- Email: [email protected]
Problems with the use of ultra-violet absorption for measuring carbon compounds in a river system
14,ater Research Vol. I l, pp. 979 to 982. Pergamon Press I977. Printed in Great Britain.
PROBLEMS WITH THE USE OF ULTRA-VIOLET ABSORPTION FOR MEASURING CARBON C O M P O U N D S IN A RIVER SYSTEM C. R. C. SHEPPARD
Department of Botany, University of Durham. Durham. England (Receired 3 August 1976) Abstract--The method of Ultra-violet absorption for measuring dissolved organic compounds has been used to trace the fate of two groups of compounds in a river. The first was a group originating from the peatland catchment whose u.v. absorption was stable for at least one week. The second was a group which appeared only by a major town, which was attributed to sewage compounds. and whose absorption was short lived (less than 24 h). Interferences have been found with the method in marine and estuarine systems which appear to make the method of limited value. The interference is such that absorption is severely depressed in saline waters within the range 220-260 nm. The substance causing the interference can be concentrated using a nylon column, but loses its effect with time. INTRODUCTION AS an estimate of dissolved carbon the method of Ultra-violet absorption is thought to be superior to Chemical Oxygen demand (Banoub, 1973), and has been used increasingly frequently as a substitute for this and similar tests. Ultra-violet absorption in the range 200--350 nm is attributable in part to unshared electron pairs, eg. C = C and C = 0. linkages (Schnitzer & Khan, 1972). Several interferences are evident with this method however. The main interfering compound in fresh water is nitrate which strongly absorbs around 210 nm. In marine and estuarine systems bromide interference has been reported (Ogura & Hanya, 1967) but these also interfere at the lower end of the spectrum and can be ignored at wavelengths above 230 rim. Due mainly to the supposed extreme homogeneity of the absorbing compounds a wide range of wavelengths have been employed. Banoub (1973) measured at 260 nm; Foster & Morris (1971) used an integrated value of 250-350nm; while Ogura & Hanya (1967, 1968) used 220nm and later 230:220 and 250:220 ratios. The strong emission line of the mercury lamp at 254 nm has been used for work on sewage effluent (Dobbs, Wise & Dean 1972). Ultra-violet absorption has seldom been studied in fresh water, although Banoub (1973) used the technique with success in lake water. In estuarine systems Butler & Tibbits (1972) established that in the Tamar estuary dissolved colloids affected the u.v. absorption of the water, and that absorption was inversely proportional to salinity. In marine systems a larger amount of work has been done. Ogura & Hanya (1967, 1968) identified the various absorbing components, and found that coastal water had a higher absorption than ocean water, which was attributed to pollution. Foster & Morris (1971), however, in attempting to relate pollution to absorption in sea
water concluded that this method was limited in its use. On the North East coast of England the River Tyne drains a catchment area of about 1750 square miles, a large part of which is covered with eroding blanket peat which is a rich sout:ce of humic materials. On the estuary of the Tyne the Newcastle conurbation discharges large quantities of untreated sewage into the tidal waters. Both the upper catchment area and the city, therefore, are large sources of organic carbon, which enter the sea together. An attempt was made to distinguish the absorption of sewage from that of the natural "background" carbon. METHOD Sampling was carried out in April-June 1974. Absorption readings were measured on a Perkin-Elmer recording spectrophotometer with a pair of matched 4cm cells. Traces between 190 and 350nm were obtained, but because of the interferences from nitrate and bromide only the section above 230nm was used. With sea water samples, in which the concentration of dissolved organic materials was low, it was convenient to use an instrument amplification of x 5. For the fresh and estuarine samples the normal u 1 setting was used. Samples were filtered in the field and placed in ice until reading within six hours. Whatman 42 papers were used throughout which had been previously flushed through with 200 ml distilled water and left in their filter funnels. Such papers do not contribute absorbing or scattering materials to the samples, unlike GF/C papers, whether precombusted or not, which have been favoured by other workers. Salinity measurements were made by argentimetry, using the conversion of chlorinity into salinity given by Strickland & Parsons (1968). RESULTS
Figure 1 shows absorption traces of water samples taken from four points along the River Tyne system. Trace (a) is that of fresh spring water, which is not
979
')',~1
(~ R. (_" SHEPP~RD 1.51
1.0 ! Ecm 4cell
b
0.5
200
250
300
350
Wavelengthnm
Fig. 1. u.v. traces of stages of River Tyne. (a) Pre peat bog springs, (b) Bog water, (c) Young river, (d) River near estuary. strictly part of the river at this early point. The negligible absorption above 230 nm shows that almost no organic material was present. Trace (b) is that of a sample of water taken from the moors into which the spring water flows, and from which the River Tyne originates. A large input of organic material occurs at this point land at the same time the nitrate shown as the 205 nm peak has been removed by the vegetation). Traces (c) and (d) are absorption curves of the young and mature river respectively, and show a progressive decline in concentration of organic material downstream from the source. The whole river system is shown in Fig. 2a, which illustrates the decline in absorption in all tributaries and in the main river as the sea is approached. Figure 2b shows the relative volume of water flow at the same points on the river, on a day of average water flow. Experiments with stored samples of water from these regions showed that the dissolved compounds were stable for at least one week with regard to absorption, and since the total flow time from catchment to estuary is less than one week it is concluded that the decline in absorption is accounted for by simple dilution of the initial load from the upland catchment. The diluting water, or land run-off had little absorbing water of its own.
