The galactic cosmic rays are the main source of ionization in the Earth stratosphere and troposphere. The induced by primary cosmic ray particles ionization is important in various processes related to atmospheric physics and chemistry, specifically the minor constituents. The ion production in the atmosphere is enhanced compared to the average following major solar energetic particles events, specifically over the polar caps. During the solar cycle 23 we observed several strong ground level enhancements, one of the strongest among them been the Bastille day event on 14 July 2000. In the work presented here we apply a full Monte Carlo 3-D model in order to compute the cosmic ray induced ionization. The model is based on atmospheric shower simulation with the PLANETOCOSMICS code and the ion production rate is considered as a superposition of cosmic rays with galactic and solar origin. The ion production rate is computed as a function of the altitude above the sea level and the short, mid and long term ionization effect relative to the average due to galactic cosmic rays is computed.

The possible effect of solar variability, accordingly cosmic rays variation on atmospheric physics and chemistry is highly debated over the last years. According to several recent models the induced by cosmic rays atmospheric ionization plays a key role in several different processes. At recent, an apparent effect on minor constituents and aerosols over polar regions during major solar proton events was observed. The ion production rate during ground level enhancements is a superposition of the contribution of cosmic rays with galactic and solar origin. The solar cycle 23 provided several strong ground level enhancements. The period of end October - beginning of November 2003 was characterized by a strong cosmic ray variability, namely a sequence of three GLEs was observed. In addition, there were several Forbush decreases, which led to a suppression of galactic cosmic ray flux. As a consequence the cosmic ray induced ion production in the atmosphere and the corresponding ionization effect were subject of dynamical changes. Here we compute the complex ion production due to cosmic rays during the Halloween sequence of ground level enhancements on October-November 2003 and we estimate the ionization effect. The spectral and angular characteristics of the solar protons are explicitly considered throughout the events as well their time evolution. The ionization effect during the period is computed at several altitudes above the sea level in a region with $R_{c}$ $\le$ 1 GV and $R_{c}$ $\le$ 2 GV.

Recently an essential progress in development of physical models for cosmic ray induced ionization in the atmosphere is achieved. Basically, the models are full target, i.e. based on Monte Carlo simulation of an electromagnetic-muon-nucleon cascade in the atmosphere. In general, the contribution of proton nuclei in those models is highlighted, i.e. primary cosmic ray $\alpha$-particles and heavy nuclei are neglected or scaled to protons. The development of cosmic ray induced atmospheric cascade is sensitive to the energy and mass of the primary cosmic ray particle. The largest uncertainties in Monte Carlo simulations of a cascade in the Earth atmosphere are due to assumed hadron interaction models, the so-called hadron generators. In the work presented here we compare the ionization yield functions $Y$ for primary cosmic ray nuclei, such as protons, $\alpha$-particles, Oxygen and Iron nuclei, assuming different hadron interaction models. The computations are fulfilled with the CORSIKA 6.9 code using GHEISHA 2002, FLUKA 2011, UrQMD hadron generators for energy below 80 GeV/nucleon and QGSJET II for energy above 80 GeV/nucleon. The observed difference between hadron generators is widely discussed. The influence of different atmospheric parametrizations, namely US standard atmosphere, US standard atmosphere winter and summer profiles on ion production rate is studied. Assuming realistic primary cosmic ray mass composition, the ion production rate is obtained at several rigidity cut-offs - from 1 GV (high latitudes) to 15 GV (equatorial latitudes) using various hadron generators. The computations are compared with experimental data. A conclusion concerning the consistency of the hadron generators is stated.