Gregor Skok
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Gregor Skok has a total of 25 co-authors. Here are the top co-authors by number of shared publications:
Katarina Kosovelj
University of Ljubljana, Faculty of Mathematics and Physics, Jadranska cesta 19, 1000 Ljubljana, Slovenia
4 shared publications- Žiga Zaplotnik
Faculty of Mathematics and Physics University of Ljubljana Ljubljana Slovenia
3 shared publications Danijela Kuzmanović
Faculty of Mathematics and Physics, University of Ljubljana, Ljubljana, Slovenia
2 shared publicationsJana Banko
Slovenian Environment Agency, Ljubljana, Slovenia
2 shared publicationsNigel Roberts
MetOffice@Reading, Met Office UK
2 shared publicationsAndrej Ceglar
Center of Excellence “SPACE-SI
1 shared publication
Latest Publications
- Open Access
- 2 June 2026
- Published by Wiley in Quarterly Journal of the Royal Meteorological Society
- Vol. 152 (779) , e70212
Forecast‐Error Diagnostics in Neural Weather Models
Deep learning (DL) weather prediction models offer notable advantages over traditional physics‐based models, including auto‐differentiability and low computational cost, enabling detailed diagnostics of forecast errors. Using our convolutional encoder–decoder model, ConvCastNet, we relax selected subdomains of the forecast fields systematically towards “true” weather states (ECMWF ERA5 reanalysis) and monitor the forecast skill gain in other regions. Our results show that a medium‐range midlatitude forecast improves substantially when the stratosphere and boundary layer are relaxed, while relaxation of the tropical atmosphere has a limited effect. This underscores the need for a more accurate representation of the stratosphere and the planetary boundary layer to improve medium‐range weather predictability. Additionally, we investigate the relationship between the forecast‐error sensitivity to initial conditions and relaxation experiments. By utilising auto‐differentiability, we identify overlapping regions of large error sensitivity and strong forecast skill improvement from relaxation. Average midlatitude error sensitivity to initial conditions shows negligible influence from the Tropics, corroborating the results of the tropical relaxation experiments. The error sensitivity shows a physically consistent influence of upstream weather dynamics and sea‐surface temperatures on forecast accuracy. The latter also highlights the importance of representing the atmosphere–ocean coupling accurately in numerical weather prediction models. This combined approach could provide valuable heuristics for diagnosing neural model errors and guiding targeted model improvements.
- 2 June 2026
- Published by Wiley in Quarterly Journal of the Royal Meteorological Society
- Vol. 152 (779) , e70212
Open AccessForecast‐Error Diagnostics in Neural Weather Models
Deep learning (DL) weather prediction models offer notable advantages over traditional physics‐based models, including auto‐differentiability and low computational cost, enabling detailed diagnostics of forecast errors. Using our convolutional encoder–decoder model, ConvCastNet, we relax selected subdomains of the forecast fields systematically towards “true” weather states (ECMWF ERA5 reanalysis) and monitor the forecast skill gain in other regions. Our results show that a medium‐range midlatitude forecast improves substantially when the stratosphere and boundary layer are relaxed, while relaxation of the tropical atmosphere has a limited effect. This underscores the need for a more accurate representation of the stratosphere and the planetary boundary layer to improve medium‐range weather predictability. Additionally, we investigate the relationship between the forecast‐error sensitivity to initial conditions and relaxation experiments. By utilising auto‐differentiability, we identify overlapping regions of large error sensitivity and strong forecast skill improvement from relaxation. Average midlatitude error sensitivity to initial conditions shows negligible influence from the Tropics, corroborating the results of the tropical relaxation experiments. The error sensitivity shows a physically consistent influence of upstream weather dynamics and sea‐surface temperatures on forecast accuracy. The latter also highlights the importance of representing the atmosphere–ocean coupling accurately in numerical weather prediction models. This combined approach could provide valuable heuristics for diagnosing neural model errors and guiding targeted model improvements.
