Temporal and spatial trends in nitrogen and phosphorus inputs to the watershed of the Altamaha River,Georgia, USA

The watershed of the Altamaha River, Georgia, is one of the largest in the southeastern U.S., draining 36,718 km² (including parts of metro Atlanta). We calculated both nitrogen (fertilizer, net food and feed import, atmospheric deposition, and biological N fixation in agricultural and forest lands) and phosphorus (fertilizer and net food and feed import) inputs to the watershed for 6 time points between 1954 and 2002. Total nitrogen inputs rose from 1,952 kg N km⁻² yr⁻¹ in 1954 to a peak of 3,593 kg N km⁻² yr⁻¹ in 1982 and then declined to 2,582 kg N km⁻² yr⁻¹ by 2002. Phosphorus inputs rose from 409 kg P km⁻² yr⁻¹ in 1954 to 532 kg P km⁻² yr⁻¹ in 1974 before declining to 412 kg P km⁻² yr⁻¹ in 2002. Fertilizer tended to be the most important input of both N and P to the watershed, although net food and feed import increased in importance over time and was the dominant source of N input by 2002. When considered on an individual basis, fertilizer input tended to be highest in the middle portions of the watershed (Little and Lower Ocmulgee and Lower Oconee sub-watersheds) whereas net food and feed imports were highest in the upper reaches (Upper Oconee and Upper Ocmulgee sub-watersheds). Although the overall trend in recent years has been towards decreases in both N and P inputs, these trends may be offset due to continuing increases in animal and human populations.     

Temperature controls a latitudinal gradient in the proportion of watershed nitrogen exported to coastal ecosystems

Increased export of biologically available nitrogen (N) to the coastal zone is strongly linked to eutrophication, which is a major problem in coastal marine ecosystems (NRC (2000) Clean Coastal Waters: Understanding and Reducing the Effects of Nutrient Pollution. National Academy Press, Washington, DC; Bricker et al. (1999) National Estuarine Eutrophication Assessment. Effects of nutrient enrichment in the nation’s estuaries. NOAA-NOS Special Projects Office, Silver Spring, MD). However, not all of the nitrogen input to a watershed is exported to the coast (Howarth et al. (1996) Biogeochemistry 35:75–139; Jordan and Weller (1996) Bioscience 46:655–664). Global estimates of nitrogen export to coasts have been taken to be 25% of watershed input, based largely on northeastern U.S. observations (Galloway et al. (2004) Biogeochemistry 70:153–226; Boyer et al. (2006) Global Biogeochem Cycle 20:Art. No. GB1S91). We applied the N budgeting methodology developed for the International SCOPE Nitrogen project (Howarth et al. (1996) Biogeochemistry 35:75–139; Boyer et al. (2002) Biogeochemistry 57:137–169) to 12 watersheds in the southeastern U.S., and compared them with estimates of N export for 16 watersheds in the northeastern U.S. (Boyer et al. (2002) Biogeochemistry 57:137–169). In southeastern watersheds, average N export was only 9% of input, suggesting the need for downward revision of global estimates. The difference between northern and southern watersheds is not a function of the absolute value of N inputs, which spanned a comparable range and were positively related to export in both cases. Rather, the proportion of N exported was significantly related to average watershed temperature (% N export = 58.41 e^{−0.11 * temperature}; R ² = 0.76), with lower proportionate nitrogen export in warmer watersheds. In addition, we identified a threshold in proportionate N export at 38°N latitude that corresponds to a reported breakpoint in the rate of denitrification at 10–12°C. We hypothesize that temperature, by regulating denitrification, results in increased proportionate N export at higher latitudes. Regardless of the mechanism, these observations suggest that temperature increases associated with future climate change may well reduce the amount of nitrogen that reaches estuaries, which will have implications for coastal eutrophication.     

