Meteorological conditions affecting renewable energy

Abstract

Synoptic situation and different meteorological phenomena can highly affect renewable energy production. Investigating different phenomena will give new information on the occurrence and characteristics of specific phenomena and their impacts on renewable energy applications. Different observational data sets and numerical models can be widely used in different phases of renewable energy projects; from planning of the project to help with the operation and the maintenance of the existing wind or solar field. In this thesis a meteorological phenomena, a low-level jet, is investigated. Thesis comprises analysis of the climatological occurrence of low-level jets, their characteristics and forcing mechanisms, as well as numerical model capability to capture the phenomena. In addition, solar radiation forecasts obtained from the operational numerical weather prediction model are evaluated and the role of cloud cover forecast skill in solar radiation forecast error is investigated. Long data sets of observational data: mainly Doppler wind lidar, ceilometer, and solar radiation observations, are used, in addition to reanalysis and operational numerical weather prediction model data. A low-level jet is a wind phenomenon that can affect wind energy production. Nighttime low-level jets are a commonly known boundary-layer phenomenon occurring during stably stratified conditions over flat terrain. In this thesis, new information on the occurrence, characteristics, and forcing mechanisms of a low-level jet was gained in different conditions: in Northern Hemisphere mid-latitude and polar regions based on reanalysis data and at two different sites in Finland and Germany based on long-term Doppler lidar observations. The low-level jet identification algorithms developed in these studies can be used to repeat the studies by using different models covering different areas or at any site operating a Doppler lidar. The low-level jet identification algorithm for Doppler lidar data can also be applied to operationally detect low-level jets, which is useful information for example from wind energy point-of-view. Solar radiation and cloud cover forecasts were evaluated at one site in Finland based on long time-series of solar radiation and ceilometer observations. The role of cloud cover forecast in solar radiation forecast error is investigated. The solar radiation and cloud cover forecasts were obtained from operational numerical weather prediction model that can be used to predict the expected power production at solar field day-ahead. It was found that there is a positive bias in the forecast incoming solar radiation even if the cloud cover forecast is correct. The study can guide model improvements as the bias is likely due to underestimation in the forecast cloud liquid water content or incorrect representation of cloud optical properties. The methods created in this study can be applied to hundreds of sites globally. In addition, the algorithms developed in this study can be further used in different applications in the field of renewable energy, for example to detect potential in-cloud icing conditions.