Browse
Abstract
As tropical cyclone (TC) official and model track forecasts improve, cases still exist where model position errors and uncertainty are large. The goal of this study is to identify forecast track bifurcations in both the Coupled Ocean-Atmosphere Mesoscale Prediction System-Tropical Cyclone (COAMPS-TC) ensemble and the Global Ensemble Forecast System (GEFS), with subsequent analyses focusing on the comparison of the synoptic environments of the bifurcating cases between the ensembles.
Thirty-three bifurcating cases are identified in the COAMPS-TC ensemble, while thirty-eight are identified in the GEFS, over a time period spanning the 2020-2022 TC seasons in the Northern Hemisphere. For these bifurcating cases, early lead time position of an ensemble member relative to the ensemble mean shows little correlation to the 72 h members position relative to the ensemble mean. Rather, the TCs 72 h forecast position relative to the ensemble mean shows high sensitivity to the tropospheric deep-layer steering flow at early lead times, particularly in the COAMPS-TC ensemble. For both ensembles, a region of weak flow is located near the TCs at the forecast initial time. Minor differences in the steering flow at early lead times induced by variations in the synoptic-scale environment cause the TCs position relative to the col to vary across the ensembles members, resulting in a bifurcation. Differences in the results for the COAMPS-TC ensemble and GEFS are thought to be driven, in part, by the fact that the COAMPS-TC ensemble utilizes flow-independent initial-state perturbations, while the perturbations in the GEFS are flow-dependent.
Abstract
As tropical cyclone (TC) official and model track forecasts improve, cases still exist where model position errors and uncertainty are large. The goal of this study is to identify forecast track bifurcations in both the Coupled Ocean-Atmosphere Mesoscale Prediction System-Tropical Cyclone (COAMPS-TC) ensemble and the Global Ensemble Forecast System (GEFS), with subsequent analyses focusing on the comparison of the synoptic environments of the bifurcating cases between the ensembles.
Thirty-three bifurcating cases are identified in the COAMPS-TC ensemble, while thirty-eight are identified in the GEFS, over a time period spanning the 2020-2022 TC seasons in the Northern Hemisphere. For these bifurcating cases, early lead time position of an ensemble member relative to the ensemble mean shows little correlation to the 72 h members position relative to the ensemble mean. Rather, the TCs 72 h forecast position relative to the ensemble mean shows high sensitivity to the tropospheric deep-layer steering flow at early lead times, particularly in the COAMPS-TC ensemble. For both ensembles, a region of weak flow is located near the TCs at the forecast initial time. Minor differences in the steering flow at early lead times induced by variations in the synoptic-scale environment cause the TCs position relative to the col to vary across the ensembles members, resulting in a bifurcation. Differences in the results for the COAMPS-TC ensemble and GEFS are thought to be driven, in part, by the fact that the COAMPS-TC ensemble utilizes flow-independent initial-state perturbations, while the perturbations in the GEFS are flow-dependent.
Abstract
The time extended Empirical Orthogonal Function (EEOF) patterns of error developed in a companion paper are applied in this study for model data postprocessing. Projecting forecast anomalies from the model seasonal cycle from GEFSv12 outputs onto these EEOF patterns reconstructs the error principal components (PCs) in a manner that can be applied in realtime. When these reconstructed error anomalies are compared with original GEFSv12 200 hPa Geopotential Height (Z200) reforecast error anomalies, the projected and validation error signals tend to align temporally. When distributed globally, projection and validation error anomalies have statistically significant positive Pearson Correlation coefficients with each other at most grid points outside of the tropics. Categorical Heidke Skill Score (HSS) analysis reaffirms that by lead day 16 of a forecast, the algorithm skillfully predicts systematic errors for locations outside of the tropics. Hemispheric winter demonstrates the highest magnitude HSS values in the high latitudes, illustrating the underlying seasonality of Z200 behavior and predictability. On average, wintertime HSS values hover around 20%, with high latitude locations exceeding mean values of 40%. Comparatively summertime values generally exceed 10% improvement outside of the tropics. High latitude storm track regions, such as the North Atlantic, North Pacific, and South Pacific contain the most Z200 variance in the globe and the highest HSS values. This algorithm will be applied to realtime forecasts and additional variables in future work, with the end goal of implementation into the NOAA-CPC subseasonal forecasts.