On entering its estuary therefore the T~ne conrained a substantial :rod measurztble amount or" stable organic carbon. This can be regarded as the natural background level, onto which is superimposed the organic compounds introduced from the built Lip areas around the estuary. A drop in absorption at all wavelengths occurred as soon as the river mixed ~ith sea water. Two factors were investigated as possible causes of this: i l) a precipitation of the absorbing compounds by salts and (2) simple dilution. Salinity measurements were used to establish the ratio of fresh water to salt water in the estuarine samples. River water of known absorbance was mixed with sea water in nine ratios between 1:9 and 9:1 and left for 5 days. Reading for absorption after this period showed that the fall in readings were accounted for entirely by dilution. Filtering the samples and re-reading them did not affect their values. At all wavelengths between 230 and 350nm, therefore, only dilution reduced the absorption, and in view of the stability of these humic substances over time and in the presence of sea water it is presumed that they flow out unchanged into the sea almost in their entirety. in the estuary it is necessary to compensate for dilution caused by tide changes. This is best achieved by eliminating the dilution factor and obtaining a corrected absorption reading which gives the absorption of the river water had it not been diluted by sea water. The following equation was used: Corrected E (E) mixture x (S)sea - (S)mixture x (E) sea (S) sea - (S) mixture where E is absorption and S is salinity. (E) sea was taken as 0.01 and (S) sea was taken as 32 ppt in all cases. Following the course of the river using the corrected readings, a substantial increase in the u.v. absorption was detected along an eight mile stretch in the middle of the Newcastle conurbation. In this region 180 out-falls discharge untreated sewage, and this region frequently becomes anoxic in warm weather (James, 1973). Random samples taken from
.89 450
t
2
]
i°oo~ 1.,oI. o
I
a Fig. 2a. Absorbance readings at different points of Tyne system. Samples taken within 4h of each other on one day in April 1974. 2b. Relative volumes of river water (x 1000ma/hl on same day.
'),~1
Measuring carbon compounds in a river system near some of the outfalls showed greatly increased reading, and it was estimated that such outfaUs were sufficient to raise the absorption by the amount seen. From this maximum value in Newcastle the corrected values declined as the mouth of the fiver was approached, despite the fact that there is no decline in industry or population. By the time the mouth of the river was reached the corrected readings were slightly less than they were at the head of the estuary. To explain this, samples were collected from three points along the estuary; from the top, the most polluted part in the middle, and from the mouth of the river. After storing for 24h at 4"C the corrected values from the central polluted region had converged with and met those from the mouth of the river, and both were progressively less than those from the top of the estuary. This indicates that the sewage is being rapidly "treated" with respect to its u.v. absorption properties, leaving only the original resistant humic compounds from the upper catchment area. That treated sewage does not have any significant absorption properties was corroborated by the fact that downstream from an efficient treatment plant (the Durham City plant on the River Wear) there was no increase in absorption although a 50% increase in total phosphorus was detected. Interferences in sea water
Immediately the humic substances were followed out to sea a change in the traces became apparent. Instead of the smoothly descending line typical of the river traces, a dip was noticed at around 230 nm, rising to a small peak at 260 nm, after which it continued in a normal uninterrupted fall (see Fig. 3). This peak and dip have been recorded before (Armstrong & Boalch, 1961) although not explained. It was further found that in over 40% of the sea water traces the dip at 240 nm extended to below the zero reference line. The possibilities that this dip was due to an artifact or an absorbing contaminant in the reference cell
1-4 = Successive column 8 ~ l ~ R in litres.
