Lightning Forecast Research Articles

A Technique for Verification of Convection-Permitting NWP Model Deterministic Forecasts of Lightning Activity

Abstract This manuscript introduces a new technique for evaluating lightning forecasts from convection-permitting models. In recent years, numerical weather prediction models at the convection-permitting scales (horizontal grid resolutions of 1–5 km) have been able to produce realistic-looking forecasts of lightning activity when compared with observations. However, it is challenging to assess what value these forecasts add above standard large-scale indices. Examining this problem, it is found that existing skill scores and neighborhood verification methods are unable to cope with both the double-penalty effect and the model’s variable frequency bias. A displacement distance and a quasi-symmetric distance score are introduced based on the distance between the model and the observations, the latter showing any improvement the forecast has over a completely “hedged” forecast. This can be combined with a domain-improved contingency table and comparisons between modeled and observed lightning flashes to evaluate the forecast performance in three important dimensions: coverage, distance, and intensity. The verification metric is illustrated with a single case, which shows that the convective-scale U.K. variable resolution model (UKV) delivers improved forecasts compared with the large-scale indices in both coverage and distance. Additionally, a month-long analysis is performed, which reveals that the coverage of lightning is in good agreement with the observations; lightning is displaced by the model by a distance on the order of 50–75 km, but the model overpredicts the lightning intensity by at least a factor of 6 after observational detection efficiencies have been considered.

Performance evaluation of an operational lightning forecasting system in Europe

This study presents the evaluation of 1 year of operational lightning forecasts provided for Europe, using the Weather Research and Forecasting model coupled with a cloud-top height-based lightning parameterization scheme. Three different convective parameterization schemes were employed for parameterizing sub-grid cloud-top heights and consequently driving the lightning scheme. Triggering of the lightning scheme was controlled by means of a model-resolved microphysics-based masking filter, while the formulation for deriving lightning flash rates was also modified, assuming a single “marine” equation instead of the original equations discriminating between continental and marine lightning. Gridded lightning observations were used for evaluating model performance on a dichotomous decision basis. Analysis showed that the lightning scheme is sensitive to the parameterization of convection. In particular, the Kain–Fritsch convective scheme was found to outperform the Grell–Devenyi and Grell–Freitas schemes, showing a statistically significant better performance with respect to lightning prediction. This was most evident during the warm season, while smaller differences among the schemes were recorded during the cold season. Further, for all examined convective schemes, it was found that the application of the masking filter is desirable for improving model performance in terms of lightning forecasting. Last, the reported results revealed that the refinement of the formulation of the lightning parameterization scheme, adhering to a “global” marine equation instead of distinguishing between land and sea lightning, may be necessary in order to obtain reliable lightning forecasts.

Seasonal prediction of lightning activity in North Western Venezuela: Large-scale versus local drivers

The Lake Maracaibo Basin in North Western Venezuela has the highest annual lightning rate of any place in the world (~200flkm−2yr−1), whose electrical discharges occasionally impact human and animal lives (e.g., cattle) and frequently affect economic activities like oil and natural gas exploitation. Lightning activity is so common in this region that it has a proper name: Catatumbo Lightning (plural). Although short-term lightning forecasts are now common in different parts of the world, to the best of the authors' knowledge, seasonal prediction of lightning activity is still non-existent. This research discusses the relative role of both large-scale and local climate drivers as modulators of lightning activity in the region, and presents a formal predictability study at seasonal scale.Analysis of the Catatumbo Lightning Regional Mode, defined in terms of the second Empirical Orthogonal Function of monthly Lightning Imaging Sensor (LIS-TRMM) and Optical Transient Detector (OTD) satellite data for North Western South America, permits the identification of potential predictors at seasonal scale via a Canonical Correlation Analysis. Lightning activity in North Western Venezuela responds to well defined sea-surface temperature patterns (e.g., El Niño-Southern Oscillation, Atlantic Meridional Mode) and changes in the low-level meridional wind field that are associated with the Inter-Tropical Convergence Zone migrations, the Caribbean Low Level Jet and tropical cyclone activity, but it is also linked to local drivers like convection triggered by the topographic configuration and the effect of the Maracaibo Basin Nocturnal Low Level Jet. The analysis indicates that at seasonal scale the relative contribution of the large-scale drivers is more important than the local (basin-wide) ones, due to the synoptic control imposed by the former. Furthermore, meridional CAPE transport at 925mb is identified as the best potential predictor for lightning activity in the Lake Maracaibo Basin.It is found that the predictive skill is slightly higher for the minimum lightning season (Jan–Feb) than for the maximum one (Sep–Oct), but that in general the skill is high enough to be useful for decision-making processes related to human safety, oil and natural gas exploitation, energy and food security.

