What are detection limits (DL) and reporting limits (RL)

Analytical machines that test for chemicals in soil, air, water, and other matrices and the science behind them, analytical chemistry, have limits to their accuracy, precision, sensitivity to chemicals. Detection limits, reporting limits, and limits of quantitation are all names for various limits that describe the lowest concentrations of chemicals that can be observed with any degree of confidence. Unfortunately, there is no universal standard for defining these limits. However, it is important that they exists because they communicate to the end user of analytical chemistry data the performance of a particular laboratory, test method, and/or analyst. In making environmental decisions, it is important to understand the inherent limitations of the information you are using and paying for.

For purposes of environmental regulation, the United States Environmental Protection Agency (USEPA) defines a detection limit in Code of Federal Regulations (CFR) 40 CFR 136, Appendix B, revision 1.11. This appendix to federal regulations specifies specific steps laboratories must follow to define a detection limit. However, regulators, consultants, and property owners rarely see detection limits on laboratory reports ordered from commercial analytical laboratories. More often, laboratories use reporting limits to describe the concentration of chemicals in the environment (or rather samples from the environment). So what's the difference between detection limits and reporting limits?

Reporting limits are typically set by commercial analytical laboratories at concentration levels above detection limits. Reporting limits may or may not be based on statistically determined detection limits or limits or quantitation (LOQ). More often than not, they are built in safety factors that allows commercial laboratories to easily handle variability associated with analyzing samples from a wide variety of sources with different possible complicating factors (such as matrix effects). Commercial laboratories more often than not run a variety of samples from different sites and matrices. It would be too time consuming to statistically determine the detection limit for each individual sample.

(Why should I care about reporting limits?) Next page >>