The influence of climate on average nitrogen export from large watersheds in the Northeastern United States

The flux of nitrogen in large rivers in North America and Europe is well explained as a function of the net anthropogenic inputs of nitrogen to the landscape, with on average 20 to 25% of these inputs exported in rivers and 75 to 80% of the nitrogen retained or denitrified in the landscape. Here, we use data for average riverine nitrogen fluxes and anthropogenic inputs of nitrogen over a 6-year period (1988–1993) for 16 major watersheds in the northeastern United States to examine if there is also a climatic influence on nitrogen fluxes in rivers. Previous studies have shown that for any given river, nitrogen fluxes are greater in years with higher discharge, but this can be interpreted as storage of nitrogen in the landscape during dry years and flushing of this stored nitrogen during wet years. Our analyses demonstrate that there is also a longer-term steady-state influence of climate on riverine nitrogen fluxes. Those watersheds that have higher precipitation and higher discharge export a greater fraction of the net anthropogenic inputs of nitrogen. This fractional export ranges from 10 to 15% of the nitrogen inputs in drier watersheds in the northeastern United States to over 35% in the wetter watersheds. We believe this is driven by lower rates of denitrification in the wetter watersheds, perhaps because shorter water residence times do not allow for as much denitrification in riparian wetlands and low-order streams. Using mean projections for the consequences of future climate change on precipitation and discharge, we estimate that nitrogen fluxes in the Susquehanna River to Chesapeake Bay may increase by 3 to 17% by 2030 and by 16 to 65% by 2095 due to greater fractional delivery of net anthropogenic nitrogen inputs as precipitation and discharge increase. Although these projections are highly uncertain, they suggest a need to better consider the influence of climate on riverine nitrogen fluxes as part of management efforts to control coastal nitrogen pollution.     

Influences of climate,hydrology, and land use on input and export of nitrogen in California watersheds

Human activities have greatly increased the input of biologically available nitrogen (N) from land-based sources to aquatic ecosystems; yet few studies have examined how human actions influence N export in regions with a strong seasonality in water availability. In this study, we quantified N inputs and outputs for 23 California watersheds and examined how climate, hydrology, and land use practices influenced watershed N export. N inputs ranged from 581 to 11,234 kg N km⁻² year⁻¹ among watersheds, with 80% of total input for the region originating from agriculture (inorganic fertilizer, manure, and legumes). Of the potential N sources examined, mean annual concentrations of dissolved organic N and dissolved inorganic N in study rivers correlated most strongly with manure N input (r ² = 0.54 and 0.53, respectively). Seasonal N export varied by basin and was correlated with climate, anthropogenic N inputs, and reservoir releases. Fractional export of watershed N inputs by study rivers annually was small (median of 8%) and scaled exponentially with runoff (r = 0.66). Collectively, our results show that anthropogenic activities have altered both the magnitude and timing of watershed N export in California and suggest that targeted management in specific locations and times of the year could reduce N export to downstream systems in the region.     

Total inputs of phosphorus and nitrogen by wet deposition into Lake Taihu, China

Lake Taihu suffers from eutrophication caused by riverine nutrient inputs and air deposition. To characterize wet deposition of phosphorus (P) and nitrogen (N) to the lake, precipitation collection and measurements of total phosphorus (TP) and total nitrogen (TN) and other components at five cities around Lake Taihu were made from July 2002 to June 2003. TP and TN concentrations and deposition rates exhibited strong spatial variation in the whole catchment. An inverse correlation between station-averaged TP and TN concentrations and precipitation amount was found. Maximal TP concentration in rainfall was found in Suzhou, and maximal TN in Wuxi. However, highest wet deposition rates of TP and TN were found in Suzhou, which suggests that atmospheric nutrients are mostly from the east and northwest area of Lake Taihu. Mean TP and TN deposition rates were 0.03 and 2.0 t km⁻² year⁻¹ respectively in Lake Taihu, which are greater than reported values in other areas by comparision. Total N and P contributed to the lake by wet deposition were 75 and 4720 t per year, respectively, which represent about 7.3% and 16.5% of total annual N and P inputs via inflow rivers. Wet deposition, especially N, could have significant effects on eutrophication in the lake, which shows that air deposition should be taken into account while reducing the external nutrients in the lake.     