A 24-Hour Lightning Forecast System in Brazil

Identifying in advance and with reliability favorable atmospheric conditions for lightning occurrence is necessary to minimize injuries and hazardous situations. Different approaches based on empirical and/or statistical methods have been used to forecast lightning occurrence (Shafer and Fuelberg 2008); however, the results in general have serious limitations. In turn, the use of meteorological models with the goal of making forecasts of lightning flash is a recent topic of research that has gained increasing interest in the scientific community. With the advance of the computational capability, the current mesoscale numerical models can run with high-space and time resolutions and explicitly represent individual convective storms and their microphysical properties, although some simplifications are necessary due to the complex electrical processes associated with thunderstorm electrification and lightning initiation (Barthe and Pinty 2007).

An Evaluation of the Efficacy of Using Observed Lightning to Improve Convective Lightning Forecasts

Abstract Dynamic lightning forecasts [total and cloud to ground (CG)] were produced on a convection-allowing forecast grid with 4-km grid spacing with lightning assimilation (ASML) and without lightning assimilation [control (CNTL)]. A dynamic scheme produces time- and space-dependent potential electrical energy, which then converts this energy into lightning (e.g., number per hour per grid element). The assimilation scheme uses observed, gridded total lightning to determine how much water vapor is added at constant temperature in the mixed-phase region, leading to a convective response. ASML and CNTL lightning forecasts were compared to observed total and CG lightning. Four case studies—each representing a different type of convective regime—demonstrate that the spatial distribution and intensity of forecast lightning were improved when lightning assimilation was used. Over 3 days in March 2012, eight 18-h lightning forecasts quantified the advantages in forecast accuracy. Equitable threat scores for forecast CG lightning associated with strong [25 (3 h)−1], very strong [50 (3 h)−1], and extreme [100 (3 h)−1] events were significantly more accurate for convective storms that developed in forecasts with lightning assimilation than without. Improvements in forecasts of very strong and extreme events occurred out to 9 h of forecast time, while the forecasts of strong events showed improvement out to 15 h. Spurious convection was removed with filtering in one case study, which led to a notable improvement in the timing and intensity of the squall line. Sensitivity tests examined the utility of this filtering approach, and the importance of reducing mass imbalances liable to occur when too much water vapor mass is introduced into the model.

A lightning forecast for the London 2012 Olympics opening ceremony

WeatherVolume 69, Issue 1 p. 16-19 Research Article A lightning forecast for the London 2012 Olympics opening ceremony Jonathan M. Wilkinson, Corresponding Author Jonathan M. Wilkinson Met Office, ExeterCorrespondence to: Jonathan M. Wilkinson jonathan.wilkinson@metoffice.gov.ukSearch for more papers by this authorF. Jorge Bornemann, F. Jorge Bornemann Met Office, ExeterSearch for more papers by this author Jonathan M. Wilkinson, Corresponding Author Jonathan M. Wilkinson Met Office, ExeterCorrespondence to: Jonathan M. Wilkinson jonathan.wilkinson@metoffice.gov.ukSearch for more papers by this authorF. Jorge Bornemann, F. Jorge Bornemann Met Office, ExeterSearch for more papers by this author First published: 03 January 2014 https://doi.org/10.1002/wea.2176Citations: 13Read the full textAboutPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinked InRedditWechat Citing Literature Volume69, Issue1January 2014Pages 16-19 RelatedInformation

