Water, Air, & Soil Pollution

, 225:2090 | Cite as

Evaluation of River Water Quality: A Case Study of the Lea Navigation (NE London)

Article

Abstract

The Lea Navigation in the north-east of London, a canalized reach of the River Lea, is affected by episodes of very low levels of dissolved oxygen. The problem was detected by the Environment Agency from the confluence with Pymmes Brook (which receives the final effluent of Deephams sewage treatment works) to the Olympic site (Marshgate Lane, Stratford). In this study, possible causes and sources of the poor water quality in the Lea Navigation were investigated using algal bioassays and detailed spatial seasonal mapping of the physico-chemical parameters collected in situ. Results showed chronic pollution and identified polar compounds in the river water and high bacterial concentrations as possible causes of low dissolved oxygen levels. This study confirmed the negative impact of Deephams sewage treatment works (via Pymmes Brook) on the water quality of the Lea. However, whilst the Environment Agency had previously focused on the pollution created by the sewage treatment works, results showed evidence of other sources of pollution; in particular, Stonebridge Brook was identified as an uncontrolled source of pollution and untreated wastewater. This study demonstrates the value of conducting combined methodologies and detailed monitoring. Other possible sources include Old Moselle Brook, diffuse pollution from surface run-off, boat discharges and other undetected drainage misconnections.

Keywords

Lea Navigation River water quality Bioassay Spatial digital maps Chronic pollution Wastewater 

1 Introduction

In 2000, the European Water Framework Directive (WFD) established that all European waters should be in good conditions by the 2015. However, after 14 years only, 53 % of the European surface waters have a good ecological status (European Commission 2014). The goal could be difficult to achieve in an urban context where the water bodies are exposed to diffuse pollution resulting from a range of sources that are generally difficult to detect, determine and manage (DEFRA 2012). The restoration of a good water condition becomes even more difficult when the watercourse is a heavily modified water body (HMWB) as a result of physical changes from human activity. Urban water pollution involves a wide variety of pollutants, often of unknown nature (emergent contaminants). Chemical analysis offers a quantitative measurement of selected chemicals in the water, but it does not identify the presence of non-target pollutants (e.g. toxic metabolites). In addition, chemical analyses do not give any indication about toxicity due to interactions between compounds (synergistic/additive effects). A recent report of the European Commission stated that the chemical status is unknown for 40 % of the surface waters, indicating that the monitoring is not adequate (European Commission 2014).

A part of the diffuse pollution, which affects urban water bodies, is due to bacterial contamination coming from wastewater treatment plants, surface run-off and combined sewer overflow. A good wastewater-specific marker appears to be caffeine (Buerge et al. 2006) which is mobile and is highly eliminated through wastewater treatment plants, but with a slow degradation rate in the environment of between 3 days and 3 months (Sauvé et al. 2012) and has a high detection frequency (Hillebrand et al. 2012). Caffeine is exclusively human-specific, regularly consumed by people with coffee, tea, soft drinks, chocolate and pharmaceuticals.

In the Directive 2008/105/EC, the Council of the European Communities highlighted the importance of combining data on chemical concentration and toxicological evaluation for integrated assessment of the status (chemical and ecological) of water bodies. As a result, microbial bioassays are commonly used in ecotoxicological assessments of surface waters in line with the guidelines for the testing of chemicals (OECD 2006). They are also required in other European legislation, such as Registration, Evaluation, Authorisation and restriction of Chemicals (REACH) for the evaluation of the environmental hazard of chemicals. Moreover, bioassays can assist long-term monitoring, showing the trend in the water quality across different sampling sites over many years (Belkin 2003; Struijs et al. 2010).

Planktonic microalgae have been chosen as test organism for the following reasons (Lewis 1998; Källaqvist et al. 2008; Silva et al. 2009; Aruoja 2011): (1) In aquatic ecosystems, algae occupy the first level of the food chain, as primary producers. This implies that disturbances in their productivity and structure may provoke changes in the structure of higher ecosystem levels; (2) they are easy to grow and maintain in the laboratory. Their short life cycle allows the measurement of toxic effect over many generations; (3) they are sensitive to modification of their environment, and they have been shown to be responsive to a wider range of contaminants than invertebrates and fishes; (4) algal tests are characterized by high reliability, reproducibility and robustness.

Microalgal bioassays have been used in many different situations to test for toxicity and detect pollution: (1) laboratory wastewater (Silva et al. 2009); (2) toxicity testing of heavy metals, pesticides, pharmaceuticals (Isidori et al. 2005; Tišler et al. 2009; Daus et al. 2010; Köck et al. 2010; Santos et al. 2010; Roberts et al. 2010); and (3) chronic toxicity of river water, organic sediment extracts and sediment pore water to aquatic organisms (Källqvist et al. 2008).

The aim of the current study was to present a case study of a polluted HMWB, identifying likely sources of water contamination and providing details of water quality over a 2-year period using bioassays. At the same time, the river water was investigated by monitoring physico-chemical parameters in situ, in order to achieve a more detailed picture of the area under investigation and to produce spatial maps, presenting results understandable by a wider audience.

2 Materials and Methods

2.1 Sampling Area

The River Lea (or River Lee) is a major left-bank tributary of the Thames. It rises from springs in north Luton (Leagrave Common) and joins the River Thames near Bow, flowing through Bedfordshire, Hertfordshire and Greater London. Between the 1600s and the mid-1800s, a length of the River Lea was canalized (from Hertford to the River Thames) to allow navigation, taking the name of Lea Navigation. The surrounding area is mainly urban, and the water quality is affected by different factors including navigation, abstraction, misconnections and water run-off from roads and nearby areas. In addition, several sewage treatment works (STW) discharge their final effluents into the River Lea. A large contribution to the pollution of the Lea Navigation is from Deephams STW at Edmonton, which discharges final effluent into Lea water through Pymmes Brook. The area of the Lea under investigation, from the confluence with Pymmes Brook to Lea Bridge Road, was selected as it was identified by Environment Agency as the most polluted. In particular, the dissolved oxygen level is almost persistently low in the stretch of the Lea Navigation downstream of Pymmes Brook, suggesting chronic pollution (Snook and Whitehead 2004).