Abstract
The time extended Empirical Orthogonal Function (EEOF) patterns of error developed in a companion paper are applied in this study for model data postprocessing. Projecting forecast anomalies from the model seasonal cycle from GEFSv12 outputs onto these EEOF patterns reconstructs the error principal components (PCs) in a manner that can be applied in realtime. When these reconstructed error anomalies are compared with original GEFSv12 200 hPa Geopotential Height (Z200) reforecast error anomalies, the projected and validation error signals tend to align temporally. When distributed globally, projection and validation error anomalies have statistically significant positive Pearson Correlation coefficients with each other at most grid points outside of the tropics. Categorical Heidke Skill Score (HSS) analysis reaffirms that by lead day 16 of a forecast, the algorithm skillfully predicts systematic errors for locations outside of the tropics. Hemispheric winter demonstrates the highest magnitude HSS values in the high latitudes, illustrating the underlying seasonality of Z200 behavior and predictability. On average, wintertime HSS values hover around 20%, with high latitude locations exceeding mean values of 40%. Comparatively summertime values generally exceed 10% improvement outside of the tropics. High latitude storm track regions, such as the North Atlantic, North Pacific, and South Pacific contain the most Z200 variance in the globe and the highest HSS values. This algorithm will be applied to realtime forecasts and additional variables in future work, with the end goal of implementation into the NOAA-CPC subseasonal forecasts.
Abstract
Analysis of seasonally varying signals of model error biases, as well as the diagnosis of space-time patterns of synoptic error, here is accomplished through the development of a statistical postprocessing algorithm based on time-extended Empirical Orthogonal Functions (EEOFs) built on patterns of variance. This work analyzes GEFSv12 200 hPa Geopotential Height (Z200) model error against ERA5 reanalysis data. The mean squared error variance between GEFSv12 reforecast and ERA5 grows rapidly after day 7 of a forecast, and continues to increase through the end of the 16 day forecast period. At lead time 16, the largest variance occurs in middle to high latitude oceanic storm tracks. Variance is highest during hemispheric winter, when baroclinic energy is abundant. The seasonal cycle of error shows largest anomalies in the high latitudes during hemispheric winter. After running the titular two-step, space-time EEOF algorithm, a standardized spatial eigenspectrum shows eigenvalues increase at longer lead times after decreasing in the medium range, demonstrating that the algorithm extracts large-scale signals of systematic error in the long range. Leading wintertime space-time eigenmode pairs include high latitude blocking structures and Rossby wave trains. Results suggest that the large-scale systematic errors in GEFSv12 Z200 initiate in part from inaccurate representations of phase speeds of atmospheric waves in the polar jet.
Abstract
Analysis of seasonally varying signals of model error biases, as well as the diagnosis of space-time patterns of synoptic error, here is accomplished through the development of a statistical postprocessing algorithm based on time-extended Empirical Orthogonal Functions (EEOFs) built on patterns of variance. This work analyzes GEFSv12 200 hPa Geopotential Height (Z200) model error against ERA5 reanalysis data. The mean squared error variance between GEFSv12 reforecast and ERA5 grows rapidly after day 7 of a forecast, and continues to increase through the end of the 16 day forecast period. At lead time 16, the largest variance occurs in middle to high latitude oceanic storm tracks. Variance is highest during hemispheric winter, when baroclinic energy is abundant. The seasonal cycle of error shows largest anomalies in the high latitudes during hemispheric winter. After running the titular two-step, space-time EEOF algorithm, a standardized spatial eigenspectrum shows eigenvalues increase at longer lead times after decreasing in the medium range, demonstrating that the algorithm extracts large-scale signals of systematic error in the long range. Leading wintertime space-time eigenmode pairs include high latitude blocking structures and Rossby wave trains. Results suggest that the large-scale systematic errors in GEFSv12 Z200 initiate in part from inaccurate representations of phase speeds of atmospheric waves in the polar jet.
