Ecological restoration aims to reverse damage caused by humans to ecosystems, whereas restoration ecology is the discipline that studies the application of ecological theory to restoring ecosystems and using information from restorations to refine or advance theory. There is increasing evidence that soil ecological knowledge may be key to restoration success, or failure. We review several theories in ecology that require soil ecological knowledge to test and apply to restoration. For example, stoichiometric theory predicts relationships among nutrients for many levels of biological organization to result in different demands, dynamics, and changes in limiting nutrients as restoration proceeds. Plant nutrient use theory predicts litter quality feeds-back to influence nutrient cycling and primary productivity, but does the assumption of steady state conditions limit applicability to restoration? Alternatively, plant input-output theory suggests the immediate effects of plants on nutrient cycling in soil occur through more direct mechanisms. When diversity is a restoration goal, can coexistence theory be applied to steer soil nutrient status, prevent dominance, and promote species diversity? These and other examples will be used to lay a framework for considering soil as a manageable filter, critical threshold, or irreversible tipping point in ecological restoration.
Ecological restoration provides a means to test changes in biogeochemical cycling and biodiversity predicted by theory during ecosystem development. Change in stoichiometry of ecosystem components can reflect changes in biogeochemical cycling. For example, increasing C:N ratios in plant biomass and soil indicates development of progressive N limitation, and when this does not occur indicates constraint or altered trajectory of ecosystem recovery. Restorations accumulating low quality plant inputs often exhibit low N supply, but contrary to plant nutrient-use theory predictions, higher N mineralization in fertilized soil may not result from higher quality plant inputs. Further, reducing soil N availability may not result in lower N supply rates according to plant nutrient use theory due to shifts in community composition, supporting plant input-output theory. A long-term manipulation of soil resource availability provided weak support for the hypothesized positive effect of environmental heterogeneity on plant diversity. Thus, application of coexistence theory to restoring plant diversity may require more knowledge of plant-soil feedbacks. These examples demonstrate that the degree to which soil is a manageable filter depends on whether the constraint to restoration can be manipulated and knowledge of competition and diversity outcomes in response to changing soil properties and processes during restoration.