I used the KCl extraction method to extract nitrate from soil samples which were then analysed using a skalar continuous flow analyser. I am doing this to calculate the rate of nitrification from soil samples incubated in situ for a month (amount of nitrate in month 2-amount of nitrate in month 1).
From my understanding, the analyser works by reducing nitrate to nitrite, which forms a dye, and the concentration of the dye is then measured at 540nm.
Some of the values of NO2-NO3 given (ml/L) are e.g. 6mg/L. Other values are stated as
Can you look at your actual calibration curve ? Is it straight ? How does it look at the low end ?
Many analysts do not pay sufficient attention to the predictive quality of their calibration data, especially as Sal suggested, at the low concentration end. Regression fits the bigger values more closely than the smaller values and if the range is large, R2 values can be 0.99 yet therre can be appreciable bias at low concentrations. This shows up in plots of residuals. In addition, some calibration programs 'force' the fits through zero.
In addition, the efficiency of conversion (reduction) of nitrate to nitrate can vary during a 'run' i.e., the sensitivity can change, added to which the instrument can 'drift' over time. To accommodate this a mixed standard of nitrate and nitrite should be measured every few samples, the frequency being based on the actual performance.
My advice is to get the raw instrument data and analyse it yourself. You may hen need to chat with the analyst and arrange to have the instrument operated to meet your requirements.
However, ultimately as the concentration approaches zero there will be small positive and neg values.
negative values in the measurement of nitrate concentration can arise due to errors in the analytical procedure or instrument noise. It is important to identify the source of negative values to ensure accurate interpretation of the results.
One possibility is that the negative values could be due to baseline drift or interference from other compounds in the soil. In such cases, it is recommended to repeat the analysis to verify the results or consider using a different analytical method.
Another possibility is that the negative values could be due to the subtraction of the blank, which is a correction factor used to account for any interference from the reagents or the instrument. In such cases, it is important to ensure that the blank value is subtracted correctly and that the concentration of nitrate in the sample is not lower than the blank value.
In general, values that are below the limit of detection (LOD) should be reported as such and not replaced with 0mg/L. If a significant number of samples have values below the LOD, it may be necessary to adjust the analytical method to increase sensitivity or consider using a different method.
It is recommended to consult with a qualified analytical chemist or statistician to properly handle and interpret the data, particularly when dealing with negative values or values below the LOD.
For some simple advice on how to handle censored values (such as values below the nominal detection limit), see USEPA 2000 ( https://www.epa.gov/quality/guidance-data-quality-assessment ), section 4.7.
Many apologies for my delay in checking these answers, thank you all for your contributions, I now understand where these values came from and the next steps I need to take to improve this data.
- Badges
- Science topic
- Similar topics
- Chemistry
- Organic Chemistry