Abstract
This study analyzes the operational predictability of warm-season (May-August) progressive derechos. A subset of 47 derechos occurring between 2010 and 2022 were selected based on NCEI Storm Data and radar imagery. The Storm Prediction Center Convective Outlooks issued from five days before each derecho to the day of the event were used to determine if the severe weather and derecho predictability was low, moderate, or high, for each case. The cases were also categorized based on the synoptic-scale midlevel flow pattern. Composite maps of synoptic patterns associated with the derecho cases and distributions of severe weather parameters in the derecho inflow region were generated. About 72% of the 47 derechos selected for this study were associated with low predictability, and the remainder were categorized with moderate predictability. None of the cases had high predictability. Most derechos occurring under northwest and zonal flow regimes (80% and 85%, respectively) had low predictability. Composites of low- and moderate-predictability derechos indicate that derechos formed in the equatorward entrance region of the upper-level jet, where low-level warm advection exists in conjunction with an equivalent potential temperature maximum. A midlevel trough is located upstream of the derecho initiation point in many of the moderate-predictability composites, but is absent in the low-predictability composites, which indicated weaker synoptic-scale forcing for ascent in low-predictability cases. The distributions of severe weather parameters for low- and moderate-predictability derechos are similar, suggesting that these parameters alone are not particularly useful as a discriminator of derecho predictability.
Abstract
This study analyzes the operational predictability of warm-season (May-August) progressive derechos. A subset of 47 derechos occurring between 2010 and 2022 were selected based on NCEI Storm Data and radar imagery. The Storm Prediction Center Convective Outlooks issued from five days before each derecho to the day of the event were used to determine if the severe weather and derecho predictability was low, moderate, or high, for each case. The cases were also categorized based on the synoptic-scale midlevel flow pattern. Composite maps of synoptic patterns associated with the derecho cases and distributions of severe weather parameters in the derecho inflow region were generated. About 72% of the 47 derechos selected for this study were associated with low predictability, and the remainder were categorized with moderate predictability. None of the cases had high predictability. Most derechos occurring under northwest and zonal flow regimes (80% and 85%, respectively) had low predictability. Composites of low- and moderate-predictability derechos indicate that derechos formed in the equatorward entrance region of the upper-level jet, where low-level warm advection exists in conjunction with an equivalent potential temperature maximum. A midlevel trough is located upstream of the derecho initiation point in many of the moderate-predictability composites, but is absent in the low-predictability composites, which indicated weaker synoptic-scale forcing for ascent in low-predictability cases. The distributions of severe weather parameters for low- and moderate-predictability derechos are similar, suggesting that these parameters alone are not particularly useful as a discriminator of derecho predictability.
Abstract
As artificial intelligence (AI) methods are increasingly used to develop new guidance intended for operational use by forecasters, it is critical to evaluate whether forecasters deem the guidance trustworthy. Past trust-related AI research suggests that certain attributes (e.g., understanding how the AI was trained, interactivity, and performance) contribute to users perceiving the AI as trustworthy. However, little research has been done to examine the role of these and other attributes for weather forecasters. In this study, we conducted 16 online interviews with National Weather Service (NWS) forecasters to examine (i) how they make guidance use decisions and (ii) how the AI model technique used, training, input variables, performance, and developers as well as interacting with the model output influenced their assessments of trustworthiness of new guidance. The interviews pertained to either a random forest model predicting the probability of severe hail or a 2D convolutional neural network model predicting the probability of storm mode. When taken as a whole, our findings illustrate how forecasters’ assessment of AI guidance trustworthiness is a process that occurs over time rather than automatically or at first introduction. We recommend developers center end users when creating new AI guidance tools, making end users integral to their thinking and efforts. This approach is essential for the development of useful and used tools. The details of these findings can help AI developers understand how forecasters perceive AI guidance and inform AI development and refinement efforts.