- Open Access
- 20 October 2025
- Published by Copernicus GmbH in Geoscientific Model Development
- Vol. 18 (20) , 7417-7433
Smoothing and spatial verification of global fields
Forecast verification plays a crucial role in the development cycle of operational numerical weather prediction models. At the same time, verification remains a challenge as the traditionally used non-spatial forecast quality metrics exhibit certain drawbacks, with new spatial metrics being developed to address these problems. Some of these new metrics are based on smoothing, with one example being the widely used Fraction Skill Score (FSS) and its many derivatives. However, while the FSS has been used by many researchers in limited area domains, there are no examples of it being used in a global domain yet. The issue is due to the increased computational complexity of smoothing in a global domain, with its inherent spherical geometry and non-equidistant and/or irregular grids. At the same time, there clearly exists a need for spatial metrics that could be used in the global domain as the operational global models continue to be developed and improved, along with the new machine-learning-based models. Here, we present two new methodologies for smoothing in a global domain that are potentially fast enough to make the smoothing of high-resolution global fields feasible. Both approaches also consider the variability of grid point area sizes and can handle missing data appropriately. This, in turn, makes the calculation of smoothing-based metrics, such as FSS and its derivatives, in a global domain possible, which we demonstrate by evaluating the performance of operational high-resolution global precipitation forecasts provided by the European Centre for Medium-Range Weather Forecasts.
- 20 October 2025
- Published by Copernicus GmbH in Geoscientific Model Development
- Vol. 18 (20) , 7417-7433
Open AccessSmoothing and spatial verification of global fields
Forecast verification plays a crucial role in the development cycle of operational numerical weather prediction models. At the same time, verification remains a challenge as the traditionally used non-spatial forecast quality metrics exhibit certain drawbacks, with new spatial metrics being developed to address these problems. Some of these new metrics are based on smoothing, with one example being the widely used Fraction Skill Score (FSS) and its many derivatives. However, while the FSS has been used by many researchers in limited area domains, there are no examples of it being used in a global domain yet. The issue is due to the increased computational complexity of smoothing in a global domain, with its inherent spherical geometry and non-equidistant and/or irregular grids. At the same time, there clearly exists a need for spatial metrics that could be used in the global domain as the operational global models continue to be developed and improved, along with the new machine-learning-based models. Here, we present two new methodologies for smoothing in a global domain that are potentially fast enough to make the smoothing of high-resolution global fields feasible. Both approaches also consider the variability of grid point area sizes and can handle missing data appropriately. This, in turn, makes the calculation of smoothing-based metrics, such as FSS and its derivatives, in a global domain possible, which we demonstrate by evaluating the performance of operational high-resolution global precipitation forecasts provided by the European Centre for Medium-Range Weather Forecasts.
- Open Access
- 29 August 2025
- Published by Copernicus GmbH
- Vol. 2025, 1
Responses to reviwers comments
Forecast verification plays a crucial role in the development cycle of operational numerical weather prediction models. At the same time, verification remains a challenge as the traditionally used non-spatial forecast quality metrics exhibit certain drawbacks, with new spatial metrics being developed to address these problems. Some of these new metrics are based on smoothing, with one example being the widely used Fraction Skill Score (FSS) and its many derivatives. However, while the FSS has been used by many researchers in limited area domains, there are, as of yet, no examples of it being used in a global domain. The issue is due to the increased computational complexity of smoothing in a global domain, with its inherent spherical geometry and non-equidistant and/or irregular grids. At the same time, there clearly exists a need for spatial metrics that could be used in the global domain as the operational global models continue to be developed and improved along with the new machine-learning-based models. Here, we present two new methodologies for smoothing in a global domain that are potentially fast enough to make the smoothing of high-resolution global fields feasible. Both approaches also consider the variability of grid point area sizes and can handle missing data appropriately. This, in turn, makes the calculation of smoothing-based metrics, such as FSS and its derivatives, in a global domain possible, which we demonstrate by evaluating the performance of operational high-resolution global precipitation forecasts provided by the European Centre for Medium-Range Weather Forecasts.
