Floods and livelihoods: The impact of changing water resources on wetland agro-ecological production systems in the Tana River Delta,Kenya

Wetlands are highly dynamic and productive systems that have been under increased pressure from changes in land use and water management strategies. In Eastern Africa, wetlands provide resources at multiple spatial and temporal levels through farming, fishing, livestock ownership and a host of other ecosystem services that sustain the local economy and individual livelihoods. As part of a broader effort to describe future development scenarios for East African coastal wetlands, this qualitative study focuses on understanding the processes by which river water depletion has affected local food production systems in Kenya's Tana River Delta over the past 50 years, and how this situation has impacted residents’ livelihoods and well-being. Interviews performed in six villages among various ethnic groups, geographical locations and resource profiles indicated that the agro-ecological production systems formerly in place were adapted to the river's dynamic flooding patterns. As these flooding patterns changed, the local population diversified and abandoned or adopted various farming, fishing and livestock-rearing techniques. Despite these efforts, the decrease in water availability affected each subcomponent of the production systems under study, which led to their collapse in the 1990s. Water depletion negatively impacted local human well-being through the loss of food security. The current study provides a detailed account of the dynamics of agro-ecological production systems facing the effects of river water depletion in a wetland-associated environment in Sub-Saharan Africa.     

A case study of using Raman lidar measurements in high-accuracy GPS applications

This paper investigates the impact of rapid small-scale water vapor fluctuations on GPS height determination. Water vapor measurements from a Raman lidar are used for documenting the water vapor heterogeneities and correcting GPS signal propagation delays in clear sky conditions. We use data from four short observing sessions (6 h) during the VAPIC experiment (15 May–15 June 2004). The retrieval of wet delays from our Raman lidar is shown to agree well with radiosonde retrievals (bias and standard deviation (SD) were smaller than 1 and 2.8 mm, respectively) and microwave radiometers (from two different instruments, bias was 6.0/−6.6 mm and SD 1.3/3.8 mm). A standard GPS data analysis is shown to fail in accurately reproducing fast zenith wet delay (ZWD) variations. The ZWD estimates could be improved when mean post-fit phase residuals were removed. Several methodologies for integrating zenith lidar observations into the GPS data processing are also presented. The final method consists in using lidar wet delays for correcting a priori the GPS phase observations and estimating a scale factor for the lidar wet delays jointly with the GPS station position. The estimation of this scale factor allows correcting for a mis-calibration in the lidar data and provides in the same way an estimate of the Raman lidar instrument constant. The agreement of this constant with an independent determination using radiosonde data is at the level of 1–4%. The lidar wet delays were derived by ray-tracing from zenith pointing measurements: further improvement in GPS positioning is expected from slant path lidar measurements that would properly account for water vapor anisotropy.     

The use of transplanted cultured tropical oysters (Saccostrea commercialis) to monitor Cd levels in North Queensland coastal waters (Australia)

Bivalves are commonly used to detect metal pollution in the marine environment. Commercially cultured Milky oysters (Saccostrea commercialis) were transplanted in various sites along the North Queensland coast and analyzed for two metals of potentially anthropogenic origin (Cd, Zn). To provide additional information, naturally occurring Black Lip oysters (Saccostrea echinata) were also collected at the transplantation sites. The study demonstrated that the oysters species transplanted are good bioindicators of these metal concentrations in tropical waters, sensitive to variations in the environment at concentrations which are much smaller than pollution signals commonly reported for temperate waters. Three transplant experiments were carried out from May 1999 to February 2000. Milky oysters transplanted to offshore areas (Orpheus Is., Kelso Reef) accumulated Cd in the soft parts whereas oysters sampled from cages placed in Ross Creek and the Herbert River estuaries showed a decrease in Cd concentration, which resulted from an increase in dry weight. Dry weight appeared to be an important covariant affecting Cd concentration in the oysters whereas it does not unambiguously affect Zn concentrations. For the duration of the experiments, oysters sampled from the Magnetic Is. reference site showed effectively constant Cd concentrations and total Cd contents which indicates that any seasonal cycle affecting metal concentration is weak. It was found that Cd accumulation in oysters increased as ambient dissolved Cd concentration decreased, from which it was concluded that for these oysters, the predominant source of Cd was from the particulate phase rather than the dissolved phase.     

