Integrating active restoration with environmental flows to improve native riparian tree establishment in the Colorado River Delta

•The Colorado River Delta riparian corridor is a highly human-impacted ecosystem due to altered hydrology, nonnative species introduction, and land use change.•Low-magnitude environmental flows to the Delta did not substantively restore natural fluvial processes of erosion, sediment deposition, or vegetation scour, but did provide wetted surface soils, shallow groundwater, and low soil salinity—critical requirements for Populus and Salix recruitment.•Environmental flow releases combined with active restoration, particularly nonnative vegetation removal, can improve Populus and Salix recruitment success.•Key management and flow design recommendations from this study include: 1) use canal infrastructure for effective flow delivery to target restoration sites; 2) clear undesirable vegetation prior to flow releases; 3) apply consistent base flow volumes over the growing season; 4) use water control structures to manage inundation and groundwater recession; and 5) use machinery to create more gradual channel slopes in river and backwater areas. Drastic alterations to river hydrology, land use change, and the spread of the nonnative shrub, tamarisk (Tamarix spp.), have led to the degradation of riparian habitat in the Colorado River Delta in Mexico. Delivery of environmental flows to promote native cottonwood (Populus spp.) and willow (Salix spp.) recruitment in human-impacted riparian systems can be unsuccessful due to flow-magnitude constraints and altered abiotic–biotic feedbacks. In 2014, an experimental pulse flow of water was delivered to the Colorado River in Mexico as part of the U.S.-Mexico binational agreement, Minute 319. We conducted a field experiment to assess the effects of vegetation removal, seed augmentation, and environmental flows, separately and in combination, on germination and first-year seedling establishment of cottonwood, willow, and tamarisk at five replicate sites along 5 river km. The relatively low-magnitude flow deliveries did not substantively restore natural fluvial processes of erosion, sediment deposition, and vegetation scour, but did provide wetted surface soils, shallow groundwater, and low soil salinity. Cottonwood and willow only established in wetted, cleared treatments, and establishment was variable in these treatments due to variable site conditions and inundation duration and timing. Wetted soils, bare surface availability, soil salinity, and seed availability were significant factors contributing to successful cotto