Diatom and geochemical indicators of acidification in a tropical forest stream, singapore 3

Unlike other environmental archives, such as documentary historical sources, the paleolimnological approach, involving studies based on the biological, chemical, and physical information

3.4 Geochemical Analysis – Total Sulphur Content 41

polycyclic aromatic hydrocarbons

3.1 Overview

With the causes of and impacts from acid deposition examined in chapter two, chapter three looks at how the possible acidification of freshwater ecosystems is investigated The chapter begins by looking at the difficulties faced

by researchers investigating acid deposition on a freshwater ecosystem, namely that there is a lack of direct data monitoring the acidity of a water body and there

is also a lack of suitable study sites Researchers therefore use sedimentary records from suitable water bodies to trace potential acidification

The chapter then looks at the methodology employed in this study to examine the potential acidification of Jungle Falls stream – diatoms and geochemical analysis The majority of paleolimnological studies into lake acidification involve the use of diatom analysis These microscopic algae are often preserved well in freshwater ecosystems and, as they are highly sensitive

to changing environmental conditions, they make an excellent proxy for environmental conditions within an ecosystem Other techniques involve examining the geochemistry of the record, specifically the variation in sulphur, lead, zinc, potassium, sodium, iron and manganese levels within the sediment This will help track the levels of atmospheric pollution and contamination going into a water body

Lastly, chapter three briefly elaborates on alternative paleolimnological methods used to study the acidification of freshwater ecosystems that are not employed in this study These include the use of spheroidal carbonaceous particles (SCPs) and polycyclic aromatic hydrocarbons (PAHs), other biological evidence, and mineral magnetic analysis

3.2 Investigating acid deposition

Investigators studying acid deposition and its effects on freshwater ecosystems face two significant hurdles Firstly, as acid deposition spreads over

a broad area and can cover entire regions, there are a lack of control lakes for

comparison (Mitchell et al, 1985) In the tropical environments, this is

compounded by another related issue – there are significantly less lakes than in temperate areas, as lakes of glacial origin are extremely rare in the tropics (Lewis, 1996) By studying available maps, Lewis (1996) estimates that no more than 10% of lakes worldwide are tropical, demonstrating the importance of glaciation in the formation of lakes at temperature latitudes In tropical latitudes, most lakes have a riverine origin, with other lakes having volcanic, coastal, man-made or aeolain origins (Lewis, 1996)

Another hurdle faced when studying acid deposition is that historical data monitoring the changes in water chemistry in a lake or river are often unavailable

or imprecise (Pienitz et al, 2006) This is because many environmental changes

and impacts are rarely foreseen and consequently are not monitored during the period of change from pristine to present day conditions (Renberg and Battarbee, 1990) Thus, baseline data of pre-acidification conditions are often nonexistent and yet are exceedingly vital (Mannion, 1999) In this absence of long-term monitoring, lake sediments offer one of the few reliable and effective ways of identifying the onset, rate and variation of environmental contamination in a

freshwater ecosystem (Charles et al, 1987; Rose and Rippey, 2002)

Unlike other environmental archives, such as documentary historical sources, the paleolimnological approach, involving studies based on the biological, chemical, and physical information preserved in lake sediments, often provides a record that is continuous, can cover both short and long timescales, and usually accumulates rapidly enough to provide a high resolution record (Renberg and Battarbee, 1990) This approach is effective in reconstructing past changes in water chemistry variables because biota and geochemical processes

respond in predictable ways to changes in lake water chemistry (Antoniades, 2007)

Since acidification is a dynamic process, “effective management will require analyses of trends over different time scales, including estimates of pre-acidification conditions” (Smol, 2008: 92) Prior to intervention and management, scientists need to prove conclusively that a lake had been acidified through anthropogenic pollution and that it was not naturally acidic or acidified through natural processes (Antoniades, 2007) Thus, “without the historical perspective that paleolimnology can provide, many naturally acidic lakes may be unjustifiably limed, resulting in massive alterations to specialised ecosystems and food webs that have persisted for thousands of years in a naturally low pH state” (Smol, 2008: 105)

During the 1980s, due to rising concerns about the effects of acid deposition on freshwater ecosystems, two major paleolimnological projects were started – the Surface Waters Acidification Programme (SWAP) in Europe and the Paleoecological Investigation of Recent Lake Acidification (PIRLA) project in North America The SWAP project focussed on tracing the recent (post-1800) history of a number of carefully chosen lakes in Norway, Sweden and the UK in order to assess the causes of acidification rather than focussing on the evidence

of acidification per se (Battarbee and Charles, 1987; Renberg and Battarbee,

1990) The PIRLA project looked into the history and effects of acid deposition, spatially and temporally, on lakes in eastern North America in order to determine the relative role anthropogenically induced atmospheric acid deposition played in

causing recent acidification (Battarbee and Charles, 1987; Moser et al, 1996)

