Abstract:Abstract As a percentage of the total global energy supply, wind energy facilities could provide 10% of the total global energy supply by 2050 as reported in IEA World Energy Outlook (2022). Considering this, a just transition to renewable and sustainable energy in South Africa is a genuine possibility if steps are taken immediately to achieve this. The Eastern Cape Province exhibits a strong wind resource which can be exploited towards expediting such a just energy transition. No research and related modelling have, to date, been undertaken in quantifying and relating the detailed P50 energy yield analyses of representative wind energy facilities in temporal and spatial dimensions to the occurrence of specific synoptic types in South Africa. To quantify this energy meteorology climatology for a suitably sized geospatial area in the Eastern Cape Province of South Africa (spatial focus area, latitude −30 to −35, longitude 20 to 30), the approach of using self-organising maps is proposed. These maps are used to identify the most common synoptic circulation types occurring in the Eastern Cape and can subsequently be mapped onto an equivalent time resolution wind energy production timeseries calculated based on probable wind energy facility sites. This paper describes comprehensive methodologies used to model the wind energy facilities, calculate with high confidence the P50 energy production, and then identify the predominant synoptic weather types responsible for the wind energy production in this spatial focus area. After quantifying the energy production, running a self-organising map software generates a purposely selected 35 node map that characterises archetypal synoptic patterns over the 10-year period. The synoptic types can be ranked by the highest energy production. It is shown that in this spatial area, monthly wind energy production is higher during the winter months. When the well-established high-pressure cells move northward, synoptic types associated with higher energy production are frequent and include tropical and temperate disturbances across South Africa, patterns resembling a ridging anticyclone off the west coast of South Africa and low-pressure cells occurring to the north and south. Low energy producing patterns show characteristics of the high-pressure cells moving southwards producing fine weather and mildly disturbed conditions. The purpose of this methodology is that it provides the foundation required to derive long-term frequency changes of these synoptic weather systems using global climate model ensembles and thus changes in wind energy production.

Spatio-temporal insights for wind energy harvesting in South Africa

Directional Gaussian spatial processes for South African wind data

Benefits of spatio-temporal modelling for short term wind power forecasting at both individual and aggregated levels

Constructing low-dimensional stochastic wind models through hierarchical spatial temporal decomposition

A Non-Gaussian Spatio-Temporal Model for Daily Wind Speeds Based on a Multivariate Skew-t Distribution

Using the lasso method for space-time short-term wind speed predictions

Joint Modeling of Wind Speed and Power via a Nonparametric Approach

Spatial and temporal patterns of global onshore wind speed distribution

A spatio-temporal model for predicting wind speeds in Southern California

Vector Autoregressive Models with Spatially Structured Coefficients for Time Series on a Spatial Grid

Spatio-temporal estimation of wind speed and wind power using machine learning: predictions, uncertainty and technical potential

Spatial Temporal Approach for High-Resolution Gridded Wind Forecasting across Southwest Western Australia

Spatio-Temporal Modeling of Wind Speed Using EOF and Machine Learning

Spatiotemporal analysis of bidimensional wind speed forecasting: Development and thorough assessment of LSTM and ensemble graph neural networks on the Dutch database

Wind energy potential of weather systems affecting South Africa’s Eastern Cape Province

Bayesian spatial modelling of geostatistical data using INLA and SPDE methods: A case study predicting malaria risk in Mozambique

AIRU-WRF: A Physics-Guided Spatio-Temporal Wind Forecasting Model and its Application to the U.S. Mid Atlantic Offshore Wind Energy Areas

Improving probabilistic wind speed forecasting using M-Rice distribution and spatial data integration

Inference for spatial processes using imperfect data from measurements and numerical simulations

A CFD Model for Spatial Extrapolation of Wind Field over Complex Terrain—Wi.Sp.Ex

Modelling potential land suitability of large-scale wind energy development using explainable machine learning techniques: Applications for China, USA and EU