Unmanned aerial systems (UASs), or drones, are increasingly being used for field sampling in wildlife research. However, to date there have been few studies using drones in avian nest searching, and none where a fixed survey methodology has been applied across multiple habitats spanning a large geographic region to locate avian nests. We tested nest searching using drone-based thermal imagery for a variety of game (Anseriformes, Galliformes), and nongame bird species (Passerines and Charadriids) at sites located in the Dakotas, North Carolina, and Alberta. We used a manual, point-transect survey method for all study sites in habitats ranging from marine-coastal marshlands to upland grasslands. We found that 77-100% of known nests can be detected based on thermal signatures in multiple habitat types and across an array of different avian species. However, detectability varies with a variety of site, weather, atmospheric conditions, and the effectiveness of the method is impeded by a high false positive rate (60-95%). There are several identified challenges to be addressed for future thermal drone research, and our study provides observations and advice for planning studies. Drone technology has the potential for field applications by wildlife managers and conservationists, and as it matures will enhance the field practitioner's toolset.