Six sampling sites were selected in order to investigate both the in-stream variation of the water quality and the discharges along the Lea Navigation downstream of the confluence with Pymmes Brook (Fig. 1). The choice of the sampling stations was based on preliminary algal tests (Sect. 3.1) and the in situ physico-chemical data (Sect. 3.3), in particular where there were low dissolved oxygen levels. River water was collected from (Fig. 1): (1) Lea Navigation, at Tottenham Hale upstream of Tottenham Lock; (2) Pymmes Brook, at the confluence with Lea Navigation; (3) Lea Navigation at Springfield Park, downstream of the Marina; (4) Lea Navigation, at the confluence with Stonebridge Brook; (5) Lea Navigation, at Lea Bridge weir; and (6) River Lea, downstream of the weir at Hackney Marshes, located outside the channel, in a natural river bed.
Fig. 1

Map of the six sampling sites. Five stations were located along the Lea Navigation from Tottenham Hale, upstream of Pymmes Brook, to the Lea Bridge weir. One station was located downstream of the weir, in the River Lea

The Tottenham Hale site was located upstream of the confluence of the Lea Navigation and Pymmes Brook. Generally, when monitoring the effects of inflow water in a receiving body, the water upstream of the incoming water is considered as control. Tottenham Hale monitoring station was chosen as the control site for four reasons: (1) It was located upstream of Pymmes Brook; (2) it presented the same physical characteristics of the sites downstream of Pymmes Brook; (3) it was easily accessible; and (4) at this site, the water quality was better than the water quality downstream of Pymmes Brook, as described by historic data presented on the Environment Agency website. The good quality of the river water at Tottenham Hale was further confirmed by ecotoxicity analysis (Sect. 3.1, Fig. 2) and the collection of in situ physico-chemical parameters (Tables 1, 2, 3 and 4) conducted in this study.
Fig. 2

Box plot representing the results of P. subcapitata growth inhibition test conducted with water collected on 06 September 2010. Box represents middle two quartiles; the horizontal line in the box represents the median. The ends of the whiskers represent the minimum and maximum data points. The level of inhibition was calculated with respect to the algal growth in the OECD medium (control). T Tottenham Hale, P Pymmes Brook, S Springfield Park

Table 1

Physico-chemical parameters collected in situ on the 22 August 2011 at 23 sites in the area under investigation

ID sampling site

Time (h)

Temperature water (°C)

Conductivity (μS/cm)

Dissolved oxygen (%)

Dissolved oxygen (mg/L)

pH

Total ammonia (mg/L)

0—Tottenham Hale

12:28

19

761

119

11.0

8.37

0.49

1—Pymmes Brook

12:20

21

1,027

66

5.9

7.38

1.13

 2

13:46

21

993

67

5.9

7.40

0.98

 3

12:09

20

993

57

5.1

7.36

1.14

4—Old River Lea

13:26

20

645

115

10.4

8.82

0.38

 5

11:48

20

964

42

3.8

7.28

1.19

 6

11:38

20

937

43

3.9

7.30

1.13

 7

11:24

19

892

48

4.4

7.36

0.99

8—Old Moselle Brook

11:13

19

879

46

4.2

7.36

0.98

 9

11:04

19

891

49

4.5

7.37

0.96

10—Stonebridge Brook

10:54

19

884

48

4.4

7.38

0.96

 11

10:41

19

882

48

4.4

7.37

0.93

 12

10:23

19

911

45

4.1

7.30

0.99

 13

10:16

19

912

45

4.1

7.30

1.01

 14

09:46

19

913

44

4.0

7.28

1.03

 15

09:56

16

738

28

2.8

7.25

0.44

16—Springfield Park

09:34

19

911

43

4.0

7.28

1.03

 17

14:26

20

895

43

3.9

7.28

1.04

 18

14:44

20

904

41

3.7

7.28

0.98

 19

15:03

20

910

39

3.5

7.28

0.98

20—Lea Bridge weir

15:14

20

911

37

3.3

7.28

0.96

 21

15:25

20

883

28

2.5

7.30

1.01

22—River Lea

15:42

20

907

57

5.2

7.36

0.95

Table 2

Physico-chemical parameters collected in situ on the 31 October 2011 at 23 sites in the area under investigation

ID sampling site

Time (h)

Temperature water (°C)

Conductivity (μS/cm)

Dissolved oxygen (%)

Dissolved oxygen (mg/L)

pH

Total ammonia (mg/L)

0—Tottenham Hale

09:43

13

900

87

9.2

8.13

1.11

1—Pymmes Brook

10:20

18

1,156

48

4.5

7.26

2.19

 2

12:20

18

1,156

53

5.0

7.28

1.72

 3

12:26

18

1,162

53

5.0

7.27

1.61

4—Old River Lea

09:56

14

795

84

8.8

7.93

0.97

 5

12:14

17

1,089

45

4.4

7.35

1.93

 6

12:08

17

1,090

44

4.3

7.35

1.91

 7

12:00

17

1,073

44

4.2

7.33

2.06

8—Old Moselle Brook

11:44

16

1,043

40

3.9

7.34

2.01

 9

11:39

16

1,024

32

3.2

7.37

2.06

10—Stonebridge Brook

11:30

16

1,014

20

2.0

7.38

2.17

 11

11:25

16

1,057

37

3.6

7.30

2.22

 12

11:16

17

1,086

34

3.3

7.26

2.32

 13

11:07

17

1,092

33

3.2

7.24

2.32

 14

10:49

17

1,089

34

3.3

7.25

2.48

 15

10:56

15

1,010

19

1.9

7.27

1.47

16—Springfield Park

12:57

17

1,079

34

3.3

7.27

2.05

 17

13:21

17

1,093

32

3.0

7.26

2.14

 18

13:29

17

1,097

32

3.1

7.26

2.05

 19

13:40

17

1,097

31

3.0

7.27

1.09

20—Lea Bridge weir

13:50

17

1,093

30

2.9

7.28

1.62

 21

13:55

17

1,089

30

2.9

7.31

1.45

22—River Lea

14:12

17

1,090

62

6.0

7.42

1.41

Table 3

Physico-chemical parameters collected in situ on the 09 January 2012 at 23 sites in the area under investigation