.6t °.4 240 260 Wavelength nm.
Fig. 3. Traces of original sea water and column effluent on passing seawater through nylon column.
w.R. l i l t -
(
+.1+'2~.~~.~_ E
"2cOe~rr/ ,f2
ul
220
of the twin beam instrument were eliminated. Neither changing the cells about nor reading against air, both with and without the reference cuvette in position eliminated the effect. To concentrate the absorbing constituents of sea water and to elucidate this effect, fiver and sea water samples were passed through a column packed with nylon after the method of Sieburth & Jenson (1968). River water lost about 50% of its absorption on passing through the nylon. When sea water was treated, however, the column effluent showed a higher absorption that than of the initial sea water, (Fig. 3). With both fiver and sea water, as the nylon became spent the absorption readings of the column effluents approached that of their respective untreated samples, ie. the readings of fresh water effluents rose and those of sea water fell. It would thus appear that some substance is present in sea water which has the property of reducing its absorption, and which is nylon extractable. This substance can be eluted from the column with 10°/NaOH, and comes off after a fraction containing strongly absorbing material. After neutralising, the eluate was found to produce a trace typified by a very deep trough at 240 nm and a rise at 260 nm, with the whole trace being well below the distilled water reference line. The fraction thus contained a "negatively absorbing" compound, or one which caused the absorption of the solution to be recorded as less than zero when read against both the distilled water blank or against air. It has been observed that pre-estuarine samples of river water showed very little change of absorption with time. In contrast stored samples of sea water showed that while absorption at 260 nm stayed much the same over a two week period, that at 240 nm showed a marked increase during the first four days which largely eliminated the dip at 240 nm (Fig. 4). With storage at two temperatures this increase was found to be dependent on temperature with a Qt0 of 2.0. Exposure to light had no effect. The same ageing experiments done on the nylon column eluate fractions showed that the absorbance of the second "negatively absorbing" fraction rose with time in a manner identical to the 240 trace of Fig. 4. The first strongly absorbing fraction of the eluate did not change with time.
--°2
H~
÷~
4
61-6
Days
Fig. 4. Change of absorption of 240 nm and 260 nm sections of sea water trace with time. Squares = E at 260 nm, Circles = E at 240 nm. + standard errors. Absorption is E with 4 cm cells with instrument amplification of x 5.
~'2
C . R . C . SHEPPkRZ)
DISCUSSION As absorption in the u.v. at ~avelengths larger than about _230nm is largely attributable to double bonds in organic carbon molecules (Schnitzer & Khan, 1972) this method has been limited to the gross detection of dissolved carbon. However. it has been shown that the absorption given by humic compounds from the source of the River Tyne remains stable for at least one week. while that given by untreated sewage introduced into the estuary declined rapidly and was not detected after about 24 h. In the lower and well aereated part of the estuary of this river system there was no absorption remaining attributable to the sewage. and all absorption by this point was accounted for by the humic compounds from the catchment of the river. Thus it was possible to distinguish between these two sources of dissolved carbon in the upper and middle parts of the estuary. In sea water the findings indicate the presence of two groups of compounds, one absorbing light in the normal way over the entire u.v. band, and one which has its effect mainly at 240nm. The latter has a depressing effect which superimposes on the normal curve, though this effect is lost in time. Probably the substance causing the depression of the absorption curve to below the zero line is the fluorescing and phenol staining substance found by Sieburth & Jensen (1968) when investigating chromatograms of their nylon eluates. In sea water the depression caused by this interference extends across a band of up to 50 nm, although the main effect is at 240 nm. Because of the nature and construction of most spectrophotometers fluorescence at any wavelength by a substance being irradiated with u.v. will be detected by the sensor in the test beam. Thus more energy is detected by the test beam sensor than by the blank beam's sensor and a reduced absorption is recorded. This could be corrected for in such instru-
ments b', recording fluorescence energy and making the necessar3 addition to the absorption reading. In the type of instrument which detected only specified wavelengths, however, as opposed to one (the majority) which detects any wavelength but whose light source is governed, the problem possibly does not exist. Without these corrections, however, the method of carbon estimation by u.v. absorption in sea water should be limited to wavelengths above 260 nm.
