Search

There are many approaches to the investigation of oil spills, but many share the same purpose of identifying the source of the oil and determining how much oil came from the spill versus other potential sources. This chapter outlines a few of the established multivariate methods that may assist with such aims. Principal components analysis (PCA) is useful for aggregating both samples that have similar compositions and compounds that are behaving similarly in the environment, but on its own, it does not quantify how much of any one source is present in a sample. Other approaches such as polytopic vector analysis (PVA) are particularly good at this although as with any such statistical tool, knowledge of the chemistry and the environment is necessary to achieve a sensible resolution. These methods are compared to traditional univariate or bivariate methods of determining the changes in oil chemistry over time and additionally highlight the problem of multiple oils at different stages of degradation.

Statistical methods form the basis of many environmental forensic investigations. These techniques can bring clarity to complex data and assist in the identification of the environmental history of samples and locations. However, treating the methods as a black box without appreciation of the method itself and its limitations can lead to erroneous interpretations. This chapter outlines some of the major methods used in multivariate statistical analyses from the standpoint of the end user; it highlights how data should be presented, transformed, and ultimately presented. Three major methods are used in this chapter: PCA, PLS, and PVA. In summary, data should be converted to proportions to remove the concentration effects, transformed to their log10 values if needed, the presence of zero should be considered—especially why they are zero in the first place—and finally how the user can interpret the results needs to be considered.

Spilled oil in the coastal zone may be remediated through biodegradation by naturally occurring bacteria. It is possible to enhance the removal rates through addition of nutrients and biosolvents. These rates may differ within the intertidal area due to many environmental factors including surf washing. Laboratory experimentation is complex when there are so many factors involved. In a simple three factor remediation experiment, the effect of the timing of addition of a biosolvent, the type of nutrients added and the quantity of biosolvent relative to the amount of oil spilled were examined at three levels. A response surface methodology (RSM) was used to identify the key experiments to conduct and 17 separate trials were carried out with high, mid and low tide microcosms. The petroleum hydrocarbons were quantified by GC–MS methods and the data were examined with Design of Experiments (MODDE) and Partial Least Squares (PLS) Statistical (SIMCA-P) software. Different factors were important at the different intertidal locations: at low tide, the timing of application was most important while in the mid tide location, the proportion of biosolvent was most important. Of the nutrient additions, inorganic nutrients alone were more effective than organic forms of nitrogen or mixtures of urea with inorganic nutrients. Overall, the amount of biosolvent was the least important of the three factors examined. The use of RSM significantly reduced the experimental effort needed to investigate the factors and their interactions.

Generally the waters of the Ria Formosa Lagoon, Portugal have a short residence time, in the order of 0.5 days (Tett, P., Gilpin, L., Svendsen, H., Erlandsson, C.P., Larsson, U., Kratzer, S., Fouilland, E., Janzen, C., Lee, J., Grenz, C., Newton, A., Ferreira, J.G., Fernandes, T., Scory, S., 2003. Eutrophication and some European waters of restricted exchange. Continental Shelf Research 23, 1635–1671). This estimation is based on the measurements of currents and the modelling of water exchange at the outlets to the ocean. However, observations of the temperature and salinity in the inner channels imply that residence time is greater in these regions of the lagoon. To resolve this apparent contradiction, spatial measurements of the temperature and salinity were made with a meter for conductivity, temperature and depth along the principal channels of the western portion of the lagoon, with a sampling frequency of two per second. Evaporation rates of 5.4 mm day −1 were measured in a salt extraction pond adjacent to the lagoon and used to determine the residence time through salinity differences with the incoming seawater. In June 2004, the water flooding in from the ocean had an average salinity of 36.07 which contrasted with a maximum of 37.82 at mid ebb on a spring tide, corresponding to a residence time of >7 days; the mean residence time was 2.4 days. As the tide flooded into the channels, the existing water was advected back into the lagoon. Although there was a small amount of mixing with water from another inlet, the water body from the inner lagoon essentially remained distinct with respect to temperature and salinity characteristics. The residence time of the water was further prolonged at the junction between the main channels, where distinct boundaries were observed between the different water masses. As the water ebbed out, the shallow Western Channel was essentially isolated from the rest of the outer lagoon, and the water from this channel was forced down the Ramalhete Channel, from where it was unable to exit the lagoon in one tidal cycle due to the extensive path length of ∼14 km to the sea. Although the overall exchange rate of water is short in the outer lagoon, this study emphasizes that management models should take into account additional complexities that might arise from the much longer exchange rates of the inner lagoon. For example, the principal sewage discharge for the urban area of Faro is into the section of the Ramalhete Channel where efficient flushing is impeded by the relatively high residence times of the water body in this channel. The implementation of the techniques used for this study are a quick and relatively cost effective approach to testing assumptions about water quality and exchange in shallow coastal systems.

Historical data from the Ria Formosa lagoon are classified according to the EEA 2001 guidelines to provide a frame of reference to evaluate the effect of management during the implementation of the environmental legislative Directives. Water samples from the Ria Formosa lagoon were significantly enriched in nitrogen (NH4+ NO2− and NO3−) with respect to the adjacent coastal waters indicating that inputs from sewage, agricultural runoff and benthic fluxes were not fully assimilated within the lagoon. Tidal flushing was insufficient in the inner areas of the lagoon to remove or effectively dilute these inputs. Enrichment was most severe close to the urban centres of Faro and Olhao, as well as in the Gilao Estuary and the shallow extremities. Dissolved oxygen undersaturation (mean 75% during daylight hours) was associated with the area close to the sewage outlets of Faro. In the shallow west end of the lagoon during summer, dissolved oxygen supersaturation reached 140% during the day but fell to 50% at night. Classification using the EEA (2001) guidelines suggests the system is “poor” or “bad” with respect to phosphate concentrations for the majority of the year and “poor” in nitrogen contamination during the autumn rainy period. Due to the high overall nitrogen load in the lagoon, there is a net export to the coastal waters, especially during November and December, and phosphate only becomes limiting briefly during the spring bloom (April). Therefore, substantial phytoplankton populations may be supported year-round in the lagoon. The consequences of water quality deterioration in the Ria Formosa would negatively affect the lagoon as a regional resource, important for its ecological, economic and recreational value. The industries most affected would be tourism, fisheries and aquaculture. Management options include Urban Waste Water Treatment, dredging, artificial inlets, limits on urban development and changes in agricultural practices.