By continuing to browse this site you agree to us using cookies as described in About Cookies
Notice: Wiley Online Library will be unavailable on Saturday 7th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 08.00 EDT / 13.00 BST / 17:30 IST / 20.00 SGT and Sunday 8th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 06.00 EDT / 11.00 BST / 15:30 IST / 18.00 SGT for essential maintenance. Apologies for the inconvenience.
 Various forms of shallow gas have been observed in seafloor strata, on the seabed and in the water column during acoustic profiling investigations in 2007–2009 in Xiamen Bay and adjacent areas. Acoustically transparent zones, acoustic turbidity and gas seepage can be seen in seabed strata, pockmarks and accumulation bodies have been found on the seafloor, and hummocky features and mushroom shaped gas signatures can be identified in the water column. This evidence shows that shallow gas is widely distributed in and around Xiamen Bay, due to degradation of the organic matter transported by the Jiulong River. The area covered by such features is roughly estimated at 150 km2, and methane flux is estimated to be 150 × 106 m3 assuming the thickness of gas bearing formations to be 1 m. This study shows that even small rivers flowing out onto a continent shelf contribute to recognizable methane flux and are linked to identifiable gas reservoirs in the shallow seabed. More detailed studies are required to understand the role of such systems as a component of the global carbon cycle.
 Shallow gas has been found on the continental shelf of China. Most studies using acoustic techniques have concentrated on the dangers to offshore petroleum exploration and drilling, the construction of bridges, the laying of pipelines, etc. [Ye et al., 2003; Gu et al., 2006; Li et al., 2010]. Acoustic techniques have been proven to provide quick and efficient identification and positioning of submarine gas emissions, and have also proven to be an appropriate tool for methane emission evaluations [Hagen and Vogt, 1999; Ergün et al., 2002; Sauter et al., 2006].
 The characteristics of shallow gas in acoustic profiles and the distribution of shallow gas accumulations in and around Xiamen Bay are presented here. These data suggest that submarine methane is not negligible in a small river estuary. In fact, such small river estuaries may be critical in understanding and evaluating methane emissions from continent shelves.
2. Regional Setting
 Xiamen Bay, located in the south of Fujian Province in China, can be divided into an inner port (including the Jiulong River estuary) and an outer port. The former is located in the west of Xiamen Island and the latter in the southeast of Xiamen Island (Figure 1). Xiamen Bay is a semi-enclosed bay, connecting the Jiulong River in the west with the open sea on the east. There are many islands and reefs in the bay, such as Jinmen, Dadan, and Xiamen islands, located mostly in the northeast, perpendicular to the main stream of the Jiulong River.
 The Jiulong River is a small mountain river, located in central part of Taiwan Strait. The main course of the Jiulong River is about 263 km long, the second largest river in Fujian Province with a drainage area of about 1.4 × 104 km2. Flow and sediment flux has inter-annual and annual variation. The average runoff to the ocean is 148 × 108 m3/yr, and the average annual suspended sediment flux into the sea is about 307 × 104 t, mainly concentrated in the flood season (June to September). The Jiulong River estuary is trumpet-shaped with an east–west orientation; there are low hills to the south and north, and a delta plain in the west. The estuary has submarine shoal development and the water depth is commonly 2 to 5 m. Since the Holocene, the sediment environment in the Xiamen Bay went through three stages: hilly river, fresh lake, and marine bay. Basal layer is weathered layer of granite parent rock or bedrock, overlying a layer of unequal thickness river alluvial sand. Above the river sand for freshwater lake mud, the top is the bay facies with sand and mud layer [Cai et al., 1987]. And the thickness of the Holocene is always less than 30 meters in the Xiamen Bay according to the cores [Cai et al., 1987; Hong and Chen, 2003]. Jiulong River estuary is connected to the open sea, more than 50 km away, through Xiamen Bay.