Significance Statement
We used a mixed-methods quantitative and qualitative approach to understand how National Weather Service (NWS) forecasters 1) make guidance use decisions within their operational forecasting process and 2) assess the trustworthiness of prototype guidance developed using artificial intelligence (AI). When taken as a whole, our findings illustrate that forecasters’ assessment of AI guidance trustworthiness is a process that occurs over time rather than automatically and suggest that developers must center the end user when creating new AI guidance tools to ensure that the developed tools are useful and used.
Abstract
As artificial intelligence (AI) methods are increasingly used to develop new guidance intended for operational use by forecasters, it is critical to evaluate whether forecasters deem the guidance trustworthy. Past trust-related AI research suggests that certain attributes (e.g., understanding how the AI was trained, interactivity, and performance) contribute to users perceiving the AI as trustworthy. However, little research has been done to examine the role of these and other attributes for weather forecasters. In this study, we conducted 16 online interviews with National Weather Service (NWS) forecasters to examine (i) how they make guidance use decisions and (ii) how the AI model technique used, training, input variables, performance, and developers as well as interacting with the model output influenced their assessments of trustworthiness of new guidance. The interviews pertained to either a random forest model predicting the probability of severe hail or a 2D convolutional neural network model predicting the probability of storm mode. When taken as a whole, our findings illustrate how forecasters’ assessment of AI guidance trustworthiness is a process that occurs over time rather than automatically or at first introduction. We recommend developers center end users when creating new AI guidance tools, making end users integral to their thinking and efforts. This approach is essential for the development of useful and used tools. The details of these findings can help AI developers understand how forecasters perceive AI guidance and inform AI development and refinement efforts.
Significance Statement
We used a mixed-methods quantitative and qualitative approach to understand how National Weather Service (NWS) forecasters 1) make guidance use decisions within their operational forecasting process and 2) assess the trustworthiness of prototype guidance developed using artificial intelligence (AI). When taken as a whole, our findings illustrate that forecasters’ assessment of AI guidance trustworthiness is a process that occurs over time rather than automatically and suggest that developers must center the end user when creating new AI guidance tools to ensure that the developed tools are useful and used.
Abstract
Forecasting the Week 3/4 period presents many challenges, resulting in a need for improvements to forecast skill. At this distance from initial conditions, numerical models struggle to present skillful forecasts of temperature, precipitation, and associated extremes. One approach to address this is to utilize more predictable large-scale circulation regimes to make forecasts of temperature and precipitation anomalies, using the association between the regimes and surface weather obtained from reanalysis products. This study explores the utility of k-means cluster analysis on geopotential heights and their ability to make skillful regime predictions in the Week 3/4 period. Using 14-day running means of European Centre for Medium-Range Weather Forecasts (ECMWF) Reanalysis v5 (ERA5) 500-hPa geopotential heights for the wintertime December-February (DJF) period, circulation regimes are identified using k-means clustering. Each period is assigned a cluster number, allowing compositing of any reanalysis or observation variable to form cluster maps. Maps of 500-hPa height, 2-m temperature, precipitation, and storm track anomalies are some of the variables composited. The utility of these relationships in a dynamical forecast setting is tested via Global Ensemble Forecast System v12 (GEFSv12) hindcasts and real-time ensemble suite forecasts. Week 3/4 deterministic and probabilistic experimental forecasts are then derived from cluster assignments using several methods. We find, via a conditional skill analysis, forecasts strongly correlated with a cluster exhibit greater skill for both dynamical model and cluster derived forecasts. Our preliminary results represent a step forward to aid forecasters make more skillful assessments of the circulation regime and its associated surface weather for this challenging forecast time scale.