Application Study of Machine Learning in Lightning Forecasting

Application Study of Machine Learning in Lightning Forecasting

Predicting Cloud-to-Ground and Intracloud Lightning in Weather Forecast Models

Abstract A new prognostic, spatially and temporally dependent variable is introduced to the Weather Research and Forecasting Model (WRF). This variable is called the potential electrical energy (Ep). It was used to predict the dynamic contribution of the grid-scale-resolved microphysical and vertical velocity fields to the production of cloud-to-ground and intracloud lightning in convection-allowing forecasts. The source of Ep is assumed to be the noninductive charge separation process involving collisions of graupel and ice particles in the presence of supercooled liquid water. The Ep dissipates when it exceeds preassigned threshold values and lightning is generated. An analysis of four case studies is presented and analyzed. On the 4-km simulation grid, a single cloud-to-ground lightning event was forecast with about equal values of probability of detection (POD) and false alarm ratio (FAR). However, when lighting was integrated onto 12-km and then 36-km grid overlays, there was a large improvement in the forecast skill, and as many as 10 cloud-to-ground lighting events were well forecast on the 36-km grid. The impact of initial conditions on forecast accuracy is briefly discussed, including an evaluation of the scheme in wintertime, when lightning activity is weaker. The dynamic algorithm forecasts are also contrasted with statistical lightning forecasts and differences are noted. The scheme is being used operationally with the Rapid Refresh (13 km) data; the skill scores in these operational runs were very good in clearly defined convective situations.

Characterization of Lightning Occurrence in Alaska Using Various Weather Indices for Lightning Forecasting

Alaska lost 10% of its boreal forest area due to vigorous forest fires in 2004 and 2005. Repeated lightningcaused forest fires adversely impact residents and influence earth’s atmosphere in every fire season. The authors have reported on the weather conditions of Alaska’s most severe lightning occurrence in mid June 2005. This paper examines a range of weather indices like soar, instability, ‘dry lightning’ and other factors that could clearly explain lightning characteristics in Alaska. First, lightning occurrence days from May to September were classified into ‘non or small lightning’ days and ‘lightning’ days to determine threshold values. Second, ‘lightning’ days were categorized into ‘less severe,’ ‘severe,’ ‘very severe,’ and ‘extremely severe’ to notice controlling factors on the lightning severity. Based on this analysis, the lifted index (LIFT) was selected as sensitive to assess upper air instability, and Te850 (environmental temperature at 850 hPa) was selected as sensitive to assess warm and moist air masses. Finally, the possibilities of lightning forecasts in Alaska are discussed using lightning occurrence and LIFT and Te850 in 2005. As there is a time-lag between LIFT measurements (14:00) and the lightning peak (-17:00), and around one day time-lag between Te850 and lightning occurrence, lightning forecasts using LIFT and Te850 could provide a simply applicable forecast index for Alaska.

Radar Nowcasting of Cloud-to-Ground Lightning over Houston, Texas

Abstract Ten years (1997–2006) of summer (June–August) daytime (1400–0000 UTC) Weather Surveillance Radar-1988 Doppler data for Houston, Texas, were examined to determine the best radar-derived predictors of the first cloud-to-ground lightning flash from a convective cell. Convective cells were tracked using a modified version of the Storm Cell Identification and Tracking (SCIT) algorithm and then correlated to cloud-to-ground lightning data from the National Lightning Detection Network (NLDN). Combinations of three radar reflectivity values (30, 35, and 40 dBZ) at four isothermal levels (−10°, −15°, −20°, and updraft −10°C) and a new radar-derived product, vertically integrated ice (VII), were used to optimize a radar-based lightning forecast algorithm. Forecasts were also delineated by range and the number of times a cell was identified and tracked by the modified SCIT algorithm. This study objectively analyzed 67 384 unique cells and 1 028 510 lightning flashes to find the best lightning forecast criteria. Results show that using 30 dBZ at the −15° or −20°C isotherm on cells within 75 km of the radar that have been tracked for at least two consecutive scans produces the best lightning forecasts with a critical success index (CSI) of 0.68. The best VII predictor values were 0.42 or 0.58 kg m−2 on cells within 75 km of the radar that have been tracked for at least two consecutive scans, producing a CSI of 0.67. Lead times for these predictors were 10.0 and 13.4 min, respectively. Lead times greater than 10 min occurred with less stringent predictors (e.g., 30 dBZ at −10°C or VII greater than 0.25 kg m−2 on cells within 125 km with a minimum track count of 2), but lower CSI values result. In general, cells tracked for multiple scans provide higher CSIs and lead times than decreasing the range from the radar or changing the reflectivity threshold and height.