REFERENCES
Armstrong F. A. J. & Boatch G. T. (1961) The Ultra-violet absorption of sea water. J. mar. biol. Ass. U:K. 41, 59 I-7. Banoub M. W. (1973) Ultra-violet absorption as a measure of organic matter in natural waters in Bodensee. Arch. Hydrobiol, 71, 2, 159-16:;. Butler E. & Tibbits S. (1972) Chemical Survey of the Tamer Estuary. J. mar. biol. Ass. U.K. 52, 681-699~ Dobbs R. A., Wise R. H. & Dean R. B. (1972) The use of Ultra-,Aolet absorbance for monitoring the total organic carbon content of water and waste water. Water Res. 6, 1173-1180. Foster P. & Ivlorris A. W. (197[) The use of Ultra-violet absorption measurements for the estimation of organic pollution in inshore sea waters. Water Res. 5, 19-27. James A. (1973) Paper read to Challenger Society. Newcastle University 1973. Ogura N. & Hanya T. (1967) Ultra-violet absorption of sea water in relation to organic and inorganic matters. Int. J. Oceanogr. Limnol. 1, 91-102. Ogura N. & Hanya T. (1968) Ultra-violet absorbance as an index of the pollution of sea water. J. War. Poll, Control Fed. 40, 3(1), 464. Schnitzer M. & Khan S. U. (1972) Humic Substances in the Environment. Marcel Dekker Inc. New York. Sieburth J. M. & Jens0n A. (1968) Studies in algal substances in the sea. Gelbstoff (Humic materials) in terrestrial and marine waters. J. exp. mar. BiOl. Ecol. 2, 174--189.
Strickland J. D. H. & Parsons T. R. (1968) A practical handbook of seawater analysis. Bulletin 167 Fisheries Res. Bd Canada.
PROBLEMS WITH THE USE OF ULTRA-VIOLET ABSORPTION FOR MEASURING CARBON C O M P O U N D S IN A RIVER SYSTEM C. R. C. SHEPPARD
Department of Botany, University of Durham. Durham. England (Receired 3 August 1976) Abstract--The method of Ultra-violet absorption for measuring dissolved organic compounds has been used to trace the fate of two groups of compounds in a river. The first was a group originating from the peatland catchment whose u.v. absorption was stable for at least one week. The second was a group which appeared only by a major town, which was attributed to sewage compounds. and whose absorption was short lived (less than 24 h). Interferences have been found with the method in marine and estuarine systems which appear to make the method of limited value. The interference is such that absorption is severely depressed in saline waters within the range 220-260 nm. The substance causing the interference can be concentrated using a nylon column, but loses its effect with time. INTRODUCTION AS an estimate of dissolved carbon the method of Ultra-violet absorption is thought to be superior to Chemical Oxygen demand (Banoub, 1973), and has been used increasingly frequently as a substitute for this and similar tests. Ultra-violet absorption in the range 200--350 nm is attributable in part to unshared electron pairs, eg. C = C and C = 0. linkages (Schnitzer & Khan, 1972). Several interferences are evident with this method however. The main interfering compound in fresh water is nitrate which strongly absorbs around 210 nm. In marine and estuarine systems bromide interference has been reported (Ogura & Hanya, 1967) but these also interfere at the lower end of the spectrum and can be ignored at wavelengths above 230 rim. Due mainly to the supposed extreme homogeneity of the absorbing compounds a wide range of wavelengths have been employed. Banoub (1973) measured at 260 nm; Foster & Morris (1971) used an integrated value of 250-350nm; while Ogura & Hanya (1967, 1968) used 220nm and later 230:220 and 250:220 ratios. The strong emission line of the mercury lamp at 254 nm has been used for work on sewage effluent (Dobbs, Wise & Dean 1972). Ultra-violet absorption has seldom been studied in fresh water, although Banoub (1973) used the technique with success in lake water. In estuarine systems Butler & Tibbits (1972) established that in the Tamar estuary dissolved colloids affected the u.v. absorption of the water, and that absorption was inversely proportional to salinity. In marine systems a larger amount of work has been done. Ogura & Hanya (1967, 1968) identified the various absorbing components, and found that coastal water had a higher absorption than ocean water, which was attributed to pollution. Foster & Morris (1971), however, in attempting to relate pollution to absorption in sea