- 29 August 2025
- Published by Copernicus GmbH
- Vol. 2025, 1
Open AccessResponses to reviwers comments
Forecast verification plays a crucial role in the development cycle of operational numerical weather prediction models. At the same time, verification remains a challenge as the traditionally used non-spatial forecast quality metrics exhibit certain drawbacks, with new spatial metrics being developed to address these problems. Some of these new metrics are based on smoothing, with one example being the widely used Fraction Skill Score (FSS) and its many derivatives. However, while the FSS has been used by many researchers in limited area domains, there are, as of yet, no examples of it being used in a global domain. The issue is due to the increased computational complexity of smoothing in a global domain, with its inherent spherical geometry and non-equidistant and/or irregular grids. At the same time, there clearly exists a need for spatial metrics that could be used in the global domain as the operational global models continue to be developed and improved along with the new machine-learning-based models. Here, we present two new methodologies for smoothing in a global domain that are potentially fast enough to make the smoothing of high-resolution global fields feasible. Both approaches also consider the variability of grid point area sizes and can handle missing data appropriately. This, in turn, makes the calculation of smoothing-based metrics, such as FSS and its derivatives, in a global domain possible, which we demonstrate by evaluating the performance of operational high-resolution global precipitation forecasts provided by the European Centre for Medium-Range Weather Forecasts.
- Open Access
- 16 July 2025
- Published by Copernicus GmbH
Spatial verification of global precipitation forecasts
Verification of global high-resolution precipitation forecasts is challenging. Spatial verification techniques address some shortcomings of traditional verification. However most existing methods do not account for the non-planar geometry of a global domain, or their computational complexity is too large for global assessments. We present an adaptation of the recently developed Precipitation Attribution Distance (PAD) metric, designed for verifying precipitation, enabling its use on the Earth's spherical geometry. PAD estimates the magnitude of location errors in the forecasts employing the mathematical theory of Optimal Transport, as it provides a close upper bound for the Wasserstein distance. The method is fast and flexible with time complexity O(n log(n)). Its behavior is analyzed using idealized cases and 7 years of operational global deterministic 6-hourly precipitation forecasts from the Integrated Forecasting System (IFS) of the European Centre for Medium-Range Weather Forecasts. Due to the lack of availability of high-resolution, high-quality global precipitation observations, we employed the short-term forecasts by the same model as verification truth. The summary results for the whole period show how location errors in the IFS model grow steadily with increasing lead time for all analyzed regions. Moreover, by examining the time-series of the results, we can determine the trends in the score's value and identify the regions where there is a statistically significant improvement (or worsening) of the forecast performance. The results can also be analyzed separately for different intensities of precipitation. Overall, the PAD provides meaningful results for estimating location errors in global high-resolution precipitation forecasts at an affordable computational cost.
- 16 July 2025
- Published by Copernicus GmbH
Open AccessSpatial verification of global precipitation forecasts
Verification of global high-resolution precipitation forecasts is challenging. Spatial verification techniques address some shortcomings of traditional verification. However most existing methods do not account for the non-planar geometry of a global domain, or their computational complexity is too large for global assessments. We present an adaptation of the recently developed Precipitation Attribution Distance (PAD) metric, designed for verifying precipitation, enabling its use on the Earth's spherical geometry. PAD estimates the magnitude of location errors in the forecasts employing the mathematical theory of Optimal Transport, as it provides a close upper bound for the Wasserstein distance. The method is fast and flexible with time complexity O(n log(n)). Its behavior is analyzed using idealized cases and 7 years of operational global deterministic 6-hourly precipitation forecasts from the Integrated Forecasting System (IFS) of the European Centre for Medium-Range Weather Forecasts. Due to the lack of availability of high-resolution, high-quality global precipitation observations, we employed the short-term forecasts by the same model as verification truth. The summary results for the whole period show how location errors in the IFS model grow steadily with increasing lead time for all analyzed regions. Moreover, by examining the time-series of the results, we can determine the trends in the score's value and identify the regions where there is a statistically significant improvement (or worsening) of the forecast performance. The results can also be analyzed separately for different intensities of precipitation. Overall, the PAD provides meaningful results for estimating location errors in global high-resolution precipitation forecasts at an affordable computational cost.
