Adaptive management in variable environments

Adaptive management (AM) encourages land managers to modify plans for the future based on carefully evaluating what happened in the past. Extreme abiotic variation overwhelms effects of management, making AM extremely difficult to implement. We demonstrate how passive AM, involving monitoring over a few years and using data to inform future decisions, was used to determine seeding rates of native plants in a restoration experiment conducted in the Irvine Ranch National Landmark in Southern California. We found that seedling density depended more on the year in which seeding was conducted than on seeding rates. Analysis of longer term data, using weather as independent variables in analyses, can help managers tease apart the relative importance of management actions and of precipitation. Seeding in multiple years provided a bet hedging strategy for increasing success of the restoration project, given inter-annual variation in precipitation. Limitations of our seeding rate example could be resolved through active AM, in the form of controlled, manipulative experiments testing different seeding rates and replicated over many years. Effective AM requires a willingness to repeat management actions that failed due to abiotic conditions in a single year. What failed last year may work well in the future, and evaluating the role of environmental variation requires repetition across multiple years.