One of the main focuses of these paleolimnology programmes was therefore to test alternative or additional causes for lake acidification (Renberg

and Battarbee, 1990) They have been instrumental in identifying the major cause

of acid deposition as fossil fuel combustion (Mannion, 1992) The main methods employed in these paleolimnological investigations of acidification are biological analyses (diatom, pollen, scaled chrysophytes, caldoceran and chironomid analysis), geochemical analysis (such as sulphur concentrations and heavy metal concentrations), examining SCPs and PAHs along with dating techniques (Renberg and Battarbee, 1990; Mannion, 1992; Smol, 2008)

Overall, the SWAP and PIRLA projects found that at individual sites, recent acidification always postdates the beginning of major industrialisation in the late 18th and early 19th century Diatoms are often the first indicator to respond to atmospheric contamination and, when comparing diatom-inferred pH trends with regional patterns of sulphur deposition, “recently acidified sites are found in areas of high S deposition and no recently acidified sites have been reported from areas of very low S deposition“ (Renberg and Battarbee, 1990: 296)

3.3 Diatom Analysis

Biological evidence, in particular diatoms, has been key in reconstructing past environments and is based on the principle of uniformitarianism, “namely that a knowledge of factors that influence the abundance and distribution of contemporary organisms enables inferences to be made about environmental controls on plant and animal populations in the past” (Lowe and Walker, 1997: 162) Three criteria should be considered for biological proxies to be useful for environmental reconstruction – “the material must withstand decomposition, exhibit sufficient morphological differences to be of taxonomic significance and provide sufficient quantities to reflect the nature of the entire assemblage from which it is derived” (Rovner, 1971: 343-4) Often fulfilling the three criteria above, diatoms have thus proved valuable in paleolimnological acidification research

Diatoms are microscopic algae found in almost all aquatic environments

(Battarbee et al, 2001) and are the dominant algal group in freshwater systems

(Smol, 2008) They have a resistant siliceous outer shell and are thus a popular biological proxy in paleolimnological reconstructions (Mannion, 1982; Korhola, 2007) Diatoms have six characteristics that make them particularly useful:

1 Large number of species: There are thousands of diatom species (Smol, 2008) This makes their assemblages taxon-rich, increasing the ecological information obtained and strengthening the confidence of the environmental reconstructions (Korhola, 2007)

2 Easily identified: As diatoms have been comprehensively documented and classified, scientists are able to identify them down to species or even subspecies levels (Korhola, 2007) When well preserved, diatoms are also readily identified and counted (O’Hara, 2000)

3 Sensitive indicators: Diatoms cover a wide range of environmental conditions, yet, different taxa have different environmental optima and tolerances (Mannion, 1982) Since this optima and tolerance is usually well defined and narrow, diatoms are very responsive to changing

environmental conditions (Moser et al, 1996)

4 Short lag time: Diatoms have short life cycles of approximately two weeks (Korhola, 2007) They also migrate rapidly and are able to colonise a habitat quickly (Smol, 2008) This means that they will respond to any changes in the environment swiftly; a contrast to pollen analysis whereupon vegetation may take as long as centuries to be in equilibrium with climate (Tibby and Haberle, 2007)

5 Good preservation rates: Because silica is resistant to degradation, diatoms are often well preserved in various sedimentary environments (Smol, 2008)

6 Reflects local changes: While pollen analysis will provide scientists with a regional picture of the environment, diatom analysis “generally relate to the lake being studied, providing a more detailed view of change on a local scale” (O’Hara, 2000: 135)

Diatoms have been studied for approximately two centuries and began with a focus on systemic and taxonomic studies This was later supplemented with ecological data concerning habitat and environmental conditions of specific species before their paleoecological significance was recognised in the 1920s (Mannion, 1982) Their size ranges from 2µm to 1-2mm and their shape varies from round (Centrales) to needle-like (Pennales) (Crosta and Koç, 2007) The distribution of diatoms is related to a number of variables such as temperature, turbulence, light availability, pH levels, nutrient availability and salinity (Jones, 2007)

Diatoms are particularly good indicators of changing acidity levels in ecosystems and is the most widely employed technique to investigate the acidification history of a lake (Battarbee, 1984) This is because their distribution

in freshwater habitats have been shown in numerous studies, conducted since the 1930s, to be strongly correlated to pH or to factors that co-vary with pH, like

alkalinity and concentration of aluminium (Battarbee and Charles, 1987; Moser et

al, 1996) While freshwater diatom assemblages are also influenced by other

physical and chemical factors, in particular salinity and nutrient availability (Lowe and Walker, 1997), pH reconstructions have provided the most convincing results (Battarbee and Charles, 1987) Diatoms are also well preserved in acid conditions and have a high concentration in acid lake sediments (Battarbee, 1984) Thus, changes between the assemblage of old diatom samples and modern diatom samples can be used to examine whether acidification has occurred in an area (Battarbee, 1984)