water concluded that this method was limited in its use. On the North East coast of England the River Tyne drains a catchment area of about 1750 square miles, a large part of which is covered with eroding blanket peat which is a rich sout:ce of humic materials. On the estuary of the Tyne the Newcastle conurbation discharges large quantities of untreated sewage into the tidal waters. Both the upper catchment area and the city, therefore, are large sources of organic carbon, which enter the sea together. An attempt was made to distinguish the absorption of sewage from that of the natural "background" carbon. METHOD Sampling was carried out in April-June 1974. Absorption readings were measured on a Perkin-Elmer recording spectrophotometer with a pair of matched 4cm cells. Traces between 190 and 350nm were obtained, but because of the interferences from nitrate and bromide only the section above 230nm was used. With sea water samples, in which the concentration of dissolved organic materials was low, it was convenient to use an instrument amplification of x 5. For the fresh and estuarine samples the normal u 1 setting was used. Samples were filtered in the field and placed in ice until reading within six hours. Whatman 42 papers were used throughout which had been previously flushed through with 200 ml distilled water and left in their filter funnels. Such papers do not contribute absorbing or scattering materials to the samples, unlike GF/C papers, whether precombusted or not, which have been favoured by other workers. Salinity measurements were made by argentimetry, using the conversion of chlorinity into salinity given by Strickland & Parsons (1968). RESULTS
Figure 1 shows absorption traces of water samples taken from four points along the River Tyne system. Trace (a) is that of fresh spring water, which is not
979
')',~1
(~ R. (_" SHEPP~RD 1.51
1.0 ! Ecm 4cell
b
0.5
200
250
300
350
Wavelengthnm
Fig. 1. u.v. traces of stages of River Tyne. (a) Pre peat bog springs, (b) Bog water, (c) Young river, (d) River near estuary. strictly part of the river at this early point. The negligible absorption above 230 nm shows that almost no organic material was present. Trace (b) is that of a sample of water taken from the moors into which the spring water flows, and from which the River Tyne originates. A large input of organic material occurs at this point land at the same time the nitrate shown as the 205 nm peak has been removed by the vegetation). Traces (c) and (d) are absorption curves of the young and mature river respectively, and show a progressive decline in concentration of organic material downstream from the source. The whole river system is shown in Fig. 2a, which illustrates the decline in absorption in all tributaries and in the main river as the sea is approached. Figure 2b shows the relative volume of water flow at the same points on the river, on a day of average water flow. Experiments with stored samples of water from these regions showed that the dissolved compounds were stable for at least one week with regard to absorption, and since the total flow time from catchment to estuary is less than one week it is concluded that the decline in absorption is accounted for by simple dilution of the initial load from the upland catchment. The diluting water, or land run-off had little absorbing water of its own.
On entering its estuary therefore the T~ne conrained a substantial :rod measurztble amount or" stable organic carbon. This can be regarded as the natural background level, onto which is superimposed the organic compounds introduced from the built Lip areas around the estuary. A drop in absorption at all wavelengths occurred as soon as the river mixed ~ith sea water. Two factors were investigated as possible causes of this: i l) a precipitation of the absorbing compounds by salts and (2) simple dilution. Salinity measurements were used to establish the ratio of fresh water to salt water in the estuarine samples. River water of known absorbance was mixed with sea water in nine ratios between 1:9 and 9:1 and left for 5 days. Reading for absorption after this period showed that the fall in readings were accounted for entirely by dilution. Filtering the samples and re-reading them did not affect their values. At all wavelengths between 230 and 350nm, therefore, only dilution reduced the absorption, and in view of the stability of these humic substances over time and in the presence of sea water it is presumed that they flow out unchanged into the sea almost in their entirety. in the estuary it is necessary to compensate for dilution caused by tide changes. This is best achieved by eliminating the dilution factor and obtaining a corrected absorption reading which gives the absorption of the river water had it not been diluted by sea water. The following equation was used: Corrected E (E) mixture x (S)sea - (S)mixture x (E) sea (S) sea - (S) mixture where E is absorption and S is salinity. (E) sea was taken as 0.01 and (S) sea was taken as 32 ppt in all cases. Following the course of the river using the corrected readings, a substantial increase in the u.v. absorption was detected along an eight mile stretch in the middle of the Newcastle conurbation. In this region 180 out-falls discharge untreated sewage, and this region frequently becomes anoxic in warm weather (James, 1973). Random samples taken from
.89 450
t
2
]
i°oo~ 1.,oI. o
I
a Fig. 2a. Absorbance readings at different points of Tyne system. Samples taken within 4h of each other on one day in April 1974. 2b. Relative volumes of river water (x 1000ma/hl on same day.
