In situ pumps

Sampling artifacts. The use of in situ pumps to collect water samples for " Th analysis permits simultaneous collection (and separation) of different particle fractions as well as dissolved Th. As pumping systems have been modified to permit determination of POC on the pump filters, it became possible to compare POC determined from the pump samples with conventional POC determinations made on small volume samples (0.5 - 2 L) taken from hydrocasts. The JGOFS data from multiple studies show large discrepancies between these two sample collection methods, with pump POC values 3 to 100 times lower than bottle POC values. Possible artifacts with each approach have been identified. For example Moran et al. (1999) have suggested that DOC is adsorbed onto... 

B Bottle IM In situ pump/membrane filter (53, 70 p,m) 1C In situ pump/cartridge filter (0.5,1 am) ST sediment trap 5°S-5°Nand25°W-5°E... 

Dawson et al. [227] described samplers for large-volume collection of sea water samples for chlorinated hydrocarbon analyses. The samplers use the macroreticular absorbent Amberlite XAD-2. Operation of the towed fish type sampler causes minimal interruption to a ship s programme and allows a large area to be surveyed. The second type is a self-powered in situ pump which can be left unattended to extract large volumes of water at a fixed station. 

Weinstein, S.E. and S.B. Moran. 2004. Distribution of size-fractionated particulate trace metals collected by bottles and in-situ pumps in the Gulf of Maine-Scotian Shelf and Labrador Sea. Mar. Chem. 87 121-135. 

A high vacuum chamber contains a liquid N2-cooled surface (A= 400 cm2) to provide a high, in situ, pumping speed for water molecules desorbing from its surface. Calculate the pumping speed for water vapour at 50 °C if all the impinging molecules are trapped. 

The improvement in water sampling techniques such as the development of the Kiel in-situ Pump (KISP) [122] and high resolution gas chromatography (such as MDGC-ECD) has resulted in a series of papers on the concentration of di-, mono-, and occasionally non-ortho CB levels in world waters. This is detailed in Table 4. 

Below about 200 m, PON concentrations tend to asymptote to similar levels of about 0.1 to 0.3 pmol kg", irrespective of the large differences in PON in the euphotic zone. Overall, PON dynamics in the aphotic zone tends to follow that of particulate organic carbon (POC), with a stoichiometric C N ratio of about 7.1 0.1 (equivalent to 16 114 2) for the particles sinking out of the euphotic zone (Schneider et al, 2003) (see also Hebei and Karl (2001) for long-term observations of the POC and PON dynamics in an oligotrophic environment). In their analysis of sediment trap samples and particles collected by in situ pumps, Schneider et al (2003) also noticed a slight increase of this ratio with depth of about 0.2 per 1000 m, so that particles sinking across 1000 m have a C N ratio very close to the 117 16 ratio proposed by Anderson and Sarmiento (1994). 

In summary, rFIA seems best suited for analysis of near-surface samples where concentrations are low. In situ pumps can be used to sample the... 

Particulate organic carbon is certainly being resuspended into the water column by the processes discussed above. These resuspension processes sometimes make a distinct delineation of the interface between the seawater and the sediment difficult. Lee et al. (1979) report measurements of sterols made on a surface floe material from the top of cores from the western North Atlantic. The sterol composition of this material, as well as the fatty acid and hydrocarbon content (J.W. Farrington, unpublished data) show that this floe material is of different composition than the underlying material. Whether this material is recently deposited and about to become part of the permanent deposit or whether it is material which is continually resuspended and moved along laterally to areas more conducive to permanent deposition is unknown. The measurement of specific organic compounds in this floe material and in samples from deep sediment traps or deep in situ pump filters should provide more insight into the question of net transport by resuspension in the deep sea. 

Another problem is the lack of representation of rarer particles in the volumes which can be sampled. In the course of some twenty years of work on particulate matter, I have examined thousands of filters under the microscope I have yet to see my first foraminiferal test from samples taken below the surface layer, even in regions where the sediment was greater than 80% calcium carbonate. To relate what is happening at the sea surface to what has resulted at the sea floor, very much greater volumes of water must be sampled than is practical with water bottle samplers. This, of course, is what the in situ pumps were designed for in the first place, and what they do very well. The pumps give us a snapshot of what is in the water column at a particular time. With the newer pumps, which can be deployed at a series of depths simultaneously, the losses due to solution, remineraHzation, or predation in the distances between pump depths can be estimated, after some aUowances have been made for the differences in background particulate burden in different water masses. What cannot be obtained from the pumps is any measurement of rate. 

Figure 4.18 Pressure decrease in a VIP due to the in situ pumping effect of the COMBOGETTER.

Traditional adsorption preconcentration techniques are based aroimd the passage of filtered (and sometimes acidified or otherwise pretreated) water through a column of adsorbent. After passage of the sample the compounds of interest are eluted with an appropriate solvent or solvent mixture. Such techniques can be laboratory based, in situ (pumped systems or, more rarely, in passive towed units), or, in more recent systems, can be used in online applications. 

In situ filtration is reviewed briefly in Chapter 2 (Section 2.1.3.3). Here we outline the procedure for the collection of suspended particulate TE with the Kiel in situ Pump (KISP) introduced by Petrick et al. (1996) and also technically described in Chapter2 (see Fig. 2-2). The system has been developed primarily for the determination of organic compounds at ultra-trace levels (see also Chapter 22). 

Filtration. Except for the in situ pumps, large-volume filtration has to be performed on board, for which a variety of pressure or suction techniques is available (compare with Chapter 2 and Chapter 12, Section 12.6). In order to prevent the loss of any particulate or dissolved activity, the risk of settling of particles and adsorption/adhesion to walls of the sampling bottles has to be considered. If a water pump is used before the filter, it should be carefully checked, e.g., by microscopy, whether the pump leaves the particles intact. Centrifugal pumps are notoriously harmful to plankton. An elegant filtration method has been described by Dehairs et al. (1997). A 30L calibrated Perspex cylinder with a conical bottom is filled with the sample obtained from a Rosette cast. The outlet is connected to a 142 mm Teflon filtration unit. Controlled (50-100 kPa) air pressure is applied to a series of these units from a simple small compressor. 

Water for the analysis of suspended particles may be collected with any clean water sampler used for oceanographic work, e.g., Niskin bottles, Go-Flo bottles, Nansen bottles. Hydrobios water samplers (see also Chapter 1). The sample volume will depend on the expected concentration of POM. In nearshore and/or biologically productive water, 0.5-2 L usually is an adequate sample volume. Ten litres may be required in particle-poor open ocean waters such as the Sargasso Sea. For sampling large volumes of water, 30 L Niskin bottles are recommended. Even larger volumes may be sampled with in situ pumps. (See Chapters 1,2 and 13). 

The Kiel in situ pump system (KISP) for filtration and extraction of trace organics at the depth of sampling Petrick et al, 1996) is suitable for volumes of up to 2000 L or more (depending on particle concentration) and depths in excess of 6000 m. Sufficient amounts of PAH and CB may thus be collected in open-ocean waters to allow their analytical determination at concentration levels around or below 0.01 pg/L. The sampler is depicted in Fig. 22-1 (technical details can be found in Chapter 2). 

Toulhoat et al. (1989) studied the retention behaviour of ions by in situ pumping experiments in low permeability formations (aquifers) and found a clear decrease of uranium content as a result of the pumping procedure. 

---