Net N input and riverine N export from Illinois agricultural watersheds with and without extensive tile drainage

Some of the largest riverine N fluxes in the continental USA have been observed in agricultural regions with extensive artificial subsurface drainage, commonly called tile drainage. The degree to which high riverine N fluxes in these settings are due to high net N inputs (NNI), greater transport efficiency caused by the drainage systems, or other factors is not known. The objective of this study was to evaluate the role of tile drainage by comparing NNI and riverine N fluxes in regions of Illinois with similar climate and crop production practices but with different intensities of tile drainage. Annual values of NNI between 1940 and 1999 were estimated from county level agricultural production statistics and census estimates of human population. During 1945–1961, riverine nitrate flux in the extensively tile drained region averaged 6.6kgNha^–1year^–1 compared to 1.3 to 3.1kgNha^–1 for the non-tile drained region, even though NNI was greater in the non-tile drained region. During 1977–1997, NNI to the tile-drained region had increased to 27kgNha^–1year^–1 and riverine N flux was approximately 100% of this value. In the non-tile-drained region, NNI was approximately 23kgNha^–1year^–1 and riverine N flux was between 25% and 37% of this value (5 to 9kgNha^–1year^–1). Denitrification is not included in NNI and, therefore, any denitrification losses from tile-drained watersheds must be balanced by other N sources, such as depletion of soil organic N or underestimation of biological N fixation. If denitrification and depletion of soil organic N are significant in these basins, marginal reductions in NNI may have little influence on riverine N flux. If tile drained cropland in Illinois is representative of the estimated 11 million ha of tile drained cropland throughout the Mississippi River Basin, this 16% of the drainage area contributed approximately 30% of the increased nitrate N flux in the Lower Mississippi River that occurred between 1955 and the 1990s.     

Assessment of nitrogen flows into the Cuban landscape

The alteration of the nitrogen (N) cycle by human activities is widespread and has often resulted in increased flows of nitrogen to the marine environment. In this paper we have attempted to know the changes of N fluxes in Cuba by quantifying the N inputs to the landscape from (1) fertilizer applications, (2) atmospheric deposition, (3) biological nitrogen fixation and (4) net import of food and feeds. N-inputs to the country progressively increased until the end of the 20th century, reaching a peak during the 80s when low cost fertilizer imported from the former Soviet Union led to heavy rates of application. This rapid growth represented more than a 5-fold increase with respect to pristine values; higher than the two-fold global increase of anthropogenic N reported by Vitousek et al. (1997 Human alteration of the global nitrogen cycle: sources and consequences. Ecol. Appl. 7:737–750). Inorganic fertilizer was the largest single source of reactive N, followed by atmospheric deposition, biological fixation, and net imports of foods and feedstocks. Nitrogen inputs peaked in 1987 and data expressed on an area basis show that N flux to the Cuban landscape, in the 80s, was one of the highest reported in the literature. During the 90s, there was a dramatic drop in nitrogen inputs mainly associated to a decrease in the use of inorganic fertilizer. Other factors reducing nutrient inflows to Cuba, during the same period, were imports of foodstuff and livestock feeds, a decrease of nitrogen oxide emissions, and a decrease in the sugar cane crop area. Using an empirical relationship (Howarth et al. 1996 Regional nitrogen budgets and riverine N & O fluxes for the drainages to the North Atlantic Ocean: Natural and human influences. Biogeochemistry 35:75–139) we present a very preliminary estimate of N-inputs to coastal waters and discuss the consequences of these changes on the coastal zone.     

Proton production from nitrogen transformation drives stream export of base cations in acid-sensitive forested watersheds

The biogeochemical cycles of nitrogen (N) and base cations (BCs), (i.e., K⁺, Na⁺, Ca²⁺, and Mg²⁺), play critical roles in plant nutrition and ecosystem function. Empirical correlations between large experimental N fertilizer additions to forest ecosystems and increased BCs loss in stream water are well demonstrated, but the mechanisms driving this coupling remain poorly understood. We hypothesized that protons generated through N transformation (PPR_N)—quantified as the balance of NH₄⁺ (H⁺ source) and NO₃⁻ (H⁺ sink) in precipitation versus the stream output will impact BCs loss in acid-sensitive ecosystems. To test this hypothesis, we monitored precipitation input and stream export of inorganic N and BCs for three years in an acid-sensitive forested watershed in a granite area of subtropical China. We found the precipitation input of inorganic N (17.71 kg N ha⁻¹ year⁻¹ with 54% as NH₄⁺–N) was considerably higher than stream exported inorganic N (5.99 kg N ha⁻¹ year⁻¹ with 83% as NO₃⁻–N), making the watershed a net N sink. The stream export of BCs (151, 1518, 851, and 252 mol ha⁻¹ year⁻¹ for K⁺, Na⁺, Ca²⁺, and Mg²⁺, respectively) was positively correlated (r = 0.80, 0.90, 0.84, and 0.84 for K⁺, Na⁺, Ca²⁺, and Mg²⁺ on a monthly scale, respectively, P < 0.001, n = 36) with PPR_N (389 mol ha⁻¹ year⁻¹) over the three years, suggesting that PPR_N drives loss of BCs in the acid-sensitive ecosystem. A global meta-analysis of 15 watershed studies from non-calcareous ecosystems further supports this hypothesis by showing a similarly strong correlation between ∑BCs output and PPR_N (r = 0.89, P < 0.001, n = 15), in spite of the pronounced differences in environmental settings. Collectively, our results suggest that N transformations rather than anions (NO₃⁻ and/or SO₄²⁻) leaching specifically, are an important mediator of BCs loss in acid-senstive ecosystems. Our study provides the first definitive evidence that the chronic N deposition and subsequent transformation within the watershed drive stream export of BCs through proton production in acid-sensitive ecosystems, irrespective of their current relatively high N retention. Our findings suggest the N-transformation-based proton production can be used as an indicator of watershed outflow quality in the acid-sensitive ecosystems.     