'),~1
Measuring carbon compounds in a river system near some of the outfalls showed greatly increased reading, and it was estimated that such outfaUs were sufficient to raise the absorption by the amount seen. From this maximum value in Newcastle the corrected values declined as the mouth of the fiver was approached, despite the fact that there is no decline in industry or population. By the time the mouth of the river was reached the corrected readings were slightly less than they were at the head of the estuary. To explain this, samples were collected from three points along the estuary; from the top, the most polluted part in the middle, and from the mouth of the river. After storing for 24h at 4"C the corrected values from the central polluted region had converged with and met those from the mouth of the river, and both were progressively less than those from the top of the estuary. This indicates that the sewage is being rapidly "treated" with respect to its u.v. absorption properties, leaving only the original resistant humic compounds from the upper catchment area. That treated sewage does not have any significant absorption properties was corroborated by the fact that downstream from an efficient treatment plant (the Durham City plant on the River Wear) there was no increase in absorption although a 50% increase in total phosphorus was detected. Interferences in sea water
Immediately the humic substances were followed out to sea a change in the traces became apparent. Instead of the smoothly descending line typical of the river traces, a dip was noticed at around 230 nm, rising to a small peak at 260 nm, after which it continued in a normal uninterrupted fall (see Fig. 3). This peak and dip have been recorded before (Armstrong & Boalch, 1961) although not explained. It was further found that in over 40% of the sea water traces the dip at 240 nm extended to below the zero reference line. The possibilities that this dip was due to an artifact or an absorbing contaminant in the reference cell
1-4 = Successive column 8 ~ l ~ R in litres.
.6t °.4 240 260 Wavelength nm.
Fig. 3. Traces of original sea water and column effluent on passing seawater through nylon column.
w.R. l i l t -
(
+.1+'2~.~~.~_ E
"2cOe~rr/ ,f2
ul
220
of the twin beam instrument were eliminated. Neither changing the cells about nor reading against air, both with and without the reference cuvette in position eliminated the effect. To concentrate the absorbing constituents of sea water and to elucidate this effect, fiver and sea water samples were passed through a column packed with nylon after the method of Sieburth & Jenson (1968). River water lost about 50% of its absorption on passing through the nylon. When sea water was treated, however, the column effluent showed a higher absorption that than of the initial sea water, (Fig. 3). With both fiver and sea water, as the nylon became spent the absorption readings of the column effluents approached that of their respective untreated samples, ie. the readings of fresh water effluents rose and those of sea water fell. It would thus appear that some substance is present in sea water which has the property of reducing its absorption, and which is nylon extractable. This substance can be eluted from the column with 10°/NaOH, and comes off after a fraction containing strongly absorbing material. After neutralising, the eluate was found to produce a trace typified by a very deep trough at 240 nm and a rise at 260 nm, with the whole trace being well below the distilled water reference line. The fraction thus contained a "negatively absorbing" compound, or one which caused the absorption of the solution to be recorded as less than zero when read against both the distilled water blank or against air. It has been observed that pre-estuarine samples of river water showed very little change of absorption with time. In contrast stored samples of sea water showed that while absorption at 260 nm stayed much the same over a two week period, that at 240 nm showed a marked increase during the first four days which largely eliminated the dip at 240 nm (Fig. 4). With storage at two temperatures this increase was found to be dependent on temperature with a Qt0 of 2.0. Exposure to light had no effect. The same ageing experiments done on the nylon column eluate fractions showed that the absorbance of the second "negatively absorbing" fraction rose with time in a manner identical to the 240 trace of Fig. 4. The first strongly absorbing fraction of the eluate did not change with time.
