03:40:16","doi":"10.21203/rs.3.rs-6247948/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-06-16T15:48:05+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-06-13T14:36:15+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-06-10T12:35:32+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"102357032098669585271019917218480041864","date":"2025-05-26T23:13:28+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"166093764348759112167888358593202903193","date":"2025-05-26T10:55:17+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"210083900951729945755332988145446078438","date":"2025-05-26T07:08:53+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-05-24T07:36:03+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-05-24T07:34:58+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-04-24T17:02:15+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-04-05T17:59:17+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-04-05T17:58:10+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"8c908a24-e288-452a-945a-8ca00835a689","owner":[],"postedDate":"April 21st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":46730227,"name":"Biological sciences/Ecology/Biogeochemistry"},{"id":46730228,"name":"Biological sciences/Ecology/Freshwater ecology"},{"id":46730229,"name":"Earth and environmental sciences/Hydrology"},{"id":46730230,"name":"Earth and environmental sciences/Environmental sciences/Environmental chemistry"},{"id":46730231,"name":"Earth and environmental sciences/Environmental sciences/Environmental impact"}],"tags":[],"updatedAt":"2025-08-04T16:43:21+00:00","versionOfRecord":{"articleIdentity":"rs-6247948","link":"https://doi.org/10.1038/s41598-025-13425-1","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2025-07-31 16:13:20","publishedOnDateReadable":"July 31st, 2025"},"versionCreatedAt":"2025-04-21 03:40:16","video":"","vorDoi":"10.1038/s41598-025-13425-1","vorDoiUrl":"https://doi.org/10.1038/s41598-025-13425-1","workflowStages":[]},"version":"v1","identity":"rs-6247948","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6247948","identity":"rs-6247948","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}