Watershed modelling of nonpoint nitrogen losses from arable land to the Swedish coast in 1985 and 1994

Eutrophication problems in the Baltic Sea have drawn attention to the contribution of nutrients from surrounding countries. By using the HBV-N model in southern Sweden (145 000 km²) daily nitrogen leaching, reduction in rivers and lakes, net transport to the sea and source apportionment have been calculated in 3725 subbasins for the period 1985–1994, with calibration at 722 sites against measured time series. On average, 48% of the nonpoint losses from agriculture were reduced during the transport towards the sea, which left about 33 500 tonnes in annual mean net transport. This represents 45% of the total land-based load. Land cover and emissions for the years of 1985 and 1994 were used in two separate simulations of the 10-year period. The normalized gross leakage from arable land in 1985 was estimated to 29 kg N ha⁻¹ year⁻¹, which corresponds to 15 kg N ha⁻¹ year⁻¹ in net leakage to the sea. In 1994 these transports were reduced by 20 and 15%, and thereby the total load on the sea was decreased by 7%. This is still far from the Swedish goal of 50% reduction. The article presents the spatial variation of nitrogen leakage and retention within the southern half of Sweden, and emphasizes the importance of allocating measures where down-stream retention is low in order to achieve efficiency with respect to the sea. It is shown that the model approach may be used in the decision making process for best management practices in watersheds.     

Recruitment prediction for Pacific herring (Clupea pallasi) on the west coast of Vancouver Island, Canada

The accurate prediction of recruitment to the fishery is a very important tool within the management structure of any fish stock being exploited. In the case of the Pacific herring, Clupea pallasi, fishery in Canada, a forecast of the abundance of each herring stock is particularly important for formulating an annual catch quota. The sustainable management of the fishery and the resource is based in part on accurate recruitment forecasting because Pacific herring are short-lived and so the recruitment contributes a significant part of the total spawning run targeted by the fishery each year. Several factors are believed be important in determining the success of recruitment besides spawners biomass. Since herrings are “r” strategists, conditions related to the egg, the planktonic, or even the juvenile stage might determine the future level of recruitment. Recently a formula that defines conditions for a semi-quantitative level of recruitment forecast was elaborated using genetic algorithms and current study attempts to improve on this model. Using salinity in two quarterly periods during the planktonic and pre-recruit stages, temperature and spawning biomass for the west coast of Vancouver Island stock, classification rules that define recruitment in 3 different levels (low, medium and high) were developed with a genetic algorithm, setting low and high boundaries for each condition. A 75% success in classifying recruitment was obtained. The model was shown to be particularly effective at predicting when the recruitment would be low, which could be important from the perspective of the Precautionary Approach and the sustainable management of this stock.     

Phosphorus deficiency induced by nitrogen input in Douglas fir in the Netherlands

Summary A re-examination of earlier NPK fertilization experiments in Douglas fir stands on sandy soils shows the effects of high nitrogen input by air pollution during the last 10–15 years on plant nutrition at these sites. In 1960, experimental plots showed a positive growth reaction to nitrogen, phosphorus, and potassium fertilization. All suffered from severe phosphorus deficiency in 1984, low phosphorus in the needles was invariably accompanied by a high nitrogen content, with all N/P ratios between 20 and 30. The same conclusion emerges from an independent investigation of nutrient status of a selection of Douglas fir stands. Hence, if stand productivity and a balanced nutrient status of the trees is to be maintained, the increase in atmospheric input of nitrogen calls for supplementary fertilization. Given the current N/P ratios in the needles, a positive growth response to phosphorus fertilization is to be expected.     

