Phosphorus analyses in humic waters- 1997 discussions

Phosphorus species

Note: The following discussions during 1997 were from an email discussion group and resulted from a question put forward by Mr. Jerry Choate of the New Brunswick (Canda) Dept. of the Environment. They are included here since the opinions of various researchers may be of considerable interest to readers. The phone numbers and email addresses have been removed to avoid spamming, etc.!



Saw your query on lakes-l. A number of years ago we were worried about the potential for commercially worked peat bogs here in Ireland to cause eutrophication via UV-photolytic release of P from humic-iron-phosphate complexes.

As part of this study we carried out a regular monthly monitoring programme of a number of peat (Sphagnum) bog outflows with very high colour on occasion. Values ranged from barely detectable to approx. 0.200mg P/l with a mean of roughly 0.05 mg P/l which we now use as an average P concentration for worked bog outflows. We assume that the P is due to mineralisation of the Sphagnum - typical aquatic plant P levels would be 0.1% P dry wt in oligotrophic situations. A dry period with active oxidisation followed by flushing of mineralised nutrients could easily lead to the type of values encountered, whereas a prolonged wet period should result in lower values.

These were all acid digested samples prior to molybdate reaction so we were not too worried about the possibility of interference. We did, however, have trouble with ordinary molybdate reactive phosphate on unfiltered samples with high colour - good blanks are important here - as even well into the red wavelengths the humics can cause significant absorption if the water is highly coloured. Hope this is of some help.

Martin McGarrigle

EPA, Ireland


Dear Jerry, I have sometimes suspected that the humics interfere with the P determination--my suggestion would be to run a series of blanks using bog or river water with the P samples to identify any interference. (Perhaps the lab commonly runs only DW blanks so wouldn't correct for color?) The second possiblity is that P is adsorbed onto the dissolved humic materials and is released during the molybdenum blue determination, and that the results are indeed correct. A bioassay with P-stressed algae may help clear this up, as the algae may also be able to extract this loosely bound P... Hope this helps--as you may have guessed, I am not a chemist but have worked with P-dynamics in a previous life!


Cheers, Jeff Thornton



I don't know if the water coloring substance interferes directly with the phosphorus-molybdenum-ascorbic acid reaction. Silica can interfere, and probably other substances. I can find you the citation for water quality testing manuals that have this information if you'd like.

Or, the colored water may be absorbing light at the same wavelength as your molyb. blue solution. This is easy to test - just put some colored water into the spectrophotometer at the appropriate wavelength and see how its absorbance compares to distilled deionized water.

Are you digesting the samples before analyzing them for TP? Digestion should remove the color, if that is the problem. However, the results may not be comparable to earlier undigested samples.

Please feel free to email me directly if you waould like to discuss this further.

Laurel Ramseyer
Michigan STate University


Are you doing total P? onlly? Try tot, soluble and soluble reactive P. ie try to find out whether you have elevated particulate or solutes. How high are the high values?

Murray N. Charlton
Project Chief, Lake Remediation
CCIW


I presume by all QA/QC that you also have acceptable recoveries of a matrix spike (the "spike" being no more than about 5 times the expected concentration in the sample)? In my experience the color in humic colored waters was removed during the digestion process. If you still have colored water following digestion, do you subtract the absorbance before adding reagents? I think Standard Methods provides guidance on dealing with colored samples. Is it possible that interfering ions are present? (e.g., arsenic). Just some random thoughts that might help. Good luck.

Standard method for the examination of water and wastewater states that color usually does not interfere at the high wavelength used (880 nm -check to see if you are running at 660 nm). Although, it does state that you may prepare a blank including all reagents except antimonyl potassium tartrate and acsorbic acid subtracting its absorbance from the absorbance of your sample.

I have also seen methods that extract phosphorus with solvents before analysis to remove interferences. As well as removing color with activated carbon before analysis.

Does the color change or is it enhanced significantly with digestion? What digestion do you use?

If we can be of further assistance or you would like another lab to run a replicate for you please contact me.

Sincerely,

Ted Lewis
Research Scientist
Center for Applied Aquatic Sciences and Aquaculture
Department of Biological Sciences
State University of New York at Brockport
Brockport, NY USA 14420



We have at times measured unexpected high total phosphorus values in heavily coloured natural waters (the associated drainage areas were nutrient poor).