ID sampling site

Time (h)

Temperature water (°C)

Conductivity (μS/cm)

Dissolved oxygen (%)

Dissolved oxygen (mg/L)

pH

Total ammonia (mg/L)

0—Tottenham Hale

09:42

4

918

92

12.0

8.02

1.64

1—Pymmes Brook

09:51

10

1,125

57

6.3

7.26

2.77

 2

10:14

10

1,097

48

5.4

7.20

2.86

 3

10:07

10

1,101

52

5.8

7.22

2.75

4—Old River Lea

11:07

4

781

99

13.0

8.03

1.38

 5

11:34

10

1,091

47

5.3

7.21

2.45

 6

11:40

9

1,063

56

6.4

7.27

2.41

 7

11:50

9

1,070

53

6.0

7.24

2.50

8—Old Moselle Brook

11:56

9

1,026

55

6.4

7.29

2.41

 9

12:02

8

1,008

54

6.4

7.28

2.55

10—Stonebridge Brook

12:07

8

1,015

57

6.6

7.28

2.58

 11

12:15

9

1,019

55

6.3

7.26

2.55

 12

12:22

8

1,013

56

6.6

7.28

2.51

 13

13:05

8

1,007

53

6.2

7.27

2.42

 14

13:11

8

995

56

6.6

7.29

2.38

 15

13:19

6

965

30

3.6

7.29

2.01

16—Springfield Park

13:29

8

1,000

54

6.3

7.25

2.41

 17

13:37

9

1,027

43

5.0

7.18

2.65

 18

13:48

9

1,039

41

4.8

7.17

2.78

 19

14:21

9

1,047

40

4.5

7.16

2.69

20—Lea Bridge weir

14:30

9

1,054

34

3.9

7.14

2.79

 21

14:36

9

1,054

34

3.9

7.15

2.69

22—River Lea

14:50

9

1,052

69

8.0

7.28

2.64

Table 4

Physico-chemical parameters collected in situ on the 23 April 2012 at 23 sites in the area under investigation

ID sampling site

Time (h)

Temperature water (°C)

Conductivity (μS/cm)

Dissolved oxygen (%)

Dissolved oxygen (mg/L)

pH

Total ammonia (mg/L)

0—Tottenham Hale

10:20

11

817

136

14.6

8.89

1.25

1—Pymmes Brook

10:28

14

1,900

50

5.0

7.31

1.83

 2

10:38

14

1,003

42

4.3

7.25

1.67

 3

10:33

14

1,001

43

4.2

7.21

1.62

4—Old River Lea

10:57

12

830

90

9.6

7.96

0.80

 5

11:29

13

921

51

5.3

7.45

1.36

 6

11:34

13

928

53

5.4

7.43

1.15

 7

11:38

13

977

44

4.4

7.32

1.25

8—Old Moselle Brook

11:44

11

659

25

2.7

7.39

0.66

 9

11:48

12

873

40

3.8

7.46

0.50

10—Stonebridge Brook

11:52

12

931

36

3.7

7.45

0.41

 11

12:02

13

950

48

5.0

7.41

0.79

 12

13:04

13

955

46

4.7

7.40

1.31

 13

13:11

13

953

39

4.0

7.38

1.12

 14

13:18

13

955

37

3.8

7.36

1.12

 15

13:24

11

838

58

6.2

7.52

0.68

16—Springfield Park

13:31

13

964

32

3.3

7.33

1.08

 17

13:40

13

972

33

3.4

7.33

1.19

 18

13:50

13

978

30

3.0

7.31

1.25

 19

14:29

13

984

27

2.8

7.28

1.57

20—Lea Bridge weir

14:40

13

987

25

2.5

7.27

1.31

 21

14:44

13

956

22

2.3

7.27

1.08

22—River Lea

15:05

13

986

51

5.3

7.38

1.15

Measurements were performed after some days of rainfall, and it was raining also during the sampling

2.2 Sample Collection

The containers used to collect the water were sterilized, screw top, glass bottles. Samplings were performed from a footbridge where possible, collecting river water from the centre of the channel. Alternatively, samples were collected from water at least 1 m from the bank, with the bottle attached to a telescopic rod. In both cases, the samples were collected at least at 20 cm below the surface of the water. The bottles were rinsed three times with the river water and then filled to the top to avoid diffusion of volatile compounds into any air space. Each bottle was labelled with date, time and location. The water samples were brought to the laboratory and kept in the dark at 4 °C until analysis; all the tests were conducted within 48 h from the sample collection, according to Environment Agency guidelines (2008). The river water collection was always conducted at the exact same six stations over all the monitoring period.

2.3 Culture of Pseudokirchneriella subcapitata

New algal cultures were set up every week to ensure healthy algal populations. The green microalga Pseudokirchneriella subcapitata (formerly known as Selenastrum capricornutum, CCAP 278/4) was obtained from the Culture Collection of Algae and Protozoa, Ambleside, Cumbria, UK. Three days before the start of the test, new cultures were established by adding 1 ml from a 1-week old algal culture (with optical density ≥ 1 at 550 nm) to 50 ml of 3 N-BBM + V medium (Bold’s Basal Medium with threefold nitrogen and vitamins). The optical density was measured with an Eppendorf BioPhotometer plus at 550 nm, using semi-microcuvettes (PS, one styrofoam, Fisher Brand). Cultures were set up in sterile glass conical flasks, plugged with sterile, non-absorbent cotton wool to allow aeration. All the algal cultures were incubated in a shaking incubator at 150 rpm (Model G25, New Brunswick, Scientific Co. Inc, Edison, NJ, USA) under continuous “cool white” fluorescent illumination and at a constant temperature of 23 ± 2 °C (OECD 2006).

