The temperature of a British river upstream and downstream of a heated discharge from a power station

Summary

• 1. The River Severn is 320 km long from its source to the sea. It is warmed in its lower-middle reaches by the cooling-water from Ironbridge \ldA\rd power station.

• 2. Hourly records of natural river temperature have been collected for 20 years. Temperatures in the river 2 km downstream of the cooling water outfalls were measured using a continuous recorder in 1966. Surveys showed that complete mixing of warm and cold water occurred at the recorder.

• 3. The Severn at Ironbridge is fairly swift, and shallow, turbulent rapids (A, C and E on Fig. 1) alternate with slower, deeper reaches. Current velocities varied from 0.5 to 0.8 m/sec. Daily flow rates during the period from January 1965 \2- June 1967 varied from 5.6 cu m/sec. to 480 cu m/sec. The power station normally circulated between 5% and 50% of the available river flow.

• 4. Records of natural temperature from January 1965 \2- June 1967 are analysed with respect to seasonal and short-term changes in weather conditions. Data obtained from the recorder are used to compute an equation from which \lddownstream\rd temperatures were calculated for periods when the recorder was not in use.

• 5. Four natural \ldseasons\rd are described covering periods of \ldlow\rd, \ldrising\rd, \ldhigh\rd and \ldfalling\rd temperatures. The rate of change varied from year to year but a mean weekly temperature of 10\dgC was reached in the same week (week 18) in 1965, 66 and 67. Lowest water temperatures were recorded in December and January, the highest in June, July or August. Winter temperatures varied from 0\dg to 8\dg with a mean around 4\dgC. Highest temperature over ten years was 22.8\dgC in 1957, 59 and 60, but the maximum during 1965\2-67 was 20.6\dg in 1965.

• 6. Mean air and water temperature move generally in the same direction, but in summer water temperature is consistently the higher.

• 7. Daily records show that there is a regular diurnal variation in water temperature on sunny days in early summer, probably due to the direct warming influence of the sun. No variation occurs in winter. The largest single daily increase was 2.5\dgC, though summer rainfall caused a fall in temperature of 3.5\dgC. This was due to a direct localized cooling effect of the rain. Steady river temperatures were associated with overcast skies and mean air temperatures close to that of the water.

• 8. Water temperature followed the trends of air temperature during the equinoxes, with some time lag. Other weather conditions during spring and autumn were similar.

• 9. The effect of a short summer spate was to depress the normal diurnal variation without causing a sharp change in river temperature.

• 10. Compared with small streams and a Lake District tarn, the temperature of the Severn is relatively stable, with less weekly and diurnal variation. Of the three types of water body, the tarn was generally warmer in summer and cooler in winter. Of the running waters the Severn was slightly warmer in summer and cooler in winter. Maximum stream temperatures were almost always lower than 20\dgC.

• 11. The short term effects of the cooling water on the river were to exaggerate normal summer diurnal variation by up to 100% and to establish diurnal patterns in other seasons when these were not found naturally. The greatest daily rise 2 km downstream of the station was 7.2\dgC above ambient. More normally the daily increase was 0.5\dgC to 3.5\dgC above ambient. Maximum downstream temperature was 24.5\dgC in the summer of 1965.

• 12. Long term effects were to produce certain river temperatures, e.g., 10\dgC for short periods some weeks before they occurred naturally, though during the night downstream temperatures fell to ambient. If degree-hours above 0\dgC were totalled over the year there were increases in the downstream reaches of some 29% in 1965 and 11% in 1966. Arbitrary total numbers of degree hours could be reached some 2\2-5 weeks earlier downstream than upstream. The rate of accumulation of degree hours varied with season and the effects of the cooling water on this \ldrate\rd were more marked in winter than in summer.

• 13. The stability of the natural water temperature in the River Severn compared with that of Hodson's Tarn and small streams, means that single daily measurements or weekly maximum and minimum recordings can be of considerable value in determining temperature patterns.