The use of diatoms in paleolimnological investigations of lake acidification

is most often based on Hustedt’s classical study on the diatom flora of Java, Bali and Sumatra, conducted in the late 1930s (Battarbee and Charles, 1987) He divided diatoms into five categories based on their individual pH preferences (Lowe and Walker, 1997):

1 Alkalibiontic diatoms: occur at pH values >7

2 Alkaliphilous diatoms: occur at pH values of about 7 but with widest distributions at pH >7

3 Indifferent (circumneutral) diatoms: occur equally above and below a pH

3 pH 5-6: alkaliphilous and indifferent diatom forms are much less numerous, the frequent forms comprise up to 75% acidophilous and acidobiontic diatoms

4 pH 4-5: alkaliphilous forms have disappeared, the indifferent forms still comprise only about 20% of the frequent forms, whereas about 80% are acidophilous and acidobiontic diatoms

5 pH 4: the number of diatom forms is very small, and these are solely acidobiontic

While this characterisation is generally perceived to be somewhat inaccurate, with modern-day statistical transfer functions providing more precise pH reconstructions, it has found general acceptance and use among diatomists and can act as a guide to interpret diatom findings (Battarbee and Charles, 1987)

The use of fossil diatom assemblages to infer lake acidity history originated with Scandinavian and Swiss studies (Battarbee and Charles, 1987) For instance, a study of lake acidification in the Swedish west coast used diatom evidence to demonstrate increase acidification in Lake Stora Skarsjön over the last three decades (Mannion, 1982) In Gårdsjön, South Sweden, historical records stated that the pH of the lake in July 1949 was 6.25 When routine monitoring of the site started in 1970, the pH of the lake had already decreased

to 4.5-4.8 (Battarbee and Charles, 1987) Based on diatom reconstruction of lake acidity levels, Renberg and Hellberg (1982) were able to deduce that rapid acidification of the lake began in the 1950s, with planktonic taxa and circumneutral non-planktonic taxa decreasing and being replaced by acidophilous and acidobiontic diatoms

In the PIRLA project conducted in North America, diatoms and environmental information, including pH levels, were gathered from over 700 lakes in the eastern region and used to create calibration sets and generate transfer functions to infer pH changes in lakes from fossil diatom assemblages

(Moser et al, 1996) This study found that the pH of most lakes investigated

decreased following the commencement of acid deposition between 1850-1960 For instance, in the Adirondacks, New York, 12 lakes displayed decreasing pH between 1920 and 1970 with the fastest rate of acidification occurring during

1950 Moser et al (1996: 41) concluded that “the patterns and the extent of

acidification appeared to be largely a function of the magnitude of the acidic deposition load and the natural background pH level of the lake”

In a review of diatom analysis and lake acidification, Battarbee (1984) found that a decrease in the diatom plankton component is often a first major sign

of acidification, though the cause of this decrease is unclear At a pH of about 5.5, circumneutral diatom taxa decline and by a pH of 4.5, these are unlikely to constitute more than 10% of the diatom assemblage Between a pH of 5.0-5.5,

these circumneutral taxa are replaced by acidophilous species such as Frustulia

rhomboides, and beyond pH 5.0, the population of acidobiontic taxa gradually

expands Acidobiontic taxa are not found at pH levels above 5.5, making then a good indicator of acidity

Issues with diatom analysis revolve around the representativeness of the record Because diatom valves are light and easy to transport, sediments may contain diatoms derived from outside the lake ecosystem, brought in by streams and catchment soils (Lowe and Walker, 1997) Preservation levels also influence this representativeness Selective dissolution of diatoms would cause the death assemblages to be biased in favour of the “stronger and more heavily silicified forms” (Lowe and Walker, 1997: 177) Variables that affect diatom preservation include pH levels, salinity, temperature, silica content of the cell and the concentration gradient of dissolved silica between the sediment and overlying water (Jones, 2007) Besides preservation levels, other factors that will affect the record include the removal of sediments by erosion (Jones, 2007), the resuspension and reworking of older sediments (Jones, 2007) and grazing by

herbivores (Lower and Walker, 1997) These factors need to be taken into account in any environmental interpretations

Ultimately, lake sediments contain a vast pool of information on the extent, rate, and causes of lake acidification Diatom analysis is likely to remain the most widely used and powerful paleolimnological technique for pH reconstruction The analysis of other biological parameters and geochemical analysis can complement this data and enhance the environmental interpretations gathered (Battarbee, 1984)