- Open Access
- 26 May 2025
- Published by Copernicus GmbH
- Vol. 2025, 1
Comment on egusphere-2025-1525
Forecast verification plays a crucial role in the development cycle of operational numerical weather prediction models. At the same time, verification remains a challenge as the traditionally used non-spatial forecast quality metrics exhibit certain drawbacks, with new spatial metrics being developed to address these problems. Some of these new metrics are based on smoothing, with one example being the widely used Fraction Skill Score (FSS) and its many derivatives. However, while the FSS has been used by many researchers in limited area domains, there are, as of yet, no examples of it being used in a global domain. The issue is due to the increased computational complexity of smoothing in a global domain, with its inherent spherical geometry and non-equidistant and/or irregular grids. At the same time, there clearly exists a need for spatial metrics that could be used in the global domain as the operational global models continue to be developed and improved along with the new machine-learning-based models. Here, we present two new methodologies for smoothing in a global domain that are potentially fast enough to make the smoothing of high-resolution global fields feasible. Both approaches also consider the variability of grid point area sizes and can handle missing data appropriately. This, in turn, makes the calculation of smoothing-based metrics, such as FSS and its derivatives, in a global domain possible, which we demonstrate by evaluating the performance of operational high-resolution global precipitation forecasts provided by the European Centre for Medium-Range Weather Forecasts.
- 26 May 2025
- Published by Copernicus GmbH
- Vol. 2025, 1
Open AccessComment on egusphere-2025-1525
Forecast verification plays a crucial role in the development cycle of operational numerical weather prediction models. At the same time, verification remains a challenge as the traditionally used non-spatial forecast quality metrics exhibit certain drawbacks, with new spatial metrics being developed to address these problems. Some of these new metrics are based on smoothing, with one example being the widely used Fraction Skill Score (FSS) and its many derivatives. However, while the FSS has been used by many researchers in limited area domains, there are, as of yet, no examples of it being used in a global domain. The issue is due to the increased computational complexity of smoothing in a global domain, with its inherent spherical geometry and non-equidistant and/or irregular grids. At the same time, there clearly exists a need for spatial metrics that could be used in the global domain as the operational global models continue to be developed and improved along with the new machine-learning-based models. Here, we present two new methodologies for smoothing in a global domain that are potentially fast enough to make the smoothing of high-resolution global fields feasible. Both approaches also consider the variability of grid point area sizes and can handle missing data appropriately. This, in turn, makes the calculation of smoothing-based metrics, such as FSS and its derivatives, in a global domain possible, which we demonstrate by evaluating the performance of operational high-resolution global precipitation forecasts provided by the European Centre for Medium-Range Weather Forecasts.
- Open Access
- 26 May 2025
- Published by Copernicus GmbH
- Vol. 2025, 1
Comment on egusphere-2025-1525
Forecast verification plays a crucial role in the development cycle of operational numerical weather prediction models. At the same time, verification remains a challenge as the traditionally used non-spatial forecast quality metrics exhibit certain drawbacks, with new spatial metrics being developed to address these problems. Some of these new metrics are based on smoothing, with one example being the widely used Fraction Skill Score (FSS) and its many derivatives. However, while the FSS has been used by many researchers in limited area domains, there are, as of yet, no examples of it being used in a global domain. The issue is due to the increased computational complexity of smoothing in a global domain, with its inherent spherical geometry and non-equidistant and/or irregular grids. At the same time, there clearly exists a need for spatial metrics that could be used in the global domain as the operational global models continue to be developed and improved along with the new machine-learning-based models. Here, we present two new methodologies for smoothing in a global domain that are potentially fast enough to make the smoothing of high-resolution global fields feasible. Both approaches also consider the variability of grid point area sizes and can handle missing data appropriately. This, in turn, makes the calculation of smoothing-based metrics, such as FSS and its derivatives, in a global domain possible, which we demonstrate by evaluating the performance of operational high-resolution global precipitation forecasts provided by the European Centre for Medium-Range Weather Forecasts.