2.4 Algal Growth Inhibition Test with P. subcapitata

Tests were performed following the guidelines stated by the Organization for Economic Co-operation and Development (OECD 2006) and the Environment Agency protocol (Environment Agency 2008). River water samples were first pre-treated by centrifugation at 3,500 rpm and a rotor radius of 12 cm for 15 min in order to remove suspended solids; then, all test solutions were supplemented with the same concentrations of OECD nutrient medium (OECD 2006). Investigations were performed in 100-ml conical glass flasks containing 25 ml of the test sample, with at least four replicates for each testing sample. Flasks were capped with cotton air-permeable stoppers. The pH should increase between 0.5 and 1.5 units, as reported by OECD guidelines (2006). Finally, an algal inoculum was added to give a starting optical density (OD) of 0.05 at 550 nm (corresponding to 12 × 104 cells/ml). Test vessels were placed in an incubator shaker at 23 ± 2 °C, under continuous cool white fluorescent light, whilst being shaken at 150 rpm. The test duration was of 72 ± 4 h, and the OD was checked every 24 h. The cell density values were then converted into cell concentrations by the following experimental equation:
$$ \mathrm{Cell}\ \mathrm{concentration}\ \left(\mathrm{cell}/\mathrm{ml}\right)=\left(2,496,759\times {\mathrm{OD}}_{550}\right)+4,224 $$
(1)

Inhibition of growth was determined by comparing the optical densities of the environmental samples with those of the controls.

2.5 Solid Phase Extraction

To further investigate the nature of the compounds responsible for inhibition of algal growth, some river water samples were treated to separate polar from non-polar pollutants by solid phase extraction (SPE). SPE was performed using silica-based packing (ENVI-18, 5 g, SUPELCO), which allowed a reversed phase separation, with a polar or moderately polar sample matrix (mobile phase) and a non-polar stationary phase. Organic analytes from polar mixture (as water) are retained because of non-polar–non-polar attraction forces (Van der Waals forces). The water samples were previously centrifuged for 10 min at 3,500 rpm to remove particles and avoid the column blockage. According to the manufacturer’s extraction protocol (Charlton Scientific, Independent Laboratory Suppliers, http://www.charltonsci.co.uk/), the column was first conditioned with 5 ml of methanol (Sigma 154903) and then washed with 5 ml of distilled water. Then, 100-ml river water samples were passed though the column: The polar compounds were drawn out of the cartridge, while the non-polar compounds were retained in the silica bed. All the 100-ml samples were collected to test the toxicity of the polar river sample fraction. Finally, the non-polar analytes were eluted with 5 ml of acetone (Sigma 34480), a non-polar solvent. After acetone evaporation by heating and stirring, the solutes collected were re-suspended in 100 ml of distilled water, and this solution was used to test the non-polar river sample fraction toxicity by algal growth inhibition test.

2.6 In Situ Physico-Chemical Parameter Monitoring

Physico-chemical parameters were determined at several locations along the river using a multiparametric probe (YSI 6820), provided by the Environment Agency. Data were collected approximately every 50 m from Tottenham Hale site to Hackney Marshes, also sampling in the proximity of likely working discharges. The final dataset included 23 sampling stations.

Measurements were carried out by dipping the probe at a distance of 1.5 m from the bank (left or right depending upon the accessibility) or from the middle of footbridges where possible. Only one measuring point in the middle of the water column was taken.

The parameters monitored were as follows: (1) dissolved oxygen (mg/l and %, optical sensor), (2) pH, (3) conductivity (μS/cm), (4) total ammonia (mg/l) and (5) temperature (°C). The probe was calibrated by the EA and according to the manufacturer’s instructions. The dissolved oxygen sensors were calibrated in the field before data recording.

Surveys were conducted seasonally starting from summer 2011 for 1 year. Specifically, the physico-chemical parameters were collected on 22 August 2011 (summer), 31 October 2011 (autumn), 09 January 2012 (winter) and 23 April 2012 (spring).

The data collected in the field were used to produce a series of spatial maps of the stretch of the Lea under investigation. The data were interpolated with the inverse distance weighted algorithm. Using the following equation, a value can be predicted in an un-sampled location using the nearby samples:
$$ u={\displaystyle \sum_{i=0}^N\frac{w_i\cdot {x}_i}{{\displaystyle {\sum}_{j=0}^N{w}_i}}} $$
(2)
where u is a value in an un-sampled location, xi is a sample at the ith location and wi is the weight assigned to each sample point. The weights are assigned based only upon the distance of each sample to the point to be estimated, according to the following equation:
$$ {w}_i=\frac{1}{d{\left( u;{x}_i\ \right)}^p} $$
(3)

where d(u; xi) is the distance between the sample xi and the un-sampled location u and p is the power parameter. In this study, a power of 2 was used, because it is the most common power function employed in surface mapping and it gives higher weight to the surrounding points (compared with the weights of distant values), presenting a good level of closeness to the observed data (Schloeder et al. 2001; Perry and Hollis 2005). With this interpolation method, it was possible to estimate a regular grid of points with a horizontal resolution of 10 m from Tottenham Hale to the natural part of the river after Lea weir. The results of this study are a series of maps, for each measured physico-chemical property and for each sampling survey, i.e. one map for each season.

3 Results and Discussions

3.1 Preliminary P. subcapitata Inhibition Growth Tests

Preliminary algal growth tests were conducted five times during autumn 2010, using fresh river water samples each time from three monitoring sites, around the confluence with Pymmes Brook. OECD nutrient medium was used as control. All the five tests showed high algal inhibition growth after 24 h at Pymmes Brook and Lea Navigation at Springfield Park, compare to the algal growth in the nutrient medium. However, by the end of the test (after 72 h), algal growth had recovered, which could be explained by absorption phenomena due to low toxicant levels and small test solution volumes (OECD 2006). Nevertheless, after 24 h, the inhibition was evident, indicating distress for the P. subcapitata. Figure 2 shows an example of algal growth test results.

Moreover, Fig. 2 shows evidence that water sample collected from Tottenham Hale exhibited little or no algal inhibition and the percentage of algal growth in this water sample was very similar to the growth in the nutrient medium. For these reasons, Lea Navigation at Tottenham Hale was set as control site for the future tests.