Using the 21 June 2008 California Lightning Outbreak to Improve Dry Lightning Forecast Procedures

Abstract An anomalous lightning event occurred on 20–21 June 2008 in central and northern California. Most of the thunderstorms produced frequent lightning with little precipitation, resulting in over 1500 new fires. Many of these fires became large, and massive firefighting efforts continued through August in order to contain all the fires. This event was not well forecast, with only isolated lightning strikes expected. Several computer model forecast parameters pointed toward the impending lightning outbreak, including satellite and lightning observations, as well as model data from 19 June. An existing procedure to forecast dry thunderstorms showed the strong possibility of a thunderstorm outbreak over California. This procedure proved to be helpful, but has been modified to better handle moisture advection and low-level forcing, which provides a more comprehensive framework for forecasting potential dry thunderstorm outbreaks.

Evaluation of European Storm Forecast Experiment (ESTOFEX) forecasts

Three years of forecasts of lightning and severe thunderstorms from the European Storm Forecast Experiment (ESTOFEX) have been evaluated. The forecasts exhibit higher quality in summer than in winter and there is some evidence that they have improved over the course of the evaluation. Five individual forecasters made the majority of the forecasts and differences in their forecasts are on the order of the overall variability of the forecast quality. As a result, the forecasts appear to come from a single unit, rather than from a group of individuals. The graphical description of the probability of detection and frequency of hits recently developed by Roebber is a valuable tool for displaying the time series of lightning forecast performance. It also appears that, even though they are not intended for that purpose, using the lightning forecasts as a low-end forecast of severe thunderstorms is potentially useful for decision makers.

A comparison of lightning activity and convective indices over some monsoon-prone areas of China

Using 10-year lightning localization data observed by the TRMM (Tropical Rainfall Measuring Mission) Lightning Imaging Sensor (LIS), the relationship between lightning activity and a series of convective indices was investigated over nine monsoon-prone areas of China in which high-impact weather (HIW) events are frequently observed. Two methods were used to verify and reconstruct LIS lightning data. First, LIS lightning flash data were verified by both surface thunderstorm reports and ground-based lightning detection data. Seasonal, monthly, and 5-day distributions of LIS observed lightning activity agree well with the surface reports and ground-based lightning observations. Second, due to LIS's low sampling frequency, a data reconstruction and compensation scheme for LIS lightning observations was designed using both LIS lightning seasonal diurnal cycles and surface thunderstorm reports. After data reconstruction, five lightning products were derived: daily mean and maximum LIS flash rate, daily mean and maximum LIS lightning cell rate, and number of lightning days per five day period. Then, a series of convective indices describing convection conditions were derived from radiosonde data according to atmospheric instability and convective potential analysis. Correlation analysis for each study region was done between 10-year lightning derived products and corresponding convective indices by 5-day periods. The correlation analysis results show that higher lightning flash rate and lightning probability are associated with more unstable air and smaller vertical wind shear in a nearly saturated lower layer in most of the study regions. But the correlation varies from region to region. The best correlation between lightning activity and convective indices was found in eastern and southern China, whereas the correlation is lowest in some inland or basin topography regions in which topographic effects are more significant. Moreover, ambient moisture plays a much more important role in the convective development of thunderstorms in southern China than other regions. Thunderstorm development mechanism differences among regions were also discussed. Based on the close relationship between lightning activity and convective indices, some regression equations for forecasting 5-day mean or maximum LIS lightning flash rate and lightning area (a thunderstorm cell) rate, and 5-day lightning days for the study regions were developed using convective indices as predictors. The verifications show that the convective index-based lightning forecast methods can provide a reasonable lightning outlook including probability and lightning flash rate forecasts for a 5-day period.