- 26 May 2025
- Published by Copernicus GmbH
- Vol. 2025, 1
Open AccessComment on egusphere-2025-1525
Forecast verification plays a crucial role in the development cycle of operational numerical weather prediction models. At the same time, verification remains a challenge as the traditionally used non-spatial forecast quality metrics exhibit certain drawbacks, with new spatial metrics being developed to address these problems. Some of these new metrics are based on smoothing, with one example being the widely used Fraction Skill Score (FSS) and its many derivatives. However, while the FSS has been used by many researchers in limited area domains, there are, as of yet, no examples of it being used in a global domain. The issue is due to the increased computational complexity of smoothing in a global domain, with its inherent spherical geometry and non-equidistant and/or irregular grids. At the same time, there clearly exists a need for spatial metrics that could be used in the global domain as the operational global models continue to be developed and improved along with the new machine-learning-based models. Here, we present two new methodologies for smoothing in a global domain that are potentially fast enough to make the smoothing of high-resolution global fields feasible. Both approaches also consider the variability of grid point area sizes and can handle missing data appropriately. This, in turn, makes the calculation of smoothing-based metrics, such as FSS and its derivatives, in a global domain possible, which we demonstrate by evaluating the performance of operational high-resolution global precipitation forecasts provided by the European Centre for Medium-Range Weather Forecasts.
- Open Access
- 12 May 2025
- Published by Wiley in Quarterly Journal of the Royal Meteorological Society
- Vol. 151 (771)
Spatial verification of global precipitation forecasts
Verification of global high‐resolution precipitation forecasts is challenging. Spatial verification techniques address some shortcomings of traditional verification. However, most existing methods do not account for the non‐planar geometry of a global domain, or their computational complexity is too large for global assessments. We present an adaptation of the recently developed precipitation attribution distance (PAD) metric, designed for verifying precipitation, enabling its use on the Earth's spherical geometry. PAD estimates the magnitude of location errors in the forecasts employing the mathematical theory of optimal transport, as it provides a close upper bound for the Wasserstein distance. The method is fast and flexible with time complexity . Its behavior is analyzed using idealized cases and 7 years of operational global deterministic six‐hourly precipitation forecasts from the Integrated Forecasting System (IFS) of the European Centre for Medium‐Range Weather Forecasts. The summary results for the whole period show how location errors in the IFS model grow steadily with increasing lead time for all regions analyzed. Moreover, by examining the time series of the results, we can determine the trends in the score's value and identify the regions where there is a statistically significant improvement (or worsening) of the forecast performance. The results can also be analyzed separately for different intensities of precipitation. Overall, the PAD provides meaningful results for estimating location errors in global high‐resolution precipitation forecasts at an affordable computational cost.
- 12 May 2025
- Published by Wiley in Quarterly Journal of the Royal Meteorological Society
- Vol. 151 (771)
Open AccessSpatial verification of global precipitation forecasts
Verification of global high‐resolution precipitation forecasts is challenging. Spatial verification techniques address some shortcomings of traditional verification. However, most existing methods do not account for the non‐planar geometry of a global domain, or their computational complexity is too large for global assessments. We present an adaptation of the recently developed precipitation attribution distance (PAD) metric, designed for verifying precipitation, enabling its use on the Earth's spherical geometry. PAD estimates the magnitude of location errors in the forecasts employing the mathematical theory of optimal transport, as it provides a close upper bound for the Wasserstein distance. The method is fast and flexible with time complexity . Its behavior is analyzed using idealized cases and 7 years of operational global deterministic six‐hourly precipitation forecasts from the Integrated Forecasting System (IFS) of the European Centre for Medium‐Range Weather Forecasts. The summary results for the whole period show how location errors in the IFS model grow steadily with increasing lead time for all regions analyzed. Moreover, by examining the time series of the results, we can determine the trends in the score's value and identify the regions where there is a statistically significant improvement (or worsening) of the forecast performance. The results can also be analyzed separately for different intensities of precipitation. Overall, the PAD provides meaningful results for estimating location errors in global high‐resolution precipitation forecasts at an affordable computational cost.