3.2 P. subcapitata Inhibition Growth Tests

Algal growth inhibition tests were conducted at Tottenham Hale, Pymmes Brook and Springfield Park over a period of almost 2 years. Samples were taken from September 2010 to July 2012.

The algal growth at Tottenham Hale was around 1.3 units per day, and it was considered as control to evaluate the level of inhibition (%). Since the highest percentage of inhibition happened after the first day of test, only results obtained after 24 h are given. Figure 3 shows the level of inhibition detected in Pymmes Brook and Springfield Park water samples during the period of investigation. Pymmes Brook water was not sampled on 18 July 2011, due to inaccessibility of the sampling site. Results did not show any seasonal trend in the two stations. Moreover, the level of inhibition at both the sampling sites was not higher than usual during summer, the period when the river water level would normally be low and composed mostly by STW discharge. The tests did not give any evidence of a clear pattern regarding which sampling station was the most polluted. Pymmes Brook and Springfield Park did not present any significant differences in the level of inhibition. In addition, changes in levels of inhibition were detected in samples collected in consequent weeks, indicating that the pollutant load coming from the STW was not constant.
Fig. 3

Box plot representing P. subcapitata level of inhibition (%) in water samples collected at Pymmes Brook and Springfield Park, after 24 h. The percentage of inhibition was calculated with respect to the growth in Tottenham Hale water samples (control). Box represents middle two quartiles; the horizontal line in the box represents the median. The ends of the whiskers represent the minimum and maximum data points. The algal growth differs statistically (*p < 0.05, t test) between the two stations. White Pymmes Brook, grey Springfield Park

From October 2011 to July 2012, river water samples were collected from the Lea Navigation at the confluence with Stonebridge Brook, at Lea Bridge weir, and from River Lea at Hackney Marshes. The result shown in Fig. 4 showed that Stonebridge Brook was a major source of pollution, in particular in two monitored events when the level of inhibition was higher than 100 % (30 January 2012 and 16 July 2012), indicating a lower algae density compared to the algal aliquot added at time 0 to the river water sample (Fig. 4a). Throughout the sampling period, the discharge from Stonebridge Brook presented a whitish colour, and an odour of raw sewage was identified at this station during the last two sample collections, suggesting that Stonebridge Brook receives either water from combined sewer overflows (CSOs), misconnections or both.
Fig. 4

Box plots representing P. subcapitata level of inhibition (%) after 24 h, in water samples collected from Lea Navigation at Stonebridge Brook (a), at Lea Bridge weir (b) and from River Lea at Hackney Marshes (c). The percentage of inhibition was calculated with respect to the growth in Tottenham Hale water samples (control). Box represents middle two quartiles; the horizontal line in the box represents the median. The ends of the whiskers represent the minimum and maximum data points

The levels of inhibition at Lea Bridge weir (Fig. 4b) and in the River Lea at Hackney Marshes (Fig. 4c) were around the 40 %, with highest inhibition level slightly greater than 50 % compared to the algal growth in Tottenham Hale waters.

At each monitoring station, a decrease in the inhibition values was noticed for the last two samples (2 July 2012 and 16 July 2012), possibly due to a dilution effect due to heavy rainfall. However, the percentage of inhibition at Stonebridge Brook site was high even on those two sampling days.

The algal tests conducted along the Lea Navigation showed an almost constant level of inhibition in this part of the channel over the 2 years of investigation. There was evidence that the pollutants were dissolved in the water column, since the algal assessments tested the river water. The chronic levels of pollution detected by the algal growth inhibition tests gave an indication of the impact contaminants present in the river water would have on the primary producers in the channel (algae and aquatic plants) which would have contributed to the harmful reduction in dissolved oxygen levels.

3.2.1 Investigation of Polar and Non-Polar River Water Fractions

Algal growth inhibition investigations were carried out with SPE pre-treated waters and untreated river samples in parallel, using river water samples collected from Lea Navigation at Tottenham Hale and from Lea Navigation at Springfield Park. The P. subcapitata growth in the OECD medium was used as control. Since acetone showed some inhibition of the algal growth over 72 h, it was evaporated by stirring and heating. The methanol appeared to have little effect on the algal population, since the algal cells in Tottenham Hale polar water fraction showed just a small level of inhibition (Fig. 5).
Fig. 5

Example of results of algal growth inhibition test conducted with river water pre-treated by solid phase extraction. The water samples were collected on 09 May 2011 from the Lea Navigation at Tottenham Hale and at Springfield Park. The percentage of inhibition was calculated with respect to the growth in the medium (control). Box represents middle two quartiles; the horizontal line in the box represents the median. The ends of the whiskers represent the minimum and maximum data points. T Tottenham Hale, TP Tottenham Hale polar fraction, TNP Tottenham Hale non-polar fraction, S Springfield Park, SP Springfield Park polar fraction, SNP Springfield Park non-polar fraction

The test was repeated twice by analyzing new water samples each time (collected on 11 April 2011 and 09 May 2011). Results from the two repetitions showed similar levels of inhibition. There was evidence that the polar water fraction at Springfield Park was negatively affecting the algal population (Fig. 5), since it showed the highest inhibition level compared to the other samples. Moreover, the inhibition percentage in Springfield Park polar fraction was much higher than the level of inhibition in Tottenham Hale polar fraction, indicating that polar pollutants were the likely major cause of negative effects on the algal population. Inhibition was detected also in the non-polar water fraction of both Tottenham Hale and Springfield Park samples, but the level of inhibition was smaller than in Springfield Park polar fraction.

3.3 Spatial Monitoring of Physico-Chemical Parameters

The collection of physico-chemical parameters in situ helped in the selection of the sampling stations. The Environment Agency had only three automated monitoring stations in the area under investigation, which were not enough to have an exhaustive picture of the water quality in that particular reach of the Lea channel. The collection of physico-chemical parameters in situ had three main aims: (1) to increase the spatial data resolution of the data collected by the three automated monitoring stations of the Environment Agency, in an effort to identify likely sources of pollution for subsequent investigation with bioassays; (2) to detect any seasonal trend in the variation of the physico-chemical data, to compare with the algal assessment results; and (3) to produce spatial maps for each physico-chemical parameter.

