Persistence is one of the core criteria in chemical exposure and hazard assessment. It is often defined as the half-life for the removal of a chemical from a specified environment by transformation. Chemicals with long transformation half-lives may pose high risks for wildlife or humans and be subject to long-range transport to remote areas. It is challenging to measure persistence directly in the field in view of the complexity of the natural environment and spatial and temporal variability in environmental conditions that may affect degradation. The mass balance approach is the most commonly used method for field measurement of persistence. In this thesis an alternative to the traditional mass balance approach that uses benchmarking is proposed and evaluated using models and field application. The benchmarking approach compares the relative behavior of chemicals, rather than measuring the absolute value of a property. The unknown property (persistence in this thesis) of test chemicals can be estimated by comparison against another chemical for which this property is known. In Paper I, the potential of benchmarking to measure persistence in the field was evaluated by modeling. A framework for applying benchmarking to measure persistence in the field was developed. Lake systems with hydraulic residence times of the order of months were identified as appropriate field sites to measure the persistence of chemicals that are close to the regulatory thresholds, which are also on the scale of months. Field studies in two Swedish lakes were conducted. Both are shallow lakes, whereas Norra Bergundasjön (Paper II) has a longer residence time (four months) than Boren (one to two months; Paper III). In Paper II the benchmarking approach was tested to measure the persistence of a group of chemicals that were expected to stay in the water phase. Acesulfame K (artificial sweetener) without observable degradation in the lake was used as the benchmark chemical. The persistence of 9 pharmaceuticals and one X-ray contrast agent was measured to range from &lt;1-2 days (ketoprofen) to 580-5700 days (carbamazepine). The results obtained using the benchmarking approach agreed well with the mass balance approach, indicating that the benchmarking approach can be a valid and useful method to measure persistence in the field. In Paper III the seasonality in chemical persistence was investigated by benchmarking. The seasonal difference in chemical persistence was found to be largest between spring and autumn. The persistence of 5 chemicals in spring were lower than in autumn, mainly attributed to lower temperature and less sunlight in autumn. The spatial variation of the persistence of chemicals was observed by comparing the persistence of chemicals in spring in the two lakes. Thus benchmarking is a useful tool to study the temporal and spatial variation of persistence in the real environment.   Paper IV explores the potential of benchmarking thoroughly and the application of benchmarking in a regulatory context. Benchmarking could facilitate more field measurements of persistence, leading to a better understanding of the temporal and spatial variability of persistence in various environments and a basis for lab-to-field extrapolation. Besides quantitative estimation of persistence in the field, benchmarking can be applied to determine the relative magnitude of persistence, called threshold benchmarking which could be a valuable tool in regulatory processes. / <p>At the time of the doctoral defense, the following papers were unpublished and had a status as follows: Paper 3: Submitted. Paper 4: Manuscript.</p>