The focus of most climate change impact studies to date is on changes related to mean climate conditions. In terms of climate model output, these changes are more robust than changes in climate variability, the latter of which has considerably greater uncertainty. By concentrating on climate means, however, the full impacts of climate change are probably being seriously underestimated. This report discusses and illustrates how the risk and uncertainty introduced by climate change can be incorporated into reclamation planning. Two approaches to reclamation planning are described. In the first approach, best management practices are developed using a deterministic methodology. A deterministic system is assumed to always produce the same output from a given starting condition or initial state. A corollary to this approach is that practices are geared to achieving the long-term average outcome. As long as this average satisfies management goals, variation is considered to be minimal and/or of little significance. The vegetation prescriptions provided in the Revegetation Manual are an example of a deterministic methodology. A fundamental assumption underpinning the validity of the approach is that past performance constitutes a reliable index of future performance. In the case of oil sands reclamation, this assumption is questionable for two reasons. First, oil sands reclamation soil materials possess biogeochemical properties and conditions that differ fundamentally from natural systems. Second, climate change is a source of uncertainty. It is anticipated to be a major chronic disturbance because of the northerly location of the oil sands. In the second approach, reclamation planning is undertaken using a stochastic methodology. This approach assumes that system development occurs along a trajectory dictated by one or more random variables (decision points). Each decision point thus represents an opportunity for the system trajectory to be altered by changes in the value of its random variables. Climate and climate change are likely the most important random variables influencing the developmental trajectory of reclaimed ecosystems. In this respect, the impact of climate as a driver of ecosystem performance needs to be considered. Under a stochastic, risk-based approach, the two basic principles of reclamation planning are: 1. That it represents the balance between the probability of an undesirable outcome and the marginal improvement in outcome from an additional unit of investment (increased capping depths or higher planting densities, for example), and 2. The greater uncertainty in outcome, the more conservative should be the management inputs (i.e., the higher the level of effort). One consideration in accounting for climate change is timescale. Over the next several decades, uncertainty in climate predictions will be predominantly a consequence of natural climatic variability. The relative effect of climate change increases significantly thereafter, which means the climate signature will become clearer and more predominant during the latter decades of this century. The implications for reclamation planning are that prescriptions suitable for establishing stands under current climate conditions may prove inadequate in the future, and short-term trends in vegetation performance may not be a reliable index of future performance. Changes in the disturbance regime associated with wildfire and insect epidemics are not given explicit consideration in reclamation planning. These risks add considerable uncertainty to assumption that current practices will be suitable for achieving long-term objectives. From a reclamation perspective, stand-level outcomes are a necessary prerequisite to successful reclamation, particularly if performance is focused on utilitarian metrics (merchantable volume, for example). Evidence suggests, however, that for the public at large, reclaimed areas are more likely to be evaluated in terms of their amenities, such as scenic beauty, ‘naturalness’, and recreational value – landscape-level attributes. The boreal mixedwood landscape has been characterized as a ‘mosaic’ of stands of differing age and species composition. At least part of this spatial heterogeneity will be created on mine sites because reclamation occurs progressively, which will ensure heterogeneity among stand ages. A second option for creating heterogeneity is to ‘plan for failure’ (PFF). Under a PFF strategy, stands are expected to vary in their developmental trajectories, with some stands transitioning to a different end land-use than originally intended. This variation constitutes the basis on which the desired level of heterogeneity is achieved. Another option is to actively manage for landscape heterogeneity by varying capping and planting prescriptions on a stand-by-stand basis. The advantages and disadvantages of these options are discussed. Changes in the disturbance regime (wildfire or insect epidemics) could largely render moot concerns around uncertainties in development trajectories. A fundamental challenge to assessing current best management practices within the context of climate change is the questionable utility of relying on historical practices for guidance. The success of a particular reclamation prescription in meeting long-term objectives can, in principle, be assessed empirically. In practice, however, many years must elapse before a reclaimed stand has developed sufficiently that a given prescription can be evaluated definitively or that interim measures are a reliable proxy for long-term outcomes. Modeling of ecosystem development is perhaps the only practical approach to resolving to this dilemma. There have been two basic approaches to predictive modelling of ecosystem response to a changing environment: empirical (statistical)- and process-based models. Here, a stochastic approach is described in which probability outcomes are derived for reclamation planning using the FORECAST Climate model and a state-and-transition model (STSM). FORECAST Climate is used to project vegetation development (i.e., ‘states’) for a given reclamation land unit (e.g., dry, moist rich, moist poor, wet rich, and wet poor) subject to current and alternative management options, disturbance regimes, and two climate change scenarios. The probabilities associated with each state transition are then be derived from these runs. The STSM simulates vegetation development for reclamation land units over time and across an entire mine footprint. By implementing a Monte Carlo experiment (i.e., repeated iterations through the STSM) in conjunction with the transition probabilities from FORECAST Climate, uncertainties in outcome for a given reclamation practice are assessed as a consequence of climate change. Model output will permit stakeholders and regulators to evaluate the efficacy of current and alternative adaptation strategies with respect to mitigating risk of undesirable outcomes due to climate change. In addition, the STSM will be provided with the capability for geospatial representation of each land unit and land unit phase. This functionality will aid mine operators in meeting approval conditions regarding integration across lease boundaries and with undisturbed areas. The tool will also be useful for wildlife habitat planning and assessment of reclamation performance with respect to re-establishing wildlife habitat (both of which have a strong spatial component).