This thesis investigates some aspects of the inverse Compton scattering process within various physical contexts in high energy astrophysics. Initially an introduction to the key results of Comptonization theory for the case of scattering in optically thick plasmas is given, using a diffusion approach, since these results are required for the interpreta- tion of Comptonized spectra. Since Comptonization in astrophysical systems is frequently treated using numerical techniques, an introduction to these is then presented. Such linear Monte Carlo photon transport codes are typically applied to scattering in plasmas without temperature and density gradients. Additionally, treating bulk motion can be difficult even for simple cases. It is demonstrated that these problems can be made tractable numerically with the use of algorithms associated with non-linear Monte Carlo codes. Such codes can already treat scattering within arbitrary velocity structures in a plasma, and an extension of the algo- rithm is proposed that enables the easy calculation of photon transport in plasmas with non-constant density as well as non-constant temperature and/or bulk motion. This algorithm and code has been developed to treat scattering in astrophysical situations where bulk motion, temperature gradients and density gradients are simultaneously present in a plasma. Both a semi-analytic approach and the numerical approach are then used to treat Comp- tonization problems of current interest. Firstly, the standard two-phase disk-corona model for the high-energy spectra of Active Galactic Nuclei is modified to include an an outflow or wind which may provide an additional source of disk cooling. Earlier slab disk-corona models predict a spectral index which is consistent with observations only if all the accretion power is dissipated in the corona. For the models investigated here, energy spectral indices that are consistent with observations can be obtained with less accretion power being dissipated in the corona, as a result of an outflow/wind. However, it is required that the wind extract large amounts of power from the disk, and it it yet to be seen if this is a plausible scenario.

This thesis investigates some aspects of the inverse Compton scattering process within various physical contexts in high energy astrophysics. Initially an introduction to the key results of Comptonization theory for the case of scattering in optically thick plasmas is given, using a diffusion approach, since these results are required for the interpreta- tion of Comptonized spectra. Since Comptonization in astrophysical systems is frequently treated using numerical techniques, an introduction to these is then presented. Such linear Monte Carlo photon transport codes are typically applied to scattering in plasmas without temperature and density gradients. Additionally, treating bulk motion can be difficult even for simple cases. It is demonstrated that these problems can be made tractable numerically with the use of algorithms associated with non-linear Monte Carlo codes. Such codes can already treat scattering within arbitrary velocity structures in a plasma, and an extension of the algo- rithm is proposed that enables the easy calculation of photon transport in plasmas with non-constant density as well as non-constant temperature and/or bulk motion. This algorithm and code has been developed to treat scattering in astrophysical situations where bulk motion, temperature gradients and density gradients are simultaneously present in a plasma. Both a semi-analytic approach and the numerical approach are then used to treat Comp- tonization problems of current interest. Firstly, the standard two-phase disk-corona model for the high-energy spectra of Active Galactic Nuclei is modified to include an an outflow or wind which may provide an additional source of disk cooling. Earlier slab disk-corona models predict a spectral index which is consistent with observations only if all the accretion power is dissipated in the corona. For the models investigated here, energy spectral indices that are consistent with observations can be obtained with less accretion power being dissipated in the corona, as a result of an outflow/wind. However, it is required that the wind extract large amounts of power from the disk, and it it yet to be seen if this is a plausible scenario.