1) The effects of hydrodynamic escape in planetary atmospheres. Any previous photochemical calculations assumed a high (“diffusion-limited”) rate of hydrogen loss to space. If the atmosphere was anoxic, hydrogen should have been lost at a much (>100 times) slower rate (Tian et al., 2004). Hydrogen loss is extremely important for understanding the evolution of the atmospheric redox state. In hydrogen-rich atmospheres, CH4 molecules could have been effectively “recycled” after the initial photolysis and methane abundance would be high even with a small source of methane. The hydrodynamic escape of hydrogen is also important for the problem of noble gases’ abundance on the terrestrial planets. 2) Modeling of mass independent fractionation (MIF) in sulfur/oxygen isotopes. MIF in sulfur is a sensitive indicator of the atmospheric redox state. Recently, the small sulfur MIF signatures were discovered in the Antarctic ice cores. We model sulfur isotopes explicitly and try to understand the transfer of non-zero sulfur MIF to the poles during large volcanic eruptions. 3) Glaciations and Mass-Extinctions due to changes in the galactic environment. Throughout its history, the Solar system encountered a number of clouds with higher hydrogen densities 10-10000 H atoms/cm3 and will inevitably encounter them in the future. It is important to understand whether those events could cause any significant climatic impact, affect the biosphere and leave any imprint in the geological record. For future research, I propose to consider three possible mechanisms pending on the density of the colliding galactic cloud. a) Collisions with dense (>1000 H atoms/cm3) clouds (a few times in Earth history). I show that dramatic climate change can be caused by interstellar dust (from the cloud) accumulating in Earth’s atmosphere during collision. The stratospheric dust layer from such interstellar particles could provide enough radiative forcing to trigger the runaway ice-albedo feedback that results in global Snowball glaciations. b) Collisions with moderate density clouds (a few hundred times in Earth history). Moderate density clouds would not suppress the heliosphere within 1 A.U. and therefore interstellar dust would not be able to penetrate into the Earth atmosphere. However, it is very plausible that moderate density clouds would destabilize the Kuiper Belt dust orbits and huge quantities of dust would be thrown into the inner parts of the Solar system and ultimately into the Earth atmosphere. It would produce a short-term but very intensive climatic perturbation. c) Collisions with low density clouds (thousands times in Earth history). Even if the density of the local interstellar medium increases to only ~10 H/cm3, the heliosphere can be effectively suppressed within 10 A.U. due to increased external hydrogen pressure. Such events probably happened thousands of times in Earth history. Here, we show that when the heliosphere was suppressed (to 5-10 A.U.), the flux of the anomalous cosmic rays (ACR) increases dramatically into the Earth atmosphere. ACR particles are energetic, can penetrate deep into the Earth’s atmosphere and have enough energy to break N2 molecules. We calculate the production rate of NOx due to ACRs in the lower stratosphere (10-30 km) as a function of latitude and showed that it should have been orders(!) of magnitude higher in comparison with the total rate of NOx production from N2O. Using 2-D photochemical model, we evaluate how such a high NOx production rate would affect the global ozone layer.