Unoccupied aerial system (UAS) imagery may serve as an additional tool towards management zone delineation. This is because UAS data collection is relatively flexible. However, it is unclear how useful UASs can be towards generating management zones, relative to preexisting tools (e.g. apparent soil electrical conductivity or EC_a). The purpose of this study, therefore, was to evaluate UAS imagery, relative to EC_a, in terms of their ability to: 1) predict cotton traits (i.e. height, seed cotton yield), and 2) define cotton management zones based on these traits. Single-season UAS images from multispectral/thermal sensors were collected and processed into Normalized Difference Vegetation Index (NDVI) and radiometric surface temperature (T_r), respectively. Management zones were also delineated using digital camera (RGB) imagery collected at periods before planting and near harvest. RGB management zones were delineated by a novel open boll mapping approach. In-season NDVI and T_r layers were significant (P?<?0.01) predictors of canopy height. Additionally, NDVI and T_r maps produced statistically different management zones during flowering and boll filling growth stages in terms of yield (P?=?0.001 or less). Open boll layers were all more accurate predictors of cotton seed yield than EC_a data—these two layers also produced statistically distinct management zones. ANOVA tests revealed that, given EC_a alone, adding UAS information via the RGB open boll map resulted in a significantly different yield prediction model (P?<?0.001). These results suggest that UAS imagery can offer valuable information for cotton management zone delineation that other techniques cannot.