Influence of submerged aquatic vegetation on size class distribution of perch (Perca fluviatilis) and roach (Rutilus rutilus) in the littoral zone of Lake Geneva (Switzerland)

The abundance of different size classes of perch and roach in the littoral zone of Lake Geneva was compared between submerged aquatic vegetation and unvegetated zones. Samples were taken with gillnets during four periods between June and October 1993. During the vegetation period (June to September), perch 9 cm and roach 10 cm were more abundant in vegetation whereas roach > 20 cm were more abundant in open water. Perch larger than 18 cm and medium roach were equally distributed in both habitats whatever the period, whereas medium perch distribution fluctuated according to the period. In October, after the decline of the vegetation, no more differences in fish distribution were observed except for small roach, which were always more abundant in the vegetated sites.     

Characterization of juvenile pyroclasts from the Kos Plateau Tuff (Aegean Arc): insights into the eruptive dynamics of a large rhyolitic eruption

Silicic pumices formed during explosive volcanic eruptions are faithful recorders of the state of the magma in the conduit, close to or at the fragmentation level. We have characterized four types of pumices from the non-welded rhyolitic Kos Plateau Tuff, which erupted 161,000 years ago in the East Aegean Arc, Greece. The dominant type of pumice (>90 vol.%) shows highly elongated tubular vesicles. These tube pumices occur throughout the eruption. Less common pumice types include: (1) “frothy” pumice (highly porous with large, sub-rounded vesicles), which form 5–10 vol.% of the coarsest pyroclastic flow deposits, (2) dominantly “microvesicular” and systematically crystal-poor pumices, which are found in early erupted, fine-grained pyroclastic flow units, and are characterized by many small (<50 μm in diameter) vesicles and few mm-sized, irregular voids, (3) grey or banded pumices, indicating the interaction between the rhyolite and a more mafic magma, which are found throughout the eruption sequence and display highly irregular bubble shapes. Except for the grey-banded pumices, all three other types are compositionally identical and were generated synchronously as they are found in the same pyroclastic units. They, therefore, record different conditions in the volcanic conduit leading to variable bubble nucleation, growth and coalescence. A total of 74 pumice samples have been characterized using thin section observation, SEM imagery, porosimetry, and permeametry. We show that the four pumice types have distinct total and connected porosity, tortuosity and permeability. Grey-banded pumices show large variations in petrophysical characteristics as a response to mingling of two different magmas. The microvesicular, crystal-poor, pumices have a bimodal bubble size distribution, interpreted as reflecting an early heterogeneous bubble nucleation event followed by homogeneous bubble nucleation close to fragmentation. Finally, the significant differences in porosity, tortuosity and permeability in compositionally identical tube and frothy pumices are the result of variable shear rates in different parts of the conduit. Differential shear rates may be the result of either: (1) pure shear, inducing a vertical progression from frothy to tube and implying a relatively thick fragmentation zone to produce both types of pumices at the same time or (2) localized simple shear, inducing strongly tubular vesicles along the wall and near-spherical bubbles in the centre of the conduit and not necessarily requiring a thick fragmentation zone.     

Assessment of the ecological quality status of soft-bottoms in Reunion Island (tropical Southwest Indian Ocean) using AZTI marine biotic indices

The ability of the two synthetic marine biotic indices, AMBI and M-AMBI, to account for changes in the ecological quality of coastal soft bottoms of Reunion Island according to disturbances was assessed from macrobenthic samples collected in five sectors between 1994 and 2004. Samples were collected under non-perturbed conditions and at two sites subjected to heavy organic enrichment. Both indices are based on a classification of macrofauna into ecological groups (EG), and their transfer to tropical waters required some adaptations. These indices proved efficient in detecting a degradation of habitat quality. Their use resulted in the classification of all sites sampled between 1996 and 1998 as "good" or "high". M-AMBI nevertheless tended to result in the attribution of a slightly worse ecological quality status than AMBI. Together with an update of the EG species list for the Indian Ocean area, our results support the extension of both indices for the assessment of tropical soft bottoms.     

222Rn as Natural Tracer for LNAPL Recovery in a Crude Oil‐Contaminated Aquifer

The objective of this study was to investigate whether ²²²Rn in groundwater can be used as a tracer for light non‐aqueous phase liquid (LNAPL) quantification at a field site treated by dual‐phase LNAPL removal. After the break of a pipeline, 5 ha of soil in the nature reserve Coussouls de Crau in southern France was contaminated by 5100 m³ of crude oil. Part of this oil seeped into the underlying gravel aquifer and formed a floating oil body of about 3.9 ha. The remediation consists of plume management by hydraulic groundwater barriers and LNAPL extraction in the source zone. ²²²Rn measurements were performed in 21 wells in and outside the source zone during 15 months. In uncontaminated groundwater, the radon activity was relatively constant and remained always >11 Bq/L. The variability of radon activity measurements in wells affected by the pump‐and‐skim system was consistent with the measurements in wells that were not impacted by the system. The mean activities in wells in the source zone were, in general, significantly lower than in wells upgradient of the source zone, owing to partitioning of ²²²Rn into the oil phase. The lowest activities were found in zones with high non‐aqueous phase liquid (NAPL) recovery. LNAPL saturations around each recovery well were furthermore calculated during a period of high groundwater level, using a laboratory‐determined crude oil–water partitioning coefficient of 38.5 ± 2.9. This yielded an estimated volume of residual crude oil of 309 ± 93 m³ below the capillary fringe. We find that ²²²Rn is a useful and cheap groundwater tracer for finding zones of good LNAPL recovery in an aquifer treated by dual‐phase LNAPL removal, but that quantification of NAPL saturation using Rn is highly uncertain.     

