Selection for genetics-based architecture traits in a native cottonwood negatively affects invasive tamarisk in a restoration field trial

Climate change and competition from invasive species remain two important challenges in restoration. We examined the hypothesis that non-native tamarisk (Tamarix spp.) reestablishment after aboveground removal is affected by genetics-based architecture of native Fremont cottonwood (Populus fremontii) used in restoration. As cottonwood architecture (height, canopy width, number of stems, and trunk diameter) is, in part, determined by genetics, we predicted that trees from different provenances would exhibit different architecture, and mean annual maximum temperature transfer distance from the provenances would interact with the architecture to affect tamarisk. In a common garden in Chevelon, AZ, U.S.A. (elevation 1,496 m), with cottonwoods from provenances spanning its elevation distribution, we measured the performance of both cottonwoods and tamarisk. Several key findings emerged. On average, cottonwoods from higher elevations were (1) two times taller and wider, covered approximately 3.5 times more basal area, and were less shrubby in appearance, by exhibiting four times fewer number of stems than cottonwoods from lower elevations; (2) had 50% fewer tamarisk growing underneath, which were two times shorter and covered 6.5 times less basal area than tamarisk growing underneath cottonwoods of smaller stature; and (3) the number of cottonwood stems did not affect tamarisk growth, possibly because the negative relationship between cottonwood stems and basal area. In combination, these findings argue that cottonwood architecture is affected by local conditions that interact with genetics-based architecture. These interactions can negatively affect the growth of reinvading tamarisk and enhance restoration success. Our study emphasizes the importance of incorporating genetic and environmental interactions of plants used in restoration.