Abstract
Forecasting the Week 3/4 period presents many challenges, resulting in a need for improvements to forecast skill. At this distance from initial conditions, numerical models struggle to present skillful forecasts of temperature, precipitation, and associated extremes. One approach to address this is to utilize more predictable large-scale circulation regimes to make forecasts of temperature and precipitation anomalies, using the association between the regimes and surface weather obtained from reanalysis products. This study explores the utility of k-means cluster analysis on geopotential heights and their ability to make skillful regime predictions in the Week 3/4 period. Using 14-day running means of European Centre for Medium-Range Weather Forecasts (ECMWF) Reanalysis v5 (ERA5) 500-hPa geopotential heights for the wintertime December-February (DJF) period, circulation regimes are identified using k-means clustering. Each period is assigned a cluster number, allowing compositing of any reanalysis or observation variable to form cluster maps. Maps of 500-hPa height, 2-m temperature, precipitation, and storm track anomalies are some of the variables composited. The utility of these relationships in a dynamical forecast setting is tested via Global Ensemble Forecast System v12 (GEFSv12) hindcasts and real-time ensemble suite forecasts. Week 3/4 deterministic and probabilistic experimental forecasts are then derived from cluster assignments using several methods. We find, via a conditional skill analysis, forecasts strongly correlated with a cluster exhibit greater skill for both dynamical model and cluster derived forecasts. Our preliminary results represent a step forward to aid forecasters make more skillful assessments of the circulation regime and its associated surface weather for this challenging forecast time scale.
Abstract
The current operational Global Ensemble Forecast System (GEFS) version 12 was implemented in National Centers for Environmental Prediction (NCEP) operations in September 2020. This is the first Unified Forecast System (UFS)-based GEFS which has demonstrated significant improvements in probabilistic forecast guidance at weather, medium-range and subseasonal scales based on the evaluation of 31-year reforecasts and 3-year retrospective forecasts. NOAA/NCEP is now planning to implement a fully coupled UFS Global Forecast System (GFS) and GEFS in 2025. Recently we have been working on the fully coupled GEFS versions following the development of UFS coupled model prototypes. Besides some updates from the model components like land model, ocean model and ice model, there are significant atmospheric model upgrades including increasing vertical resolution, PBL scheme, microphysics and gravity wave drag following the development of these model prototypes.
In this paper, we present two coupled GEFS experiments that are based on UFS coupled model prototype 5 and 8 respectively. This paper aims to evaluate and compare the results from these coupled ensemble prototypes experiments as well as stochastic physics application, configuration and tuning progress in these coupled ensemble experiments for weather and subseasonal forecast. The results from those two coupled GEFS experiments show the improvements of the 500 hPa geopotential height probabilistic forecast skills (CRPSS) and model bias for all lead-time, the improvements of the precipitation forecast on the weather time scale of reliability and Hedike scale for weekly average including weeks 3&4, the skill-error ratios and MJO prediction as well.
Abstract
The current operational Global Ensemble Forecast System (GEFS) version 12 was implemented in National Centers for Environmental Prediction (NCEP) operations in September 2020. This is the first Unified Forecast System (UFS)-based GEFS which has demonstrated significant improvements in probabilistic forecast guidance at weather, medium-range and subseasonal scales based on the evaluation of 31-year reforecasts and 3-year retrospective forecasts. NOAA/NCEP is now planning to implement a fully coupled UFS Global Forecast System (GFS) and GEFS in 2025. Recently we have been working on the fully coupled GEFS versions following the development of UFS coupled model prototypes. Besides some updates from the model components like land model, ocean model and ice model, there are significant atmospheric model upgrades including increasing vertical resolution, PBL scheme, microphysics and gravity wave drag following the development of these model prototypes.
