Data quality descriptions in this table are subjective determinations based on the authors' experience with these methods. They are based on how well the method is expected to perform for stream sediment samples with typical compositions for the United States. Opinions take into consideration known analytical problems, published detection limits, and past experience in comparing data from multiple analytical methods. The crude categories used to describe data quality are:
Excellent. The method has a low detection limit and a reputation for producing data of the highest quality; no major analytical problems are known.
Very good. The method has reasonably low detection limits; when available, other methods agree fairly well. No major analytical problems are known.
Good. The method produces reasonable data; detection limits may be somewhat higher than is desireable; there may be some known analytical problems.
Poor. Data for this method are probably not very useful except in cases of very large enrichments of the element; detection limits may be very high; major analytical problems or interferences may be known.
Variable. These data from the NURE program were most likely determined by a combination of several methods. The quality of these methods may be highly uneven, ranging from extremely poor to excellent.
Future versions of this report will contain quantitative discussions of accuracy and precision for each method.