348
site conditions and standards are met and how management actions infl uence the
common characteristics associated with Bromus persistence (see Sect. 12.2.6 ) and
thus annual weed dominance (Sheley et al. 2010 ). Plans should also assess how
restoration activities infl uence disturbance regimes, invasion resistance, and site
resilience to stress and disturbance (Abella 2014 ; Chambers et al. 2014a ).
12.3.2 Implementation
The next step in the adaptive management cycle is project implementation (Fig.
12.1 ). Multifaceted management actions are carried out as agreed upon by the man-
agers and stakeholders followed by execution of the monitoring activities identifi ed
in the planning step. Monitoring data are summarized and analyzed as the cycle
continues into the evaluation step.
12.3.3 Evaluation
After management intervention, data gathered through monitoring is used to evalu-
ate the state of the system, as described in STMs (Williams and Brown 2012 ).
Because criteria for success are defi ned in the planning process, the monitoring
results are compared to these criteria to evaluate whether the trajectory of the sys-
tem is consistent with the management objectives. If it is not, the management
approach can be adjusted. These learning-based adjustments provide the necessary
feedbacks for making iterative decisions regarding the choice of management
actions (Svejcar and Boyd 2012 ; Williams and Brown 2012 ). For example, not
every restoration decision will lead to success, but through application of monitor-
ing, evaluating, and adjusting, management actions can be modifi ed to correct site
trajectories toward restoration goals. This feedback allows project leaders to learn
from both successes and failures as they collectively learn and incorporate informa-
tion about the fundamental successional processes responsible for changes within a
restoration site (Walker et al. 2007 ).
Bromus control will also benefi t from evaluating the infl uence of weather, which
is a major contributor to the variability of restoration outcomes in arid and semiarid
systems. Although weather is relatively unpredictable from year to year, it has his-
torical characteristics that can be captured to remove signifi cant uncertainty from
management actions (Hardegree et al. 2011 , 2012 ). While a number of models have
been proposed to incorporate the accretion of knowledge through adaptive manage-
ment (Allen et al. 2011 ), most comprehensive models do not specifi cally address the
issue of annual and seasonal variability in weather (Hardegree et al. 2012 ). However,
weather records, simulations of weather variability, and seasonal weather forecast-
ing information could be incorporated into both the planning and evaluation steps of
adaptive management. This incorporation reduces uncertainty by clarifying the role
of weather variability within a climatic regime (Hardegree et al. 2012 ; Abatzoglou
T.A. Monaco et al.