Quantifying impacts of ecological disturbance on ecosystem carbon and water fluxes will improve predictive understanding of biosphere—atmosphere feedbacks. Tree mortality caused by mountain pine bark beetles (Dendroctonus ponderosae) is hypothesized to decrease photosynthesis and water flux to the atmosphere while increasing respiration at a rate proportional to mortality. This work uses data from an eddy-covariance flux tower in a bark beetle infested lodgepole pine (Pinus contorta) forest to test ecosystem responses during the outbreak. Analyses were conducted on components of carbon (C) and water fluxes in response to disturbance and environmental factors (solar radiation, soil water content and vapor pressure deficit). Maximum CO2 uptake did not change as tree basal area mortality increased from 30 to 78% over three years of beetle disturbance. Growing season evapotranspiration varied among years while ecosystem water use efficiency (the ratio of net CO2 uptake to water vapor loss) did not change. Between 2009 and 2011, canopy water conductance increased from 98.6 to 151.7 mmol H2O m−2 s−1. Ecosystem light use efficiency of photosynthesis increased, with quantum yield increasing by 16% during the outbreak as light increased below the mature tree canopy and illuminated remaining vegetation more. Overall net ecosystem productivity was correlated with water flux and hence water availability. Average weekly ecosystem respiration, derived from light response curves and standard Ameriflux protocols for CO2 flux partitioning into respiration and gross ecosystem productivity, did not change as mortality increased. Separate effects of increased respiration and photosynthesis efficiency largely canceled one another out, presumably due to increased diffuse light in the canopy and soil organic matter decomposition resulting in no change in net CO2 exchange. These results agree with an emerging consensus in the literature demonstrating CO2 and H2O dynamics following large scale disturbance events are dependent not only on tree mortality but also on the remaining and new vegetation responses because mortality and recovery occur at the same time.