In this paper, only the map of the dissolved oxygen will be present as example of a spatial map (Fig. 6); however, all the values collected for each survey are showed in Tables 1, 2, 3 and 4. The colouring on the map and the classification were chosen in agreement with the Thames 21 paper (2011). The comparison between the four seasonal maps shows a clear decrease of the dissolved oxygen levels during the monitoring in April 2012. The most likely explanation is attributable to the low rainfall, in the area during April, resulting in the Lea Navigation being mainly composed of waters from Pymmes Brook, Stonebridge Brook and other unidentified inflows.
Fig. 6

Interpolated maps of dissolved oxygen levels (%) registered during four surveys. The colouring on the map and the classification were chosen in agreement with the Thames 21 paper (2011)

The dissolved oxygen levels recorded downstream of Pymmes Brook were lower than upstream of Tottenham Hale locks and at Old River Lea (Fig. 6). The low oxygen levels detected could be due to several factors such as the following: low photosynthetic activity, organic matter decomposition and nitrification by aerobic bacteria, and high chemical oxygen demand (COD). Very low oxygen levels were identified in the Lea Navigation at the confluence with Stonebridge Brook and Old Moselle Brook, indicating these two brooks as likely sources of pollution. The detection of low oxygen levels in the area of Lea Bridge weir suggested the presence of other sources of pollution, such as misconnections and/or surface run-off from the roads. The dissolved oxygen levels at Pymmes Brook were slightly higher than downstream of it, probably due to the shallower and more turbulent water flow. Higher dissolved oxygen concentrations were detected also in the River Lea downstream of the weir, which could be due both to physical phenomena, such as aeration of the water by the passage through the weir and by turbulent and shallow flow, and natural depuration of pollutants by the vegetation and microorganisms present along the banks and along the riverbed.

The temperature downstream of Tottenham Hale was higher than at upstream of the locks and at Old River Lea. The highest temperature was detected at Pymmes Brook and in Lea Navigation immediately downstream of its confluence. This could be explained by both the presence of STW discharge waters and the lower water level in Pymmes Brook, which could facilitate the heat exchange with the air.

Pymmes Brook waters were characterized by high conductivity levels and lower pH with respect to Tottenham Hale, indicating the presence of water from the STW discharge beside the contribution of surface run-off.

There was also evidence of higher total ammonia downstream of Pymmes Brook than at Tottenham Hale, Old River Lea and Coppermill, although it was not detected during all the four surveys. The equilibrium between the non-toxic ammonium ion (NH4+) and the toxic un-ionized ammonia (NH3) depends on the pH and the temperature. The toxic un-ionized form increases at higher pH and higher temperatures (Novak and Holtze 2005). At similar pH, higher levels of un-ionized ammonia are present in warmer water. Moreover, surveys showed that the highest temperatures were at Pymmes Brook, which could results in the presence of toxic un-ionized ammonia.

3.4 General Discussion

Algal growth inhibition tests were used to investigate the water quality in the Lea Navigation over a 2-year period, upstream and downstream of the confluence with Pymmes Brook, which receives water from the discharge of Deephams STW. Such a study had not previous been undertaken on this stretch of the Lea Navigation.

Results showed inhibition after 24 h, followed by algal recovery. The most likely explanation of the recovery was the absorption of compounds resulted in the almost complete absence of bioavailable toxicant in the free water volume, as a result of working with low toxicant concentrations and small test solution volumes (OECD 2006). Uptake by the algae, sorption onto the vessel walls, volatilization and degradation are described by Simpson et al. (2003) as causes of decrease of pollutants concentration during ecotoxicity test periods.

Preliminary investigations conducted in this study showed that the water collected from the Lea Navigation upstream of the confluence with Pymmes Brook (at Tottenham Hale) had little or no negative effect on the P. subcapitata population. The algal growth trend at Tottenham Hale was stable, and it was very similar to the algal growth rate in the nutrient medium, which led to its use as control during the next tests. Moreover, those results identified the channel stretch upstream of the confluence with Pymmes Brook as not being heavily polluted.

The data collected over the 2 years of the survey allowed the investigation of likely trends in the poor water quality of this particular part of the channel. However, algal growth tests results did not show any particular seasonal trend at Pymmes Brook and Springfield Park stations over the 2 years of investigation. The level of inhibition was not higher than usual during the two monitored summers, when the river water level should be low and composed mostly of STW discharge. Results from algal bioassays conducted with water samples collected at the same sampling sites during consecutive weeks showed different levels of inhibition, indicating that the level of pollution was not constant.

The tests demonstrated that Pymmes Brook and Springfield Park had similar water quality, since they showed similar level of algal inhibition. However, algal bioassays conducted with water samples collected from the Lea Navigation at the confluence with Stonebridge Brook identified this small inflow channel as another potential source of pollution. At the last two sample dates (02 July 2012 and 16 July 2012) Stonebridge Brook samples presented high levels of inhibition while the other monitoring stations showed decreased levels of inhibition. Moreover, between Pymmes Brook and Stonebridge Brook, a significant amount of water comes into the Lea Navigation through a stretch of the Old River Lea (Fig. 1), which could dilute the water from Pymmes Brook. A likely explanation of similarities in the pollution level at Pymmes and Springfield Park sites could be the combination of pollution sources (such as Stonebridge Brook, the Marina, diffuse pollution from boats and run-off) and dilution effects (Old River Lea inflow) between the two sampling sites.

Downstream of Springfield Park, two other sites were monitored: one in the Lea Navigation at Lea Bridge weir and the second downstream of the weir on the River Lea at Hackney Marshes. Algal inhibition was detected at both stations, indicating the presence of other pollution sources, since they were located several kilometres downstream of Pymmes Brook’s inflow. The inhibition level detected at the site located in the natural River Lea at Hackney Marshes suggested traces of pollution despite the presence of vegetation along the stream, which should act as a depuration filter.