 A Probe 5001S sub-bottom profiling system and Klein 3000 sidescan sonar system were used in this study. Survey lines totaling more than 1500 km were collected during 2007–2009 in Xiamen Bay and adjacent areas (Figure 1). The sub-bottom profiling system has a high vertical resolution (20 to 30 cm), with a depth of penetration of >30 m in mud and a frequency of 3–10 kHz. The sidescan sonar has a maximum range of 600 m in the long-range (100 kHz) mode and of 150 m in high resolution (500 kHz) mode. A survey of grid of lines, 1 km apart navigated with differential GPS, was conducted in conjunction with this survey. Data were acquired at a ship speed of 4–6 knots. All seismic sub-bottom and sidescan sonar profiles were post-processed using the Geosurvey software and Sonarweb software respectively. Some methods, such as the band pass filter and TVG (Time Variation of Gain) were used to obtain good images. Then, with automatic and manual interactive modes, some geological features, such as shallow gas, bedrock positions were obtained. The results produced GIS information for analyses and mapping.
 Shallow gas occurrences can be divided into three categories according to their spatial position and characteristic features in the acoustic profiles. The first category is evidence of gas in sub-bottom strata, e.g., acoustic blanking zones (Figure 2) and acoustic turbidity. The second category consists of features manifested on the seabed, e.g., pockmarks and collapse pits caused by gas escape (Figure 3). Pockmarks have many forms (e.g., circular, oval or dish-shaped), are always distributed in groups, and generally are 30–40 m in width depending on sediment characters. The third category consists of features in the seawater column, e.g., acoustic plumes, cloudy turbidity and hummocky reflections. In the study area, shallow gas has escaped from seafloor strata into the water column in a variety of forms. Some were “cloud”-like (Figure 4a), some were “mushroom” shaped (Figure 4b), and even abnormal fluctuations can be seen in water with wedged (Figure 4c) or plumes features (Figure 4d). Some shallow gas can accumulate under the shallowest part of a sea bottom, with a gobbets state and a diffused shape. The shallow gas that accumulates in strata always masks below the sedimentary sequence, thereby obscuring underlying reflections. Shallow gas seepage reaching the seabed is found in this area, and the collapsed characters also can be seen in the sub-bottom profile and side-scan sonar image (Figure 3). These features show that shallow gas is widely spread in this area.
5.1. Abnormally Valued Assessments
 The flux of methane from the continental shelf is generally estimated using chemical sampling methods in the field [Middelburg et al., 2002; Zhang et al., 2004], and analysis of the methane in the laboratory. However, methane concentration and saturations have great spatial and temporal variability in available open reports. The maximum is several orders of magnitude larger than the minimum. For example, the methane concentration in surface water ranged from 6.9 to 173.7 nmol/L and surface saturation varied from 329 to 7896% in a study of the western Pearl River Estuary of China [Zhou et al., 2009]. The methane concentration and saturation also changes with the season in the same estuary. For example, the methane saturation in the Humber estuary during June and October of 1996 was 609–16691% and 566–21048%, respectively. The methane concentration varied correspondingly from 14.9–431 nmol/L and 15.5–666 nmol/L [Upstill-Goddard et al., 2000], compared to the minimum, with the maximum needing substantial change. The abnormal value is attributed to methane as it seeps from seafloor sediments [Zhou et al., 2009]. It is affected by pollution from allochthonous oil that enters the system from the coastal wetlands and river. In this study, the methane is observed to escape from the seabed in the acoustic profiles, which is the likely explanation for the reports with methane escaped from the Arctic continental margin seabed [Westbrook et al., 2009]. This shows that methane escape from the seabed may be a normal phenomenon in some estuaries [García-García et al., 1999; García-Gil et al., 2002]. And shallow gas was escaping during the days scale time of the investigation. Sampling can occasionally catch escaping methane, causing abnormally recorded high values of methane concentration and saturation. However, because there is no knowledge of the distribution of shallow gas deposits prior to surveying, there is less ability to observe and understand the migration of shallow gas in the water column. This can affect the assessment of its impact on the local environment using only chemical analysis methods.