--°2
H~
÷~
4
61-6
Days
Fig. 4. Change of absorption of 240 nm and 260 nm sections of sea water trace with time. Squares = E at 260 nm, Circles = E at 240 nm. + standard errors. Absorption is E with 4 cm cells with instrument amplification of x 5.
~'2
C . R . C . SHEPPkRZ)
DISCUSSION As absorption in the u.v. at ~avelengths larger than about _230nm is largely attributable to double bonds in organic carbon molecules (Schnitzer & Khan, 1972) this method has been limited to the gross detection of dissolved carbon. However. it has been shown that the absorption given by humic compounds from the source of the River Tyne remains stable for at least one week. while that given by untreated sewage introduced into the estuary declined rapidly and was not detected after about 24 h. In the lower and well aereated part of the estuary of this river system there was no absorption remaining attributable to the sewage. and all absorption by this point was accounted for by the humic compounds from the catchment of the river. Thus it was possible to distinguish between these two sources of dissolved carbon in the upper and middle parts of the estuary. In sea water the findings indicate the presence of two groups of compounds, one absorbing light in the normal way over the entire u.v. band, and one which has its effect mainly at 240nm. The latter has a depressing effect which superimposes on the normal curve, though this effect is lost in time. Probably the substance causing the depression of the absorption curve to below the zero line is the fluorescing and phenol staining substance found by Sieburth & Jensen (1968) when investigating chromatograms of their nylon eluates. In sea water the depression caused by this interference extends across a band of up to 50 nm, although the main effect is at 240 nm. Because of the nature and construction of most spectrophotometers fluorescence at any wavelength by a substance being irradiated with u.v. will be detected by the sensor in the test beam. Thus more energy is detected by the test beam sensor than by the blank beam's sensor and a reduced absorption is recorded. This could be corrected for in such instru-
ments b', recording fluorescence energy and making the necessar3 addition to the absorption reading. In the type of instrument which detected only specified wavelengths, however, as opposed to one (the majority) which detects any wavelength but whose light source is governed, the problem possibly does not exist. Without these corrections, however, the method of carbon estimation by u.v. absorption in sea water should be limited to wavelengths above 260 nm.
REFERENCES
Armstrong F. A. J. & Boatch G. T. (1961) The Ultra-violet absorption of sea water. J. mar. biol. Ass. U:K. 41, 59 I-7. Banoub M. W. (1973) Ultra-violet absorption as a measure of organic matter in natural waters in Bodensee. Arch. Hydrobiol, 71, 2, 159-16:;. Butler E. & Tibbits S. (1972) Chemical Survey of the Tamer Estuary. J. mar. biol. Ass. U.K. 52, 681-699~ Dobbs R. A., Wise R. H. & Dean R. B. (1972) The use of Ultra-,Aolet absorbance for monitoring the total organic carbon content of water and waste water. Water Res. 6, 1173-1180. Foster P. & Ivlorris A. W. (197[) The use of Ultra-violet absorption measurements for the estimation of organic pollution in inshore sea waters. Water Res. 5, 19-27. James A. (1973) Paper read to Challenger Society. Newcastle University 1973. Ogura N. & Hanya T. (1967) Ultra-violet absorption of sea water in relation to organic and inorganic matters. Int. J. Oceanogr. Limnol. 1, 91-102. Ogura N. & Hanya T. (1968) Ultra-violet absorbance as an index of the pollution of sea water. J. War. Poll, Control Fed. 40, 3(1), 464. Schnitzer M. & Khan S. U. (1972) Humic Substances in the Environment. Marcel Dekker Inc. New York. Sieburth J. M. & Jens0n A. (1968) Studies in algal substances in the sea. Gelbstoff (Humic materials) in terrestrial and marine waters. J. exp. mar. BiOl. Ecol. 2, 174--189.
Strickland J. D. H. & Parsons T. R. (1968) A practical handbook of seawater analysis. Bulletin 167 Fisheries Res. Bd Canada.