At two affiliate sites of the North American Long-Term Soil Productivity Study (Matlock, WA and Molalla, OR, USA), soil chemical properties and stand characteristics of planted coast Douglas-fir (Pseudotsuga menziesii var. menziesii) were compared 15 years after forest harvesting and application of three logging-debris configurations (dispersed, piled, or removed) combined with either initial vegetation control (IVC; year 0) or annual vegetation control (AVC; years 0 to 5). At Matlock, soil carbon (C) and nitrogen (N) concentrations each were 17% greater after IVC than after AVC; at Molalla, soil N was 13% greater where debris was removed than where it was dispersed. At Matlock, cover of nonnative Scotch broom (Cytisus scoparius) after IVC was greater where debris was removed (61%) than where it was piled (27%) or dispersed (7%), despite a control treatment in year 4. Conversely, covers of the native shrubs, trailing blackberry (Rubus ursinus) and salal (Gaultheria shallon) were 20% to 30% greater where debris was dispersed than where it was piled or removed. With AVC versus IVC, Douglas-fir stand volume was 34% to 159% greater at Matlock depending on the logging-debris treatment, and it was 30% greater at Molalla independent of debris treatments. However, Douglas-fir survival and growth after AVC did not differ among logging-debris treatments at either site. Survival of Douglas-fir growing <= 1 m from the edge of debris piles at Matlock averaged 16% greater than that of trees > 1 m from debris piles. Debris dispersal or piling at Matlock strongly mitigated Scotch broom impacts to forest productivity compared to debris removal. Our findings demonstrate how disturbance characteristics associated with forest harvesting and regeneration practices can influence vegetation recruitment and competitive relationships to place limits on longer-term forest productivity.