Connections between solar flare characteristics and their underlying magnetic drivers

The flare occurrences on the Sun have been a focal point of investigation for many years in the Solar community. Our increasing dependence on satellite technology as well as a way of living requiring frequent airplane trips, leave us susceptible to the interactions of the solar magnetic field with that of Earth's. Being able to identify parameters that could signal the occurrence of solar flares, would be advantageous towards our efforts to enhance our knowledge of flares and the principles of Space Weather. This thesis tackles two aspects of this problem. In its first half, we investigate a pattern of the solar magnetic field as it expands in the heliosphere, the Hale Sector Boundaries (HSBs). The heliospheric magnetic field typically is organised into two or four different sectors of alternating polarity. The HSB is the segment for which the change in magnetic sector polarity agrees with that of the leading and following sunspots in each hemisphere. Based on heliospheric measurements, we made use of recorded times when the sector boundaries were detected from Earth and mapped it back to the Photosphere assuming ballistic propagation of the solar wind. At these times, the HSB was at the Central Meridian back at the Sun. Using the flare positions as provided by the RHESSI instrument, we found them concentrated in the hemisphere of the Hale boundary. We then further developed a method of detecting HSBs using magnetic field extrapolations, which allows us to detect all HSBs present at all times flares were recorded. It additionally gives the chance to track the HSBs at all solar latitudes. Both approaches confirmed an association, adding to existing literature showing likewise. We find that 41% of RHESSI flares during Cycle 23 and 47% during Cycle 24 had a location within 30 degrees from their closest HSB. We later show how this pattern has evolved during the declining phase of Cycle 23 and the rising phase of Cycle 24, following the expected change between northern and southern hemispheres. Our findings show the rising phase of the solar cycle to be connected with a four-sector structure, while after solar maximum the sectors turn to a two-sector structure. That was an observational result that proved to be consistent during the times studied in this thesis. Finally, we contrast their migration paths to those of another feature of the large-scale field, Active Longitudes. Moving into a Carrington system, the HSBs were found to propagate faster than the Carrington rate at the beginning of a Cycle and slower at the declining phase, a property shared with Active Longitudes. Superimposing the paths followed by HSBs at Central Meridian to those of Active Longitudes showed no clear correlation. There were both points of overlap as well as large deviations, leaving space for further investigation. In the second half of this work, we moved into the smaller scales of the solar magnetic field, looking for potential persistent changes of its properties before microflares. We presented two different methods of fragmenting a line of sight magnetogram as well as an approach for tracking those fragments with time. We then applied our pipeline on data provided by the SDO/HMI instrument on four individual regions; one small flux emergence region used mainly for testing our code and three active regions with long series of microflares. The timelines of their bulk properties did not return any recurring trends in their patterns. We found the fragment area and unsigned flux to both increase and decrease before or after the start of the flare, with the average area and flux per fragment to be mostly unaffected by the emission of flares. We noted more changes in the positive field than the negative and mainly before a flare occurs. However, the results lacked consistency in their pattern to declare that a specific association has been spotted. Similarly, there was no specific direction the fragments were drifting towards, with the fragments of one of the regions consistently moving towards the left, while for the other two spreading apart. The Python pipeline we developed for time tracking the line of sight magnetic field did determine a number of additional properties about the fragments, which could be further studied and applied to new regions in future work. Our work shows that indeed there seems to be a stronger correlation of the HSBs with flaring activity and that there would be merit to explore further how the magnetic field is organised and results to this association. It also showed that the correlation is not as strong as to suggest that the HSBs can be used as a forecasting property by itself. The second project sets the ground for understanding the processes leading to microflares, a thoroughly unexplored field to-date. The pipeline has been created and tested, with the initial results not revealing a magnetic signature of microflares. However, the pipeline with a refinement can be used for a statistical study similar to those already existing for flares of large energetic output