Recently, examination of deoxyribonucleic acids in water samples (environmental DNA or eDNA) has shown promise for identifying fish species present in water bodies. In water, eDNA arises from bodily secretions such as mucus, gametes, and feces. I investigated whether eDNA can be effective for characterizing fish presence, relative abundance, biomass, and species composition in a large Arizona reservoir (Theodore Roosevelt Lake) and 12 small Arizona (<24 ha) waterbodies. Specifically, I compared fish presence, relative abundance (catch per unit effort [CPUE]), biomass (biomass per unit effort [BPUE]), and species composition measured through eDNA methods and established American Fisheries Society (AFS) standard sampling methods in Theodore Roosevelt Lake and 12 small waterbodies. Environmental DNA sampling resulted in detection of Gizzard Shad Dorosoma cepedianum at a higher percentage of sites than boat electrofishing, both in spring and fall. Contrarily, gill nets detected Gizzard Shad at more sites than eDNA for both spring and fall sampling in Lake Roosevelt. Boat electrofishing and gill netting detected Largemouth Bass Micropterus salmoides at more sites than eDNA, with the exception of fall gill net sites which equally detected Largemouth Bass at sites within Lake Roosevelt. Environmental DNA detected Largemouth Bass and Bluegill Lepomis macrochirus at more Arizona small lakes than detection with established gear methods. I observed no relationship between relative abundance and biomass of Largemouth Bass and Gizzard Shad measured by established methods and their DNA copies at individual sites or by lake section in Lake Roosevelt. Likewise, I found no relationship between relative abundance and biomass of Largemouth Bass and Bluegill measured by established methods and their DNA copies across 12 small waterbodies. Plot analysis conceivably illustrated that reservoir-wide catch composition (numbers and total weight of fish [g]) achieved through a combination of gear types (boat electrofishing + gill netting) for Largemouth Bass and Gizzard Shad was slightly similar to the proportion of total eDNA copies of each species for both spring and fall field sampling. Likewise, spring and fall gill net surveys somewhat portrayed total catch composition (numbers and total weight of fish [g]) of Largemouth Bass and Gizzard Shad similar to the proportion of total eDNA copies of each species. The exception was the total lack of similarity illustrated between proportions of fish caught in spring and fall boat electrofishing and total eDNA copies of each species in Lake Roosevelt. However, the deceptive similarity of all the plots were not present in the chi-square analysis with the exception of fall gill net surveys in Lake Roosevelt. In addition, eDNA did reflect the relative proportions of Largemouth Bass and Bluegill in total catch composition in some, but not all of 12 small Arizona waterbodies. The ease of eDNA sampling over established fish sampling makes it appealing to natural resource managers. Compared to current established fish sampling methods, eDNA sampling can be less laborious, less time consuming, and more cost effective. Environmental DNA sampling may be useful in sites that have difficult access such as remote sites. However, evaluation of eDNA is necessary to identify limitations and benefits in fish monitoring programs. Furthermore, field sampling protocols, filtration, DNA extraction, primer design, and DNA sequencing methods need further refinement and testing before incorporation into standard fish sampling surveys.