In order to further investigate the nature of the inhibiting analytes, water samples (Tottenham Hale and Springfield Park) were pre-treated by SPE and separated into polar and non-polar fractions, and algal growth tests were conducted. The highest level of inhibition was detected in Springfield Park polar fraction after 24 h of testing. The level of inhibition identified in the polar fraction of the Springfield Park sample was noticeably higher than the percentage of inhibition detected in the polar fraction of the Tottenham Hale sample, indicating that polar compounds were possibly the main factor responsible for the inhibition of the P. subcapitata growth in water samples collected at Pymmes Brook and downstream of its confluence with the Lea Navigation.

Results from chemical analyses, performed separately by the Environment Agency, confirmed that the Lea Navigation at Tottenham Hale (upstream of Pymmes Brook) had better water quality than the stations located downstream of the confluence with Pymmes Brook. The two stations further downstream (Lea Navigation at Lea Bridge weir and River Lea at Hackney Marshes) showed the same concentration of organic volatile compounds and the same amount of polar pollutants identified in the other sites, but the level of faecal bacteria was lower, indicating a likely low human sewage discharge (such as misconnections) or possible dilution. In the UK, there are no regulations on the coliform levels in streams, but since Lea Navigation is used for recreation purposes such as rowing, the concentration of coliform in the Lea could be interpreted under the directive for bathing waters (76/160/EEC). Whilst the Environment Agency has been concerned about Pymmes Brook and the likely pollution from Deephams STW, both algal and chemical analyses indicated Stonebridge Brook’s uncontrolled pollution input (high BOD, coliform level and total organic volatile compound concentration) to be an equally important contributor to the poor environmental quality in this reach of the channel. In particular, high levels of faecal bacteria and caffeine were detected at this site, suggesting the presence of untreated wastewater, possibly either from combined sewer overflows (CSOs) or misconnections or both.

The data collected in situ were analyzed to investigate any seasonal trend of the physico-chemical parameters and to produce spatial maps. The detection of low dissolved oxygen levels downstream of Pymmes Brook showed that its water was affecting the Lea channel water. Also, the other parameters suggested Pymmes Brook as a source of pollution: (1) The temperature of the water was higher at Pymmes Brook and the area immediately downstream; (2) the highest conductivity values were recorded at Pymmes Brook; and (3) pH values were lower downstream of Pymmes Brook confluence than upstream at Tottenham Hale. The total ammonia levels did not vary between stations. However, it is important to keep in mind that in warmer water, there is more toxic ammonia than in cooler water, at any pH (Novak and Holtze 2005). This fact draws the attention to Pymmes and the surrounding area, where the temperature of the water was higher than in the other monitoring locations. The performance of a spatially detailed survey confirmed Pymmes Brook as source of pollution, and the study of the dissolved oxygen levels enabled the detection of two other sources of pollution: Stonebridge Brook and Old Moselle Brook. Moreover, the low dissolved oxygen concentrations registered both above and after the Lea Bridge suggested the presence of other potential sources of contamination (run-off, misconnections, etc.).

Spatial maps are an “easy-to-read” presentation of the results. The interpolation of the data resulted in having a picture of the river water quality as close to reality as possible; even at locations not investigated as sampling sites by estimating a value within two measured data, the interpolation method allowed creation of a continuous dataset from discrete values. In this study, the interpolation method helped to focus the investigation at sites more polluted than others. The presentation of the results with maps allowed an immediate reading of the river water quality, whose variations over space and the time were easily identifiable by colour changes. Spatial maps could be useful tools to display data helping people to understand the water quality in the Lea Navigation as they relate the water quality of the channel with the nearest features (streets, parks and recreational areas), raising the public concern for the area and maybe raising awareness of the need to protect it.

4 Conclusions

The low water quality of the Lea Navigation between Pymmes Brook and Lea Bridge weir was confirmed. Additionally, this project demonstrated that it was possible to achieve a clear picture of the water quality in the Lea channel by combining information obtained from different methodologies and detailed monitoring. Pymmes Brook (which receives the final effluent of Deephams STW) and Stonebridge Brook (which is source of untreated sewage) were identified as two major sources of pollution, and a contribution from the urban diffuse pollution to the water quality was detected. The low dissolved oxygen levels were likely due to both oxygen consuming by large bacteria populations and inhibition of the photosynthetic activity because of polar pollutants in the river water. Moreover, this project demonstrated an easy-to-read tool to show the results to a wide audience by using spatial maps.

The present work demonstrates the need to regularly monitor Stonebridge Brook, which is a source of uncontrolled pollution and wastewater. Moreover, efforts to decrease the level of the bacterial load in the Lea water should be undertaken. It would be useful to investigate the possibility of using the natural marshes between Springfield Park and Lea Bridge to further decrease the level of bacteria by the natural depurative activity of the vegetation. Finally, further investigations are needed to determine the nature and source of polar compounds in the Lea channel.

Notes

Acknowledgments

The authors would like to thank Mr. Peter Rudd from the Environment Agency in Hatfield for his help and assistance with provision of data and equipment.