In this paper, we present two coupled GEFS experiments that are based on UFS coupled model prototype 5 and 8 respectively. This paper aims to evaluate and compare the results from these coupled ensemble prototypes experiments as well as stochastic physics application, configuration and tuning progress in these coupled ensemble experiments for weather and subseasonal forecast. The results from those two coupled GEFS experiments show the improvements of the 500 hPa geopotential height probabilistic forecast skills (CRPSS) and model bias for all lead-time, the improvements of the precipitation forecast on the weather time scale of reliability and Hedike scale for weekly average including weeks 3&4, the skill-error ratios and MJO prediction as well.
Abstract
Forecast models based on the gust factor, the ratio of peak gust to sustained wind speed, have shown promise in predicting peak wind gusts in recent years. These models assume that turbulent vertical transport forced by wind flowing over upstream terrain mixes high-momentum air aloft down to the surface. A recently constructed database of hourly peak gusts, together with approximately coincident, nearby upper-air wind observations, is used to identify mixdown altitudes, the altitudes from which peak gusts descend, during different weather scenarios at 16 locations across the United States. Median mixdown altitudes generally ranged from 50 to 450 m AGL with occasional exceptions, particularly for convective gusts and at mountainous locations where terrain effects are likely to amplify gusts. A mixdown model in which surface peak gusts are predicted by obtaining forecast upper-air winds within this altitude interval was developed and tested. Our results suggest that a mixdown model methodology for forecasting peak gusts may be feasible at locations and during weather conditions where terrain-forced turbulent mixing is the principal cause of wind gusts.
Abstract
Forecast models based on the gust factor, the ratio of peak gust to sustained wind speed, have shown promise in predicting peak wind gusts in recent years. These models assume that turbulent vertical transport forced by wind flowing over upstream terrain mixes high-momentum air aloft down to the surface. A recently constructed database of hourly peak gusts, together with approximately coincident, nearby upper-air wind observations, is used to identify mixdown altitudes, the altitudes from which peak gusts descend, during different weather scenarios at 16 locations across the United States. Median mixdown altitudes generally ranged from 50 to 450 m AGL with occasional exceptions, particularly for convective gusts and at mountainous locations where terrain effects are likely to amplify gusts. A mixdown model in which surface peak gusts are predicted by obtaining forecast upper-air winds within this altitude interval was developed and tested. Our results suggest that a mixdown model methodology for forecasting peak gusts may be feasible at locations and during weather conditions where terrain-forced turbulent mixing is the principal cause of wind gusts.
Abstract
The Laseyer is a very local and uncommon wind storm in a narrow and steep valley in northeastern Switzerland. Whereas the ambient wind is from west to north-west, the strong surface wind in the valley is from the east, leading to gust speeds that become dangerous for the local train running into the valley to the Wasserauen station. To minimize the risk of derailment and to improve passenger comfort, the train service provider Appenzeller Bahnen (AB) has developed a new warning algorithm in close collaboration with academia (ETH Zurich) and the Swiss national weather service (Meteoswiss). The aim is to accurately predict the Laseyer wind storm several hours in advance, but also to reduce the number of false alarms. The new warning system is based on the Meteoswiss operational ensemble prediction system at 1.1 km horizontal mesh size, which is then used in combination with an observation-based machine learning approach to probabilistically forecast Laseyer events up to 30 hours in advance. A particular challenge in developing the new system was to introduce the customer, AB, to the modern concept of probabilistic numerical weather prediction, which requires a careful risk assessment by the customer. Hence, the development of the warning system is a process in which the customer and the warning provider closely collaborate and specify the final warning products to be delivered operationally. The operation of the new warning system during the 2021-22 Laseyer season shows that it is working successfully, and also indicates that the warning thresholds in the warning algorithm can be adjusted in the future to minimize false alarms without increasing the number of in missed events.