Global Positioning System (GPS) Precipitable Water in Forecasting Lightning at Spaceport Canaveral

Abstract This paper evaluates the use of precipitable water (PW) from the global positioning system (GPS) in lightning prediction. Additional independent verification of an earlier model is performed. This earlier model used binary logistic regression with the following four predictor variables optimally selected from a candidate list of 23 candidate predictors: the current precipitable water value for a given time of the day, the change in GPS PW over the past 9 h, the K index, and the electric field mill value. The K index was used as a measure of atmospheric stability, which, of the traditional stability measures, has been shown to work best in the area and season under study. This earlier model was not optimized for any specific forecast interval, but showed promise for 6- and 1.5-h forecasts. Two new models were developed and verified. These new models were optimized for two operationally significant forecast intervals. The first model was optimized for the 0.5-h lightning advisories issued by the U.S. Air Force’s 45th Weather Squadron. An additional 1.5 h was allowed for sensor dwell, communication, calculation, analysis, and advisory decision by the forecaster. Therefore, the 0.5-h advisory model became a 2-h forecast model for lightning within the 45th Weather Squadron advisory areas. The second model was optimized for major ground processing operations supported by the 45th Weather Squadron, which can require lightning forecasts with a lead time of up to 7.5 h. Using the same 1.5-h lag as in the other new model, this became a 9-h forecast model for lightning within 37 km (20 n mi) of the 45th Weather Squadron advisory areas. The two new models were built using binary logistic regression and a list of 26 candidate predictor variables: the current GPS PW value, the K index, and 24 candidate variables of the change in GPS PW levels over 0.5-h increments up to 12 h. The new 2-h model found the following four predictors to be statistically significant, listed in decreasing order of contribution to the forecast: the 0.5-h change in GPS PW, the 7.5-h change in GPS PW, the current GPS PW value, and the K index. The new 9-h forecast model found the following five independent variables to be statistically significant, listed in decreasing order of contribution to the forecast: the current GPS PW value, the 8.5-h change in GPS PW, the 3.5-h change in GPS PW, the 12-h change in GPS PW, and the K index. In both models, the GPS PW parameters had better correlation to the lightning forecast than did the K index, a widely used thunderstorm index. Possible future improvements to this study are discussed.

Lightning strike forecasts

A new Global Positioning System Lightning Index could provide a better lightning forecast for the launches of the NASA Shuttle and other commercial and military launches from the Kennedy Space Center and Cape Canaveral Air Force Station in Florida, which are in a high lightning‐flashdensity region.The index measures the amount of water vapor in the atmosphere, which can help to create the positive and negative charges in a thundercloud.

Verification of Mesoscale Forecasts Made during MAP'88 and MAP'89

Two experimental forecasting projects, each called Mesoscale Applications Project (MAP), were conducted jointly by the National Severe Storms Laboratory and the National Weather Service Forecast Office at Norman, Oklahoma, during 1988 and 1989. This paper focuses primarily on the verification of the MAP'88 and MAP'89 experimental forecasts, and combines the results with those from a similar experiment run in 1987, to examine the evolution of forecast skill over that three-year period. Results suggest that the severe-weather outlooks issued on a given experiment day exhibited good skill, with the skill being fairly stable over the three-year period studied. For outlooks issued the day before, the skill was notably higher in 1987 than in the subsequent years. Convective-mode forecasts ranged from poor to moderate skill levels, and did not change significantly from results obtained in 1987. Areal lightning forecasts were attempted in 1988 and 1989, with skill increasing more or less as the valid area increased, that area being defined as a circle ranging from 10 to 40 km of Norman. Advance outlooks for lightning, issued the day before the anticipated event, showed little or no skill. Some discussion of the possible reasons for the observed forecasting skill and its trends is presented. Several aspects of forecasting experiments in general are discussed also, based on experience during the MAP experiments.

On lightning forecasting

An exact forecasting of lightning is not possible so far. On the other hand situations occur to-day more often than in earlier times in which the best lightning protection is not sufficient to prevent harmful effects of lightning strokes. For this reason it should be attempted to use informations from the field of atmospheric electricity to improve the existing possibilities of lightning forecasting. These improvements are shortly pointed out, further experiments are under way.