References

  1. Aruoja, V. (2011). Algae Pseudokirchneriella subcapitata in environmental hazard evaluation of chemicals and synthetic nanoparticles. PhD thesis, Estonian University of Life Sciences.Google Scholar
  2. Belkin, S. (2003). Microbial whole-cell sensing systems of environmental pollutants. Current Opinion in Microbiology, 6, 206–212.CrossRefGoogle Scholar
  3. Buerge, I. J., Poiger, T., Müller, M. D., & Buser, H. R. (2006). Combined sewer overflows to surface waters detected by the anthropogenic marker caffeine. Environmental Science and Technology, 40, 4096–4102.CrossRefGoogle Scholar
  4. Council directive 76/160/ECC of 8 December 1975 concerning the quality of bathing water. Available at: http://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:31976L0160&from=EN. Accessed 27 May 2013.
  5. Daus, B., Weiss, H., & Altenburger, R. (2010). Uptake and toxicity of hexafluoroarsenate in aquatic organisms. Chemosphere, 78(3), 307–312.CrossRefGoogle Scholar
  6. DEFRA (2012). Diffuse pollution in water. Available at: http://www.defra.gov.uk/environment/quality/water/water-quality/diffuse-pollution/. Accessed 8 Dec 2012.
  7. Directive 2008/105/EC of the European Parliament and of the Council of 16 December 2008 on environmental quality standards in the field of water policy, amending and subsequently repealing Council Directives 82/176/EEC, 83/513/EEC, 84/156/EEC, 84/491/EEC, 86/280/EEC and amending Directive 2000/60/EC of the European Parliament and of the Council. Available at: http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:2008:348:0084:0097:EN:PDF. Accessed 9 Dec 2012.
  8. Environment Agency (2008). The direct toxicity assessment of aqueous environmental samples using the Pseudokirchneriella subcapitata freshwater algal growth inhibition test. Bristol, UK.Google Scholar
  9. Hillebrand, O., Nödler, K., Licha, T., Sauter, M., & Geyer, T. (2012). Caffeine as an indicator for the qualification of untreated wastewater in karst system. Water Research, 46, 395–402.CrossRefGoogle Scholar
  10. Isidori, M., Lavorgna, M., Nardelli, A., Pascarella, L., & Parrella, A. (2005). Toxic and genotoxic evaluation of six antibiotics on non-target organisms. Science of the Total Environment, 346, 87–98.CrossRefGoogle Scholar
  11. Källqvist, T., Milačič, R., Smital, T., Thomas, K. V., Vranes, S., & Tollefsen, K. (2008). Chronic toxicity of the Sava River (SE Europe) sediments and river water to the algae Pseudokirchneriella subcapitata. Water Research, 42(8–9), 2146–2156.CrossRefGoogle Scholar
  12. Köck, M., Farré, M., Martínez, E., Gajda-Schrantz, K., Ginebreda, A., Navarro, A., Alda, M. L. D., & Barceló, D. (2010). Integrated ecotoxicological and chemical approach for the assessment of pesticide pollution in the Ebro River delta (Spain). Journal of Hydrology, 383, 73–82.CrossRefGoogle Scholar
  13. Lewis, M. A. (1998). Freshwater primary producers. In P. Calow (Ed.), Handbook of ecotoxicology (pp. 28–50). Oxford: Blackwell Science.Google Scholar
  14. Novak, L. L. J., & Holtze, K. E. (2005). Overview of toxicity reduction and identification evaluations for use with small-scale tests. In C. Blaise & J. F. Férard (Eds.), Small-scale freshwater toxicity investigations. The Netherlands: Springer.Google Scholar
  15. OECD (Organization for Economic Co-operation and Development) (2006). Freshwater alga and cyanobacteria, growth inhibition test. Test Guideline 201.Google Scholar
  16. Perry, M., & Hollis, D. (2005). The generation of monthly gridded datasets for a range of climatic variables over the UK. International Journal of Climatology, 25, 1041–1054.CrossRefGoogle Scholar
  17. Roberts, J. F., Van Egmond, R., & Price, O. R. (2010). Toxicity of haloacetic acids to freshwater algae. Ecotoxicology and Environmental Safety, 73, 56–61.CrossRefGoogle Scholar
  18. Santos, L. H. M. L. M., Araújo, A. N., Fachini, A., Pena, A., Delerue-Matos, C., & Montenegro, M. C. B. S. M. (2010). Ecotoxicological aspects related to the presence of pharmaceuticals in the aquatic environment. Journal of Hazardous Materials, 175, 45–95.CrossRefGoogle Scholar
  19. Sauvé, S., Aboulfadl, K., Dorner, S., Payment, P., Deschamps, G., & Prévost, M. (2012). Faecal coliforms, caffeine and carbamazepine in stormwater collection systems in large urban area. Chemosphere, 86, 118–123.CrossRefGoogle Scholar
  20. Schloeder, C. A., Zimmermann, N. E., & Jacobs, M. J. (2001). Division S-8—nutrient management & soil & plant analysis. Soil Science Society of America Journal, 65, 470–479.CrossRefGoogle Scholar
  21. Silva, A., Figueiredo, S. A., Sales, M. G., & Delerue-Matos, C. (2009). Ecotoxicity tests using the green algae Chlorella vulgaris—a useful tool in hazardous effluents management. Journal of Hazardous Materials, 167(1–3), 179–185.CrossRefGoogle Scholar
  22. Simpson, S. L., Roland, M. G. E., Stauber, J. L., & Batley, G. E. (2003). Effect of declining toxicant concentrations on algal bioassay endpoints. Environmental Toxicology and Chemistry, 22(9), 2073–2079.CrossRefGoogle Scholar
  23. Snook, D., & Whitehead, P. G. (2004). Water quality and ecology of the River Lee: mass balance and a review of temporal and spatial data. Hydrology and Earth System Sciences, 8(4), 636–650.CrossRefGoogle Scholar
  24. Struijs, J., Van der Grinten, E., & Aldenberg, T. (2010). Toxic pressure in the Dutch delta measured with bioassays: trend over the years 2000–2009. National Institute for Public Health and the Environment. Available at: http://www.rivm.nl/bibliotheek/rapporten/607013013.pdf. Accessed 8 Dec 2012.
  25. Thames 21 (2011). A water quality analysis of the River Lee and major tributaries within the perimeter of the M25, from Waltham Abbey to Bow Locks. Available at: http://www.thames21.org.uk/Downloads/A%20water%20quality%20analysis%20of%20the%20River%20Lea%20and%20major%20tributaries%20within%20the%20perimeter%20of%20the%20M25,%20from%20Waltham%20Abbey%20to%20Bow%20Locks%20-Thames21.pdf. Accessed 8 Dec 2012.
  26. Tišler, T., Jemec, A., Mozetič, B., & Trebše, P. (2009). Hazard identification of imidacloprid to aquatic environment. Chemosphere, 76(7), 907–914.CrossRefGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Deborah Patroncini
    • 1
  • Fabio Veronesi
    • 2
    • 3
  • David Rawson
    • 1
  1. 1.Institute of Biomedical and Environmental Science and Technology (iBEST)University of BedfordshireGreat MarlingsUK
  2. 2.Cranfield UniversityCranfieldUK
  3. 3.Institute of Cartographic and Geoinformation, ETH ZurichZurichSwitzerland

Personalised recommendations