Abstract
The Laseyer is a very local and uncommon wind storm in a narrow and steep valley in northeastern Switzerland. Whereas the ambient wind is from west to north-west, the strong surface wind in the valley is from the east, leading to gust speeds that become dangerous for the local train running into the valley to the Wasserauen station. To minimize the risk of derailment and to improve passenger comfort, the train service provider Appenzeller Bahnen (AB) has developed a new warning algorithm in close collaboration with academia (ETH Zurich) and the Swiss national weather service (Meteoswiss). The aim is to accurately predict the Laseyer wind storm several hours in advance, but also to reduce the number of false alarms. The new warning system is based on the Meteoswiss operational ensemble prediction system at 1.1 km horizontal mesh size, which is then used in combination with an observation-based machine learning approach to probabilistically forecast Laseyer events up to 30 hours in advance. A particular challenge in developing the new system was to introduce the customer, AB, to the modern concept of probabilistic numerical weather prediction, which requires a careful risk assessment by the customer. Hence, the development of the warning system is a process in which the customer and the warning provider closely collaborate and specify the final warning products to be delivered operationally. The operation of the new warning system during the 2021-22 Laseyer season shows that it is working successfully, and also indicates that the warning thresholds in the warning algorithm can be adjusted in the future to minimize false alarms without increasing the number of in missed events.
Abstract
The paper presents the development of a high-resolution mesoscale atmospheric numerical model Advanced Research WRF (ARW 3.7, henceforth ARW)-based operational forecast system, and evaluation of its surface wind forecasts using buoys and satellite data in the Indian Ocean. This was set up as part of the ocean state forecasting system to force the operational ocean models at the Indian National Centre for Ocean Information Services (INCOIS). Evaluation of winds is carried out by comparing the ARW forecasts with ocean buoys and Scatterometer observations during 2016. The analysis is conducted separately with coastal and open ocean buoys, revealing marginally better performance in the open ocean simulations. The comparison of ARW forecasted winds against offshore buoy winds show mean differences for wind speed and direction of −0.1 m/s and −3.3°, with RMSEs of 2.1m/s and 39.0° and correlation of 0.75 and 0.42, respectively. At coastal regimes, the mean differences for wind speed and direction are 0.6 m/s and 2.6°, with RMSEs of 2.2 m/s and 57° and correlations of 0.63 and 0.60, respectively. ARW model performs reasonably well compared to the NCMRWF model in capturing wind speed variability in the Arabian Sea compared to the Bay of Bengal. Particularly during the VSCS Vardah, the ARW model provided more accurate forecasts (high skill score 80-90%) compared to the NCMRWF model. These results indicate the efficacy of the WRF-based forecast system in predicting surface wind fields in the Indian Ocean region, particularly in the coastal areas, thus endorsing its use for an effective operational ocean forecasting at INCOIS.
Abstract
The paper presents the development of a high-resolution mesoscale atmospheric numerical model Advanced Research WRF (ARW 3.7, henceforth ARW)-based operational forecast system, and evaluation of its surface wind forecasts using buoys and satellite data in the Indian Ocean. This was set up as part of the ocean state forecasting system to force the operational ocean models at the Indian National Centre for Ocean Information Services (INCOIS). Evaluation of winds is carried out by comparing the ARW forecasts with ocean buoys and Scatterometer observations during 2016. The analysis is conducted separately with coastal and open ocean buoys, revealing marginally better performance in the open ocean simulations. The comparison of ARW forecasted winds against offshore buoy winds show mean differences for wind speed and direction of −0.1 m/s and −3.3°, with RMSEs of 2.1m/s and 39.0° and correlation of 0.75 and 0.42, respectively. At coastal regimes, the mean differences for wind speed and direction are 0.6 m/s and 2.6°, with RMSEs of 2.2 m/s and 57° and correlations of 0.63 and 0.60, respectively. ARW model performs reasonably well compared to the NCMRWF model in capturing wind speed variability in the Arabian Sea compared to the Bay of Bengal. Particularly during the VSCS Vardah, the ARW model provided more accurate forecasts (high skill score 80-90%) compared to the NCMRWF model. These results indicate the efficacy of the WRF-based forecast system in predicting surface wind fields in the Indian Ocean region, particularly in the coastal areas, thus endorsing its use for an effective operational ocean forecasting at INCOIS.