TEMPERATURE MEASUREMENT AND ... - Springer Link

Environmental Monitoring and Assessment (2006) 114: 35–47 DOI: 10.1007/s10661-006-1076-7

c Springer 2006

TEMPERATURE MEASUREMENT AND STRATIFICATION IN FACULTATIVE WASTE STABILISATION PONDS IN THE UK CLIMATE KAREN LOUISE ABIS∗ and DUNCAN MARA School of Civil Engineering, University of Leeds, Leeds, UK. (∗author for correspondence, e-mail: [email protected])

(Received 27 October 2004; accepted 20 January 2005)

Abstract. Temperature measurements were taken at three pilot-scale facultative ponds located at Esholt wastewater treatment works in West Yorkshire, UK over two one-year periods. The measurements were taken at different depths using a technology called Thermochron iButton. The iButton readings were compared with temperature measurements taken by a YSI sonde probe and maximum/minimum thermometers: they were found to be within 1 ◦ C. In the temperature range 3–18 ◦ C the mean weekly and monthly air temperatures were found to be good predictors for the mean weekly and monthly pond water-column temperatures. The level of stratification in the ponds under UK climatic conditions was assessed; the data revealed that the ponds remained thermally stratified for significant periods during the year despite their shallow depth. Keywords: air temperature, measurement, stratification, UK climate, waste stabilisation ponds, water temperature

1. Introduction Water temperature is a fundamental parameter for the prediction of biological growth rates in all water bodies. It is particularly important for biological wastewater treatment systems where it significantly influences process performance, especially that of low-energy treatment technologies, such as waste stabilisation ponds (WSP) which are particularly sensitive to changes in temperature. A WSP system is a set of interconnecting shallow basins, typically 18 ◦ C the curve levelled off, with the water temperature no longer increasing in line with the air temperature. A very similar relationship was found for streams by Mohensi and Stefan (1999); at air temperatures 20 ◦ C evaporative cooling moderates the pond temperature (at these temperatures the difference between wet and dry bulb thermometer readings was ∼3 ◦ C at a relative humidity of ∼80%). Moreover the

TEMPERATURE MEASUREMENT IN UK WASTE STABILISATION PONDS

39

lower parts of the pond water column may be prevented from becoming warm by algal shading and a lack of mixing, particularly during periods of thermal stratification. The linear relationship between the air and mean water-column temperatures is useful for the prediction of the average pond temperature in climates where the mean weekly air temperature is in the range 3–18 ◦ C; within this range the mean weekly water-column temperature = 0.96 × mean weekly air temperature (r 2 = 0.99)

Figure 2. Water-column and air temperatures: (a) hourly readings from Pond A; (b) weekly mean values from all ponds; (c) monthly mean values from all ponds; and (d) hourly readings from Pond A when the temperature gradient was zero. (Continued on next page)

40

K.L. ABIS AND D. MARA

Figure 2. (Continued )

and the mean monthly pond water-column temperature = 0.91 × monthly mean air temperature (r 2 = 0.97). Since WSP in temperate climates have long retention times (typically >20 days), it is reasonable for mean monthly temperatures to be used for design. The cause of the scatter in Figure 2a is not known, but the variance was sensibly constant at all temperatures within the linear range and therefore the scatter may be visualised as the air temperature fluctuating around its weekly mean, which in turn is approximately equal to the pond water-column temperature. The scatter was not due to changes in temperature gradient as its extent was the same even when the pond was completely mixed (Figure 2d). The air temperature varied rapidly in response to changing environmental conditions, whereas the response of the pond water column temperature response was much slower; summer and winter examples of this are shown in Figure 3.


41

Figure 3. Air and water temperature readings during one week in summer and one week in winter in 2001.

3.3. P OND

STRATIFICATION

Figures 4 and 5 show the proportion of time each pond was stratified (i.e., had a temperature gradient >1 ◦ C between the surface and mid-depth) during each month

Figure 4. The percentage of time each month when the ponds were thermally stratified during June 2001–July 2002. Data missing for Pond C in May–July 2002.