I have not seen interference before but I suppose its possible. I can suggest a few things you might try. I assume you use persulfate digestion, is the sample still colored? if so try more, or a stronger oxidant with appropriate blanks and stds etc. (see Standard Methods 19th ed) Try standard additions to make sure there is no interference in the determination. Std. Methods (19th ed) states that color or turbidity generally does not interfere at the wavelenght used, but for highly colored waters make a special blank by adding all reagents except ascobic acid and the tartrate to the sample and subtract this absorbance from your regularly prepared sample absorbance. Hope this helps.

Mark D. Mattson
UMASS


Dear Jerry,

No worry with your analyses. What you observed is quite well-known from Scandinavian lakes and streams. Among other, this implies that total phosphorus (or even molybdate-reactive dissolved P) is not a very useful indicator of neither nutrient availability nor trophic status in humic waters. You may find more information quite easily:

* Meili, M. (1992): Sources, concentrations and characteristics of organic matter in softwater lakes and streams of the Swedish forest region. -Hydrobiologia 229:23-41.

* References therein.
* Other papers in the same Hydrobiologia volume.

Yours sincerely
Markus

______________________________________________________________________
Markus.Meili, Uppsala University, Inst. of Earth Sciences,
Norbyvägen 18 B, S-752 36 Uppsala, Sweden.
______________________________________________________________________



jerry -

When you say "total" P, do you mean acid-digested+K2S2O8? If organic matter is disgested, generally, most color is eliminated. But if you are simply measuring SRP via the std molybdate, then yes, high DOC will give erroneously high absorbances.

On the other hand, terrestrial DOC can contain apprecaible levels of organic-P, and will be measured in the digest.

good luck

John Wehr
Louis Calder Center
Fordham University


Jerry, The only thing I can relate is phosphorus compounds can be more soluble in anaerobic situations such as in waterlogged soils. Bogs may for some reason have a redox which may allow the normally non soluble phosphates to be placed into solution from anaerobic bottom muds or peat.
Bill Ayer



Analyses were done using the Molybdenum-Blue method with ascorbic acid as the reductant. We are wondering if water colour in this instance could interfere causing results to indicate phosphorus concentrations much higher than actually exist.

Yes, I think I did, some 10 years ago here in northern Ontario. I used to make a colour blank, in addition to the H2O-blank as follows:
Sample + "mixed" reagents without Ascorbic acid, AA was replaced by H2O.
This way, you still have the natural colour interference but you can deal with it, i.e. substract it.

Let me know if you'd like further clarification.
_________________________________________________

Gertrud Nurnberg, Ph.D.
Freshwater Research
RR 1 Baysville ON
P0B 1A0 CANADA
_________________________________________________


Re your query wrt TP concentrations in highly colored waters..... Colored waters in NS typically have higher TP concentrations than non-colored waters. TP in 276 NS lakes ranged from <.001 ug/l to .040 ug/l with mean of .005 ug/l, while color ranged from <2.5 to 300 TCU with a mean of 45 TCU. Typical TP values for colored waters are as follows:

TPColor
.012150
.005200
.007200
.018250
.018275
.017300

As you can see there doesn't appear to be a linear coorelation necessarily and I have undertaken no stats. Moreover, the highest TP concentration in this data was .040 ug/l and the associated Color was 30 TCU.

I believe Tom Claire of Environment Canada has noted variations in TOC and TP concentrations in streams draining wetlands (with highly colored waters) depending upon relative amounts of rainfall and runoff, as did one of your respondents (from Ireland ?).


Darrell



You can quickly rule out color interference by blanking against lake water without adding P analysis reagants. Also if P is high the sample after adding reagants will be very blue colored compared with original sample water.

We have at times measured unexpected high total phosphorus values in heavily coloured natural waters (the associated drainage areas were nutrient poor).

One recent case was on a small stream in northeastern New Brunswick where colour ranged between 150 and 500, hardness 6-9 mg/l, conductivity 22-23 microseimens/cm, pH 4.94-6.41, suspended solids 0-1 mg/l, ammonia <0.01 to 0.047 mg/l, nitrate & nitrite <0.05 mg/l, total organic carbon 7 to 37 mg/l, phosphorus 0.036 to 0.084 mg/l. Samples were processed in our (New Brunswick Department of the Environment) laboratory.

QA/QC has shown phosphorus results to be highly reliable. Yet, we have at times found high values in coloured waters, particularly in samples taken directly from bogs (the stream in question drains some boggy areas).

Analyses were done using the Molybdenum-Blue method with ascorbic acid as the reductant. We are wondering if water colour in this instance could interfere causing results to indicate phosphorus concentrations much higher than actually exist.

Has anyone encountered a similar situation, or have an explanation for the apparently biased results?

Jerry Choate
NB Dept. of the Environment
Federicton, NB



Phosphorus species