5.2. Flux of Methane in Xiamen Bay and Adjacent Areas
 Xiamen Bay provides the only marine pathway between the Jiulong River and the open sea. This bay accommodates a large portion of the sediment discharge from the Jiulong River, including a large amount of organic material that accumulates in the estuary. Through the bacteriological reduction of organic matter and the geochemical transformation of buried biota [Floodgate and Judd, 1992], methane can be generated in rich organic and anoxic sediments. Such gas can accumulate in deltaic sediments, submarine channel sands and other types of reservoirs close to their depositional location, without the need for long-distance migration. Some methane gas can freely scatter in the area, forming a large-scale gas-bearing area. When shallow high-pressure gas in the strata is released all at once, this always causes the pressure on the gas bearing formation to dramatically fall. The rapid flow on the soil produces a strong flushing action, seriously disturbing the overlaying strata, and forming the collapse of the pit or other accumulation body. Pockmarks are interpreted to be the remnants of shallow gas escaping from marine sediment. In this study, in an area of about 150 km2, extensive evidence of shallow gas has been found – with the notable exception of the shallow-water portion of the Jiulong estuary. Some studies have suggested that the thickness of the shallow gas is only 1–2 m [Schubel and Scheimer, 1973]. If the thickness of shallow gas in this study is taken to be 1 m according to the study by Schubel and Scheimer , then the reservoir and escaped shallow gas will reach 150 × 106 m3 in the upper 30 m of the seabed using acoustic profiling data (Figure 4). The large scale of this accumulation shows the significance of buried and escaped shallow gas from small river estuaries and their adjacent regions.
Guo et al.  performed 56 surface water analyses and investigated four short sediment cores for methane in the Jiulong River estuary. The results show that the concentration of methane varies from 10.7 to 456.7 nmol/L, and that an important source of the methane is the mangrove intertidal wetland in the upper estuary. At the same time, the results show that most of the shallow gas is gathered far from the estuary, even more than 80 km away from the estuary, it is apparently the release of methane from the mangrove intertidal wetland has little effect in this area. Shallow gas may accumulate in the outer area for various reasons.
Milliman and Meade , Milliman and Syvitski , and Milliman et al. calculated the statistics of the river sediment discharge based on an ever-increasing database over the last 30 years. One important result is that the sedimentary discharge of small rivers (especially small mountain river) has been severely underestimated. Comparing this work with earlier concepts [Holeman, 1968], shows that previously calculated discharges from small rivers were underestimated by about three orders of magnitude [Wang, 2009]. The underestimated flux and significant increase in modeled organic-rich sediments in coastal estuarine and deltaic areas will result in a substantial increase in estimates of microbial methane generation (and seepage) [Judd et al., 2002], probably generated at short (decadal) time scales. This is confirmed by the seasonal change of flux discharge in the Jiulong River, the maximum annual runoff is two to three times greater than the minimum runoff. Average annual suspended sediment flux into the sea is mainly concentrated in the flood season (June to September) [Zeng et al., 1998].
 Another important control on gas accumulation in estuaries is human activity. With the development of an ocean economy, marine development activities have greatly intensified. Some new economic areas have developed in China in recent years, such as the West Side of the Economic Area in the Taiwan Straits, part of this study area. This development is associated with a large number of offshore construction projects. The construction of these facilities will have a direct impact on shallow gas formation and in some cases may cause instabilities in gas-bearing formations. However, we have little understanding of dynamic processes of accumulation and expulsion in these shallow gas accumulations. Little is known of how the shallow gas will escape or migrate elsewhere if mobilized by human activities.
 Evidence of shallow gas in a range of spatial configurations has been characterized by acoustic profiling in Xiamen Bay and adjacent areas. Shallow gas accumulations are distributed over a region of approximately 150 km2. If the shallow gas is taken to be a 1-m-thick layer of methane, then the buried and escaped methane flux can reach 150 × 106 m3within the upper 30 m of the seabed. However, Xiamen Bay and adjacent areas likely receive most of the sediment discharge from the Jiulong River, a small mountain river with a sediment discharge into the sea of 3 Mt/yr. Therefore, this study shows that even a small river may be capable of generating widely-distributed shallow gas deposits, highlighting the significant role of small rivers in methane escape processes.
 We have much to understand regarding the formation, migration, and distribution of shallow gas on the continent shelf. These topics are worthy of more detailed investigations, and need to be re-evaluated as a component of the global carbon cycle.
 This work was Sponsored by the Scientific Research Foundation of Third Institute of Oceanography, SOA (2009004) Chinese Polar Environment Comprehensive Investigation&Assessment Programs (CHINARE2012-01-03) and the Ocean Public Welfare Scientific Research Project (201005029).
 The Editor thanks Ana Garcia-Garcia and an anonymous reviewer for their assistance in evaluating this paper.