3.4 Geochemical Analysis – Total Sulphur Content

Sulphur is “an essential macronutrient, the sixth most abundant element

in biomass, and also integral to many biogeochemical processes” (Bindler et al,

2008: 61) Much attention has been paid to the role of sulphur in the acidification

of surface waters worldwide Research has focussed on the contemporary biogeochemical cycling of anthropogenic sulphur in the atmosphere, soils and surface waters and how these relate of acidification, along with modelling past and future atmospheric emissions and deposition of sulphur and examining the

sulphur record in natural environmental archives like lake sediments (Bindler et

al, 2008)

Studies conducted in eastern USA have found a positive linear relationship between the emission of and atmospheric deposition of sulphur

(Charles et al, 1987) This means that sulphate concentrations in precipitation,

and thus, concentrations in freshwater ecosystems, have increased with

increasing industrial emissions (Mitchell et al, 1988) As this atmospheric

deposition of sulphur has been a cause of changes in lake acidity, the inclusion of

sulphur in paleolimnological investigations is vital (Mitchell et al, 1988)

Sulphur can enter a freshwater ecosystem either directly, from atmospheric deposition, or indirectly, through streams and groundwater inputs

(Mitchell et al, 1988) Lake and reservoir sediments act as important sinks for

sulphur, accumulating approximately 0.7 and 8.0 million tonnes per year respectively (Nriagu, 1984) This amounts to roughly 10% of sulphur released annually from fossil fuel combustion (Nriagu, 1984) In regions where lake acidification has occurred, sediments at or near the surface are often found to be enriched in sulphur as compared to deeper sediments (Nriagu, 1984) This can often be observed visually, in fresh sediment cores, by a darker colour in modern

sediments (Bindler et al, 2008) For instance, surface sediments in the Great

Lakes ecosystems are about three times enriched in sulphur (Nriagu, 1984) This enrichment has been attributed to increased inputs of sulphur from pollution sources (Nriagu, 1984)

Using total sulphur content in lake sediment as a paleolimnological indicator of acidity requires limnetic sulphur concentration to be reflected in the sulphur contents of sediment Such a relationship has been proven in some lake systems, suggesting that sulphur incorporation in lake sediments can be proportional to limnetic sulphur concentrations in lakes that exhibit similar

biogeochemical cycling of this element (Mitchell et al, 1988) There are numerous

different components of sulphur within lake sediments This includes acid volatile sulphur, elemental sulphur, pyrite sulphur, HCl-soluble sulphur, ester-sulfate, carbon-bonded sulphur and chromium-reducible sulphur With increased lake acidity resulting from increasing levels of sulphur in limnetic waters, there are several pathways for increased levels of sulphur incorporation in sediments as well These include the “sedimentation of seston and dissimilatory sulphate

reduction within the sediment” (Mitchell et al, 1988: 220)

In general, lake sediments are often deficient in sulphur and any sulphate ions that are present are quickly reduced to sulphide below the oxidised microzone As inland waters normally have a large excess of iron in the reduced state over the amount of sulphur, any sulphide ions formed are rapidly tied to the iron, making them immobile Thus, the sulphur content of sediment tends to follow closely the sulphur flux into a lake (Nriagu, 1984)

A higher limnetic sulphate concentration leads to an increase in sedimentation of organic sulphur (carbon-bonded sulphur and ester-sulfates) as both heterotrophs and autotrophs exhibit an increase in organic sulphur with

higher sulfate levels in solution (Mitchell et al, 1988) Organic sulphur is an

significant source of sulphur in lake sediments of oligotrophic, mesotrophic, eutrophic and hypereutrophic lakes, constituting as much as 80% of total sulphur

(King and Klug, 1982; Mitchell et al, 1984) In addition, organic sulphur, bound to

organic material, is diagenetically immobile (Nriagu, 1984) Furthermore, higher limnetic sulphate concentrations also accelerate dissimilatory sulphate reduction

in sediments Thus, upon acidification Lake 223 in the ELA, north-western Ontario, bacterial sulphate reduction increased two to three times over that of pre-acidification levels, resulting in an increased accumulation of iron-sulphide

compounds (Mitchell et al, 1988) Hydrogen sulphide can also be directly

incorporated into organic matter in lake sediments, contributing to their high

organic sulphur levels (Mitchell et al, 1988)

This strong correlation between sedimentary sulphur levels and limnetic

sulphur concentrations has been shown by Gorham et al (1974), in their study of

20 lakes in the English Lake District They concluded that as dissolved sulphate levels increase in a lake, sedimentary sulphur increases because “productive lakes, besides being relatively rich in sulphate, develop reducing conditions in their sediments which favour both preservation of organic sulphur compounds