University of Twente Student Theses


Monitoring the activities of fumaroles using thermal infrared, meteorological, and gas chemistry data in the Olkaria Geothermal Field: a possible link to the geothermal motor?

Omwenga, Benard Mageto (2021) Monitoring the activities of fumaroles using thermal infrared, meteorological, and gas chemistry data in the Olkaria Geothermal Field: a possible link to the geothermal motor?

Link to full-text:
(only accessible for UT students and staff)
Abstract:Long-term monitoring of fumarole temperatures variation can provide insight into the subsurface behavior, especially fluxes in the geothermal and volcanic system. Conventional field-based monitoring methods, e.g., calorimetric methods, though accurate, are limited in the temporal frequency domain. The thermal infrared data, especially geostationary thermal infrared sensors’ have widely been useful for fumarole monitoring, especially in eruptive volcanic surveillance studies. However, pertinent issues such as spatial resolution and influences of atmospheric and surface effects on recorded temperature values have hampered comprehensive monitoring campaigns. For this reason, the study investigated the activities of fumarole in line with the following objectives: (i) apply high temporal resolution thermal infrared imagery to study surface temperature variation as a proxy to subsurface geothermal heat flux, (ii) To understand external factors affecting the apparent temperature using weather data (air temperature and precipitation), (iii) contextualize subsurface dynamics of the geothermal reservoir using fumarole gas discharge data. Data processing was done procedurally: Demarcation of regions of interest from specific hour-of-day TIR imagery by visual inspection of principal component analysis, temporal mean, standard deviation, and optical image collected by the daytime timelapse camera. Secondly, time series plots of the apparent radiant temperature of the demarcated regions of interest were generated and their long-term patterns analysed. The third step involved removing low-quality images based on the standard deviation value of pixels of time series TIR images. Normalization of time series plots was done to remove further daily components associated with environmental effects. Finally, the normalized time series plots were used to calculate radiative heat flux plots. Finally, fumarole gas concentration data were plotted, and temporal patterns were analysed. The results show comparable spatial patterns within indicator products of principal component analysis, temporal mean products of selected hour-of-day TIR images. The temporal of time series plots of different ROIs extracted from TIR show variation with seasonality. The variations were related to existing environmental conditions and were specifically enhanced during rainy periods. The periodicity also varied per region of interest and was optimal in thermally anomalous sections near the fumarole area. The results further show that removing images acquired during foggy conditions and the time normalization procedure eliminated anomalous signals from time series plots. The heat flux plots, calculated from normalized radiant temperature values, were useful in identifying endogenous processes associated with subsurface heat emission. Finally, fumarole gas chemistry concentration plots show variability with occasional peaks (between July 2008 to June 2020), which were attributed to subsurface processes such as dilution and boiling. Such processes were linked with the variability of subsurface fluid movement, an aspect of fluctuating speeds of the geothermal motor. In conclusion, the study revealed that TIR imagery acquired by ground-based thermal infrared cameras is a useful tool in monitoring fumarole's thermal behavior in a productive geothermal field. Besides, the study's novelty lies in isolating external influences from TIR infrared imagery time series data, analyzing the residual trends, and associating them with subsurface geothermal activities.
Item Type:Essay (Master)
Faculty:ITC: Faculty of Geo-information Science and Earth Observation
Programme:Geoinformation Science and Earth Observation MSc (75014)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page