- Open Access
- 23 December 2024
- Published by Wiley in Geoscience Data Journal
- Vol. 12 (2)
A dataset of tracked mesoscale precipitation systems in the tropics
Mesoscale Convective Systems (MCSs) are often quantified via surface‐based radar network, geostationary satellite, or low earth orbit satellite observations. However, each of these has drawbacks for detecting cloud systems such as a lack of global coverage, a lack of variables to quantify deep convective cloud and precipitation properties, and a lack of continuous observations of individual MCSs, respectively. To generate a dataset of tropical Tracked IMERG Mesoscale Precipitation Systems (TIMPS), we use the Forward in Time tracking algorithm to track precipitation systems in the Integrated Multi‐satellitE Retrievals for the Global Precipitation Mission (IMERG). IMERG is a global gridded precipitation product that incorporates observations from a constellation of satellites with passive microwave sensors and other sources, allowing the TIMPS dataset to have continuous temporal precipitation information for MCSs in a global tropical strip with data every 30 min in time and 0.1° in space. TIMPS are provided in a publicly available data base with a variety of variables including MCS size, motion, and precipitation properties, estimations of MCS life cycle stages, and their proximity to the nearest tropical cyclone. By combining the TIMPS dataset with the University of Washington Convective Features database, we also provide snapshots of information from more spatially detailed space‐borne radar coverage. The TIMPS dataset provides the means for detailed long‐term and large‐scale study of MCSs in all regions of the tropics with applications such as composite studies of MCS life cycles and the evaluation of model performance.
- 23 December 2024
- Published by Wiley in Geoscience Data Journal
- Vol. 12 (2)
Open AccessA dataset of tracked mesoscale precipitation systems in the tropics
Mesoscale Convective Systems (MCSs) are often quantified via surface‐based radar network, geostationary satellite, or low earth orbit satellite observations. However, each of these has drawbacks for detecting cloud systems such as a lack of global coverage, a lack of variables to quantify deep convective cloud and precipitation properties, and a lack of continuous observations of individual MCSs, respectively. To generate a dataset of tropical Tracked IMERG Mesoscale Precipitation Systems (TIMPS), we use the Forward in Time tracking algorithm to track precipitation systems in the Integrated Multi‐satellitE Retrievals for the Global Precipitation Mission (IMERG). IMERG is a global gridded precipitation product that incorporates observations from a constellation of satellites with passive microwave sensors and other sources, allowing the TIMPS dataset to have continuous temporal precipitation information for MCSs in a global tropical strip with data every 30 min in time and 0.1° in space. TIMPS are provided in a publicly available data base with a variety of variables including MCS size, motion, and precipitation properties, estimations of MCS life cycle stages, and their proximity to the nearest tropical cyclone. By combining the TIMPS dataset with the University of Washington Convective Features database, we also provide snapshots of information from more spatially detailed space‐borne radar coverage. The TIMPS dataset provides the means for detailed long‐term and large‐scale study of MCSs in all regions of the tropics with applications such as composite studies of MCS life cycles and the evaluation of model performance.
- Open Access
- 5 March 2024
- Published by Springer Nature in International Journal of Biometeorology
- Vol. 68 (5) , 965-977
Improving the operational forecasts of outdoor Universal Thermal Climate Index with post-processing
The Universal Thermal Climate Index (UTCI) is a thermal comfort index that describes how the human body experiences ambient conditions. It has units of temperature and considers physiological aspects of the human body. It takes into account the effect of air temperature, humidity, wind, radiation, and clothes. It is increasingly used in many countries as a measure of thermal comfort for outdoor conditions, and its value is calculated as part of the operational meteorological forecast. At the same time, forecasts of outdoor UTCI tend to have a relatively large error caused by the error of meteorological forecasts. In Slovenia, there is a relatively dense network of meteorological stations. Crucially, at these stations, global solar radiation measurements are performed continuously, which makes estimating the actual value of the UTCI more accurate compared to the situation where no radiation measurements are available. We used seven years of measurements in hourly resolution from 42 stations to first verify the operational UTCI forecast for the first forecast day and, secondly, to try to improve the forecast via post-processing. We used two machine-learning methods, linear regression, and neural networks. Both methods have successfully reduced the error in the operational UTCI forecasts. Both methods reduced the daily mean error from about 2.6 $$^{\circ }$$ C to almost zero, while the daily mean absolute error decreased from 5 $$^{\circ }$$ C to 3 $$^{\circ }$$ C for the neural network and 3.5 $$^{\circ }$$ C for linear regression. Both methods, especially the neural network, also substantially reduced the dependence of the error on the time of the day.- 5 March 2024