海南岛西海岸红树林软体动物多样性

为了掌握海南岛西海岸红树林软体动物多样性状况, 本文于2015-2016年对海南岛西海岸6个主要红树林分布区软体动物的种类组成、物种多样性和群落结构的时空变化进行了研究。共采集到软体动物57科201种, 主要优势种有珠带拟蟹守螺(Cerithidea cingulata)、绯拟沼螺(Assiminea latericea)、圆胖樱蛤(Pinguitellina cycladiformis)和斜肋齿蜷(Sermyla riqueti)等。年平均密度与平均生物量分别为324.13 ind./m²和142.88 g/m², 物种数、密度和生物量均高于北部湾沿岸的其他红树林。海南岛西海岸各红树林软体动物的密度和生物量均未呈现明显的季节变化, 软体动物密度存在空间上的显著差异。聚类分析结果显示, 在盐度、土壤结构和红树林群落存在差异的不同红树林, 软体动物种类组成相似度低。多因子AZTI海洋生物指数(M-AMBI)结果表明海南岛西海岸红树林软体动物群落未受明显扰动, 生态环境质量介于一般和优良之间, 但仍受到人类活动与外来植物入侵威胁, 因其在候鸟迁徙中的重要作用, 亟需加强管理与保护。     

Importance of sediment deposition and denitrification for nutrient retention in floodplain wetlands

Questions: Various floodplain communities may differ in their relative abilities to influence water quality through nutrient retention and denitrification. Our main questions were: (1) what is the importance of sediment deposition and denitrification for plant productivity and nutrient retention in floodplains; (2) will rehabilitation of natural floodplain communities (semi‐natural grassland, reedbed, woodland, pond) from agricultural grassland affect nutrient retention? Location: Floodplains of two Rhine distributaries (rivers Ussel and Waal), The Netherlands. Methods: Net sedimentation was measured using mats, denitrification in soil cores by acetylene inhibition and bio‐mass production by clipping above‐ground vegetation in winter and summer. Results: Sediment deposition was a major source of N and P in all floodplain communities. Highest deposition rates were found where water velocity was reduced by vegetation structure (reedbeds) or by a drop in surface elevation (pond). Sediment deposition was not higher in woodlands than in grassland types. Denitrification rates were low in winter but significantly higher in summer. Highest denitrification rates were found in an agricultural grassland (winter and summer) and in the ponds (summer). Plant productivity and nutrient uptake were high in reedbeds, intermediate in agricultural grasslands, ponds and semi‐natural grasslands and very low in woodlands (only understorey). All wetlands were N‐limited, which could be explained by low N:P ratios in sediment. Conclusions: Considering Rhine water quality: only substantial P‐retention is expected because, relative to the annual nutrient loads in the river, the floodplains are important sinks for P, but much less for N. Rehabilitation of agricultural grasslands into ponds or reedbeds will probably be more beneficial for downstream water quality (lower P‐concentrations) than into woodlands or semi‐natural grasslands.     

Ecological interaction between sympatric Mytilus species on the west coast of Canada investigated using PCR markers

M. californianus is the dominant marine mussel species on exposed rocky shores, while M. trossulus is usually the dominant mussel species in more sheltered waters on the west coast of North America. Since these species are physically indistinguishable when small (< 10.0 mm), we developed two polymerase chain reaction (PCR) -based markers to discriminate between them. Using these markers, we identified mussels taken from an exposed coast ( n = 114), a sheltered harbour ( n = 80), and an upper-intertidal pool ( n = 42) on the west coast of Vancouver Island, British Columbia, Canada. M. californianus were found only on the open coast. Small M. trossulus (< 20.0 mm) were common to all three sample sites, but were extremely rare at larger sizes (> 20.0 mm) on the open coast. Our results indicate that M. californianus are excluded from sheltered waters via early life factors, while M. trossulus are excluded from the open coast due to mortality later in life.     

Anthropogenic nitrogen sources and relationships to riverine nitrogen export in the northeastern U.S.A.

