Doctor of Philosophy
Water vapor is the most dominant greenhouse gas in Earth's atmosphere. It is highly variable and its variations strongly depend on changes in temperature. Atmospheric water vapor can be expressed as relative humidity (RH), the ratio of the partial pressure of water vapor in the mixture to the equilibrium vapor pressure of water over a flat surface of pure water at a given temperature. Liquid water can exist as super-cooled water for temperatures between 0C to -38C. Thus, RH can be measured either relative to water (RHw) or to ice (RHi). RHi measurements are important in the upper tropospheric region, where the temperature is always less than 0C, to study ice supersaturation (ISS) and its relation to the formation of cirrus clouds.
I present three studies all using a mathematical scheme called the optimal estimation method (OEM). The OEM is an inverse method that determines the most probable state consistent with the measurements and a priori knowledge. These studies use parts of a large set of existing measurements from the Raman Lidar for Meteorological Observations (RALMO) instrument located at the meteorological observatory in Payerne, Switzerland.
I first develop an OEM retrieval for temperature using RALMO's two pure rotational Raman (PRR) channel measurements. Retrieved temperatures show excellent agreement with coincident balloon-borne radiosonde measurements. A second OEM scheme is introduced to retrieve RHw directly from RALMO measurements of back-scatter due to water vapor and nitrogen. I validate the OEM retrievals developed for temperature and RHw. I then combine the OEM-retrieved temperature and RHw with data from the European Centre for Medium-Range Weather Forecasts Re-analysis (ERA5) to compute a new and improved temperature and relative humidity product. The retrieval is enhanced by assimilating it with the ERA5 data. The quality of the RHw retrievals from the assimilated OEM scheme greatly improves over retrievals which use less accurate a priori information.
Thirdly, I retrieve RHi to detect ISS layers. I find the frequency of ISS layers in the free troposphere over Payerne to be about 27% using 82.5 hours of measurements.
Summary for Lay Audience
Water vapor is the most dominant greenhouse gas in Earth’s atmosphere. It is highly variable and its variations strongly depend on changes in temperature. Accurate estimates of humidity and temperature and as well as the uncertainties associated are required for both weather and climate forecasting purposes. I present a new mathematical and statistical approach to estimate both atmospheric humidity and temperature using Raman lidar backscatter measurements. The new method provides full uncertainty budgets for each estimated temperature and relative humidity profile, that represent the errors due to instrumentation, estimation method and so on. I have also combined the Raman lidar measurements into the data from the ERA5 that is the latest major global reanalysis produced by European Centre for Medium- Range Weather Forecasts (ECMWF), to enhance the quality of the humidity and temperature estimates. My results show that the quality of the temperature and humidity retrievals are greatly improved and agree best with the measurements made by coincident radiosondes.
Mahagammulla Gamage, Shayamila N., "Development of a 1-Dimensional Data Assimilation to Determine Temperature and Relative Humidity Combining Raman Lidar Backscatter Measurements And a Reanalysis Model" (2019). Electronic Thesis and Dissertation Repository. 6356.