- Published by Springer Nature in International Journal of Biometeorology
- Vol. 68 (5) , 965-977
Open AccessImproving the operational forecasts of outdoor Universal Thermal Climate Index with post-processing
The Universal Thermal Climate Index (UTCI) is a thermal comfort index that describes how the human body experiences ambient conditions. It has units of temperature and considers physiological aspects of the human body. It takes into account the effect of air temperature, humidity, wind, radiation, and clothes. It is increasingly used in many countries as a measure of thermal comfort for outdoor conditions, and its value is calculated as part of the operational meteorological forecast. At the same time, forecasts of outdoor UTCI tend to have a relatively large error caused by the error of meteorological forecasts. In Slovenia, there is a relatively dense network of meteorological stations. Crucially, at these stations, global solar radiation measurements are performed continuously, which makes estimating the actual value of the UTCI more accurate compared to the situation where no radiation measurements are available. We used seven years of measurements in hourly resolution from 42 stations to first verify the operational UTCI forecast for the first forecast day and, secondly, to try to improve the forecast via post-processing. We used two machine-learning methods, linear regression, and neural networks. Both methods have successfully reduced the error in the operational UTCI forecasts. Both methods reduced the daily mean error from about 2.6 $$^{\circ }$$ C to almost zero, while the daily mean absolute error decreased from 5 $$^{\circ }$$ C to 3 $$^{\circ }$$ C for the neural network and 3.5 $$^{\circ }$$ C for linear regression. Both methods, especially the neural network, also substantially reduced the dependence of the error on the time of the day. - Open Access
- 6 July 2023
- Published by Copernicus GmbH
Improving the operational forecasts of outdoor UTCI with post-processing
Universal Thermal Climate Index (UTCI ) is a thermal comfort index that describes how the human body experiences ambient conditions. It has units of temperature, and it takes into account the effect of air temperature, humidity, wind, radiation, and clothes. It is increasingly used in many countries as a measure of thermal comfort for outdoor conditions, and its value is calculated together with the operational meteorological forecast. At the same time, forecasts of outdoor UTCI tend to have a relatively large error caused by the error of meteorological forecasts. In Slovenia, there is a relatively dense network of meteorological stations. Crucially, on these stations, global solar radiation measurements are performed, making estimating the true value of UTCI more reliable. We used seven years of measurements in hourly resolution from 42 stations to first verify the operational UTCI forecast for the first forecast day and, secondly, to try to improve the forecast via post-processing. The verification showed that the operational ALADIN model tends to overestimate the UTCI values. The overestimation is most pronounced in the morning when the mean error is about 8 °C. Also, a small percentage of cases have a very large error (up to 40 °C). We used two machine-learning methods to try to improve the forecasts of UTCI: linear regression and neural networks. Both methods have successfully reduced the error in the operational UTCI forecasts. Both reduced the daily mean error from about 2.6 °C to almost zero, while the daily mean absolute error decreased from 5 °C to 3 °C for the neural network and 3.5 °C for linear regression.
- 6 July 2023
- Published by Copernicus GmbH
Open AccessImproving the operational forecasts of outdoor UTCI with post-processing
Universal Thermal Climate Index (UTCI ) is a thermal comfort index that describes how the human body experiences ambient conditions. It has units of temperature, and it takes into account the effect of air temperature, humidity, wind, radiation, and clothes. It is increasingly used in many countries as a measure of thermal comfort for outdoor conditions, and its value is calculated together with the operational meteorological forecast. At the same time, forecasts of outdoor UTCI tend to have a relatively large error caused by the error of meteorological forecasts. In Slovenia, there is a relatively dense network of meteorological stations. Crucially, on these stations, global solar radiation measurements are performed, making estimating the true value of UTCI more reliable. We used seven years of measurements in hourly resolution from 42 stations to first verify the operational UTCI forecast for the first forecast day and, secondly, to try to improve the forecast via post-processing. The verification showed that the operational ALADIN model tends to overestimate the UTCI values. The overestimation is most pronounced in the morning when the mean error is about 8 °C. Also, a small percentage of cases have a very large error (up to 40 °C). We used two machine-learning methods to try to improve the forecasts of UTCI: linear regression and neural networks. Both methods have successfully reduced the error in the operational UTCI forecasts. Both reduced the daily mean error from about 2.6 °C to almost zero, while the daily mean absolute error decreased from 5 °C to 3 °C for the neural network and 3.5 °C for linear regression.