Human activities have greatly altered the nitrogen (N) cycle, accelerating the rate of N fixation in landscapes and delivery of N to water bodies. To examine relationships between anthropogenic N inputs and riverine N export, we constructed budgets describing N inputs and losses for 16 catchments, which encompass a range of climatic variability and are major drainages to the coast of the North Atlantic Ocean along a latitudinal profile from Maine to Virginia. Using data from the early 1990's, we quantified inputs of N to each catchment from atmospheric deposition, application of nitrogenous fertilizers, biological nitrogen fixation, and import of N in agricultural products (food and feed). We compared these inputs with N losses from the system in riverine export.The importance of the relative sources varies widely by catchment and is related to land use. Net atmospheric deposition was the largest N source (>60%) to the forested basins of northern New England (e.g. Penobscot and Kennebec); net import of N in food was the largest source of N to the more populated regions of southern New England (e.g. Charles & Blackstone); and agricultural inputs were the dominant N sources in the Mid-Atlantic region (e.g. Schuylkill & Potomac). Over the combined area of the catchments, net atmospheric deposition was the largest single source input (31%), followed by net imports of N in food and feed (25%), fixation in agricultural lands (24%), fertilizer use (15%), and fixation in forests (5%). The combined effect of fertilizer use, fixation in crop lands, and animal feed imports makes agriculture the largest overall source of N. Riverine export of N is well correlated with N inputs, but it accounts for only a fraction (25%) of the total N inputs. This work provides an understanding of the sources of N in landscapes, and highlights how human activities impact N cycling in the northeast region.     

Soil macrofauna and nitrogen on a sub-Antarctic island

Summary The densities, diets and habitat preferences of the soil macrofaunal species on sub-Antarctic Marion Island (47°S, 38°E) are described. Their role in N cycling on the island is assessed, using a mire-grassland community as an example. Primary production on the island is high and this leads to a substantial annual requirement of nutrients by the vegetation. This requirement must almost wholly be met by mineralization of nutrient reserves in the organic matter. Rates of peat nitrogen mineralization mediated by microorganisms alone are much too low to account for rates of N uptake by the vegetation. Although soil macroinvertebrates, and bacteria represent a very small fraction of the total N pool, their interaction accounts for most of the peat N mineralization, as indicated by the amounts of inorganic N released into solution in microcosms. Extrapolation of the microcosm results shows that the soil macrofauna (mainly earthworms) stimulate the release of enough N from the mire-grassland peat to account for maximum N mineralization rates calculated from temporal changes in peat inorganic N levels and plant uptake during the most active part of the growing season. Considering that large numbers of mesoand microinvertebrates occur and must also contribute to nutrient mineralization, the soil faunal component is clearly of crucial importance to nutrient cycling on Marion Island. This is probably true of all sub-Antarctic islands.     

Methane in the Changjiang (Yangtze River) Estuary and its adjacent marine area: riverine input,sediment release and atmospheric fluxes

Dissolved methane (CH₄) was measured in the waters of the Changjiang (Yangtze River) Estuary and its adjacent marine area during five surveys from 2002 to 2006. Dissolved CH₄ concentrations ranged from 2.71 to 89.2 nM and had seasonal variation with the highest values occurring in summer and lowest in autumn. The horizontal distribution of dissolved CH₄ decreased along the freshwater plume from the river mouth to the open sea. Dissolved CH₄ in surface waters of the Changjiang was observed monthly at the most downstream main channel station Xuliujing (121o2′E, 31o46′N), which ranged from 16.2 to 126.2 nM with an average of 71.6 ± 36.3 nM. The average annual input of CH₄ from the Changjiang to the Estuary and its adjacent area was estimated to be 2.24 mol s⁻¹ equal to 70.6 × 10⁶ mol year⁻¹. Mean CH₄ emission rate from the sediments of the Changjiang Estuary in spring was 1.97 μmol m⁻² day⁻¹, but it may be higher in summer due to hypoxia in the bottom waters and higher temperatures. The annual sea to air CH₄ fluxes from the Changjiang Estuary and its adjacent marine area were estimated to be 61.4 ± 22.6 and 16.0 ± 6.1 μmol m⁻² day⁻¹, respectively, using three different gas exchange models. Hence the Changjiang Estuary and its adjacent marine area are net sources of atmospheric CH₄.