Comments on a method allowing permanent preparations of phytoplankton for light microscopy and the accurate selection and positioning of its cells for transmission electron microscopy

A new flat embedding method is commented. This method allows embedding of the same phytoplankton sample for light and transmission electron microscopy. Rare or interesting field specimens can then be easily selected and sectioned without the need of cultures. A great range of biological material can be handled with this method.     

Isotopic and geochemical characterization of salinization in the shallow aquifers of a reclaimed subsiding zone: The southern Venice Lagoon coastland

The coastal plain bordering the southern Venice Lagoon is a reclaimed lowland characterized by high subsidence rate, and ground level and water-table depth below sea level. In this agricultural region, where the surface hydrologic network is entirely artificially controlled by irrigation/drainage canals, salinization problems have long been encountered in soils and groundwaters. Here we use isotopic and geochemical tracers to improve our understanding of the origin of salinization and mineralization of the semi-confined aquifer (0–40 m), and the freshwater inputs to this hydrological system. Water samples have been collected at different seasons in the coastal Adriatic Sea, lagoon, rivers and irrigation canals, as well as in the semi-confined aquifer at depths between 12 and 35 m (14 boreholes), and in the first confined aquifer (three boreholes drilled between 40 and 80 m depth). Stable isotopes (δ¹⁸O and δD) and conductivity profiles show that direct saline intrusion from the sea or the lagoon is observed only in a restricted coastal strip, while brackish groundwaters are found over the entire topographic and piezometric depression in the centre of the study area. Fresh groundwaters are found only in the most western zone. The sharp isotopic contrast between the western and central regions suggests disconnected hydrological circulations between these two parts of the shallow aquifer. The border between these two regions also corresponds to the limits of the most strongly subsiding zone.Our results can be interpreted in terms of a four end-member mixing scheme, involving (1) marine water from the lagoon or the open sea, (2) alpine and pre-alpine regional recharge waters carried either by the main rivers Adige, Bacchiglione and Brenta (irrigation waters) or by the regional groundwater circulation, (3) local precipitation, and (4) evaporated waters infiltrated from the surface. Infiltration from the surface is also revealed by the stratification of the electrical conductivity profiles, showing that the brackish groundwaters are overlain by a shallow layer of less saline water all over the central depression. In the first confined aquifer, the groundwaters have isotopic compositions similar to the deep groundwaters of the Venetian confined aquifers (40–400 m depth). The isotopic data and the Br/Cl ratio show that the origin of the salinization of the phreatic aquifer can be ascribed to seawater intrusion alone, with no indication of the involvement of deep brines (identified at 450 m depth) in the process.The chemical composition of the saline and brackish groundwaters is characterized by an excess of sodium and a deficit of calcium compared to conservative mixing between fresh groundwaters and seawater. This suggests that the phreatic aquifer is progressively freshening, as a consequence of the beneficial influence of the extensive irrigation/drainage network, including raised canals acting as a hydraulic barrier along the coast. This freshening tendency may have been lasting since the reclamation in the mid-twentieth century, and has probably been accelerated by the ban on groundwater abstraction since the 1970s.     

Intentional contamination of water distribution networks: developing indicators for sensitivity and vulnerability assessments

Performing a comprehensive risk analysis is primordial to ensure a reliable and sustainable water supply. Though the general framework of risk analysis is well established, specific adaptation seems needed for systems such as water distribution networks (WDN). Understanding of vulnerabilities of WDN against deliberate contamination and consumers’ sensitivity against contaminated water use is very vital to inform decision-maker. This paper presents an innovative step-by-step methodology for developing comprehensive indicators to perform sensitivity, vulnerability and criticality analyses in case of absence of early warning system (EWS). The assessment and the aggregation of these indicators with specific fuzzy operators allow identifying the most critical points in a WDN. Intentional intrusion of contaminants at these points can potentially harm both the consumers as well as water infrastructure. The implementation of the developed methodology has been demonstrated through a case study of a French WDN unequipped with sensors.