42


Figure 5. Percentage of time each month when the ponds were thermally stratified during May 2003–March 2004.

for the two separate years. The bar charts show that the ponds were stratified during both years for much of the time during the summer months, rising to >80% of the time during May–July. During June 2001–July 2002 the ponds had very long retention times (A, 65 d; B, 85 d and C, 100 d). Such long hydraulic retention times might be expected to cause stagnation effects which permit long periods of stratification. However, during May 2003–January 2004, the retention times were reduced to 20–60 d, yet the percentage of time the ponds were stratified remained high. This suggests that at retention times >20 d the effect of the wastewater inflow rate was insufficient to destroy the stratification. The wastewater treatment works at Esholt is located in the valley of the River Aire. The pilot-scale ponds are located at the highest point of the works, about half way down the north side of the valley. This location is not particularly exposed but it is subject to very strong winds which are channelled through the valley. The small surface area of the ponds reduced the effective fetch of the wind across the pond surface and so may also have been a contributory factor to the long periods of stratification. The patterns of stratification during the summer were similar to those expected in warmer sunnier climates: daytime stratification followed by night-time destratification, or continuous stratification for more than a week at a time. Gu and Stefan (1995) described the Harris ponds in Minnesota as experiencing three types of stratification: Type I which was completely mixed for at least one consecutive day and night; Type II which stratified during the day and destratified at night; and Type III which was continuously stratified during day and night. The pilot-scale ponds experienced all of these three types of stratification and, in addition, during mid

43


TABLE I Number of days with specific measured stratification and mixing characteristics as defined by Gu and Stefan (1995)a Pond Type Jul 2001 Aug Sept Oct Nov Dec Jan 2002 Feb Mar Apr May Jun Totalb Percent A

I II III IV

0 12 19 0

0 18 13 0

0 26 4 0

1 2 28 21 2 5 0 0

15 3 0 13

5 9 0 15

7 5 19 19 2 7 0 0

0 0 18 14 10 17 0 0

0 35 10 197 20 99 0 28

10 55 28 8

B

I II III IV

0 11 20 0

0 13 18 0

0 27 3 0

2 2 27 17 2 9 0 0

11 4 0 16

2 12 0 15

8 3 17 22 3 6 0 0

0 0 14 27 14 4 0 0

0 28 8 199 22 101 0 31

8 55 28 9

C

I II III IV

0 10 21 0

0 12 19 0

1 27 2 0

3 3 26 18 2 7 0 0

8 4 0 19

1 11 0 17

8 2 16 20 4 8 0 0

0 15 9 0

− − − −

9 54 25 12

− − − −

26 159 72 36

a

See text for description of each type. The number of full days during the year on which measurements were taken was 359 days for Ponds A and B and 293 days for Pond C. b

winter they experienced a fourth type where there was a temperature inversion when the surface layers were cooler than the lower layers. The number of days each pond was subject to each condition is given in Table I. The pilot-scale ponds remained stratified much longer than the Harris ponds in Minnesota which are a thousand times larger and were fed with an influent jet system; both these factors would have increased the in-pond mixing potential. Table I shows that the pilot-scale ponds had Type II stratification on over half of the days and Type III on over a quarter. Only on ≤10% of days were the ponds continuously mixed. Figure 6 shows the daily range of the temperature gradient for 19 June–31 August 2001 which indicates the occurrence of Types II and III stratification. The graph also shows how high the temperature gradient could become during the day (>20 ◦ C/m). Thus, despite the temperate climate, the small ponds were highly stratified. The stratification and destratification events appeared to be related to water surface heating and cooling with air temperature. When the air temperature fell below the mid-depth pond temperature, the surface layer temperature eventually dropped until it reached the mid depth temperature and the layers became mixed. Figure 7a illustrates this heating and cooling of the surface layers (corresponding to changes in the air temperature) leading to Type II stratification between 4–10 August 2001. Figure 7b shows the period between 11–14 August when the air temperature was consistently higher than the mid-depth pond temperature and Type III stratification occurred. Between 15–18 August the air temperature started to fall and complete

44


Figure 6. The daily temperature gradient range 19 June–31 August 2001.

mixing almost occurred in the early morning. Figure 7c shows the diurnal patterns in early November 2001 when the surface temperature responded to slight increases in air temperature but the overall low temperatures allowed the surface frequently to cool to the mid-depth temperature of the pond, resulting in mixing for ∼80% of the day. This autumnal Type II pattern continued until December when the pond temperature dropped to