## 1. Introduction

Cloud-to-ground (CG) lightning is one of the leading causes of weather-related fatalities in the United States (Holle et al. 1999). In fact, Curran et al. (2000) showed that only river and flash floods ranked higher than lightning in terms of deaths. Aside from the loss of life, CG lightning damages trees, buildings, and utility lines, often leading to power outages and disruptions to communications.

Improved forecasts of CG lightning would have many potential societal benefits. Skillful probabilistic guidance in the 3–12-h time frame would allow the public to better assess the CG lightning threat and thereby support better decision making regarding the protection of life and property. Some of the economic sectors that would benefit include organizers of outdoor sporting events, the fire weather community, aviation, the maritime industry, outdoor construction, and electric utilities. A product that produces accurate and timely CG lightning threat information should lead to reduced fatalities and injuries.

Florida has more than 2 times the number of lightning casualties of any other state (Curran et al. 2000). A majority of casualties occurs during the warm season (May–September), the climatological peak for lightning in Florida. Studies examining lightning patterns across the United States have shown that Florida annually receives more CG strikes than any other state (Orville 1994; Hodanish et al. 1997; Orville and Huffines 2001; Orville et al. 2002).

It has long been recognized that warm-season convection over Florida is regulated by low-level convergence associated with the sea breeze. Complete descriptions of the sea breeze are given by Wexler (1946) and Simpson (1994). Interactions between the sea breeze, the prevailing wind, and coastline curvature have been shown to influence lightning patterns (e.g., López and Holle 1987; Arritt 1993; Lericos et al. 2002). Numerical simulations also have revealed the effects of the synoptic flow on sea-breeze convergence (e.g., Estoque 1962; Pielke 1974; Arritt 1993). Studies by McPherson (1970), Hodanish et al. (1997), Camp et al. (1998), and Lericos et al. (2002) noted the effects of coastline shape on lightning patterns. In addition, local thermal circulations (e.g., due to water conservation areas, lakes, and rivers), urban effects (e.g., Westcott 1995; Steiger et al. 2002), and thunderstorm outflows can aid in producing deep convection. Even if one could pinpoint the exact locations that will experience convection, it does not necessarily follow that these areas will experience the most lightning, since lightning production ultimately is controlled by cloud microphysics.

Determining how convective clouds become electrified remains a challenging problem in meteorology, and many factors influencing cloud electrification still are poorly understood (MacGorman and Rust 1998). Two traditional explanations for electrification are the precipitation hypothesis (Reynolds et al. 1957) and the convection hypothesis (Vonnegut 1963). More recent studies propose that cloud electrification occurs through the noninductive ice–ice collision process (e.g., Williams 1985; Williams et al. 1989). These hypotheses depend on a vigorous updraft and a robust ice phase for charge generation (Price and Rind 1992; Petersen and Rutledge 1998). Studies suggest that a persistent and strong updraft above the −10°C level, and not just the presence of a strong updraft, is necessary for electrification (e.g., Price and Rind 1992, 1993; Solomon and Baker 1994, 1998; Zipser 1994; Gremillion and Orville 1999; Ushio et al. 2001).

A variety of statistical techniques have been used to develop forecast models for thunderstorms and lightning. For continuous predictands, the most common method is multiple linear regression (MLR) (e.g., Neumann and Nicholson 1972; Reap and MacGorman 1989). However, when the predictand is “yes” or “no,” binary logistic regression (BLR) often is employed (this technique is described in section 3). Examples include Livingston et al. (1996), Mazany et al. (2002), Lambert et al. (2005), and Shafer and Fuelberg (2006). Statistical prediction models for lightning over Canada and the northern United States also have been developed using Classification and Regression Trees (CART) (Burrows et al. 2005).

Many of the statistical studies described above utilized parameters derived from morning soundings to forecast lightning occurrence during the afternoon. However, this approach sometimes can produce large forecast errors if morning conditions change, or if the sounding is not representative of the entire forecast area. An alternative to soundings is data from numerical weather prediction (NWP) models. Since NWP models provide input data that are more location and time specific than soundings, they may produce more skillful forecasts.

Model output statistics (MOS) is an objective forecasting technique in which a statistical relationship is determined between a predictand and variables forecast by an NWP model (Glahn and Lowry 1972). The primary advantage of MOS is that model biases and local climatology are automatically built into the equations (Klein and Glahn 1974; Brunet et al. 1988). Reap (1994a) developed MOS equations predicting the spatial distribution of CG lightning over Florida during different low-level flow regimes using predictors from the Nested Grid Model (NGM). The MOS approach also has been used to generate objective thunderstorm guidance for stations across the contiguous United States based on output from NWP models (e.g., Reap and Foster 1979; Reap 1994b; Hughes 2001, 2002).

Despite its advantages, MOS has several drawbacks that can limit its forecast skill. Since NWP models are constantly changing, it often is difficult to obtain a long archive of forecasts from the same model that will be used to develop the MOS equations. Any modifications to the NWP model that change (even reduce) systematic model errors require redevelopment of the MOS equations (Wilks 2006). Furthermore, a MOS scheme tends to perform best for typical events while rare events generally are forecast poorly.

An alternative to MOS is the perfect prognosis (PP) (or “perfect prog”) method. This approach develops statistical relationships between *observed* atmospheric parameters and observations of the predictand (Klein et al. 1959; Klein 1971). Once the statistical relations are determined, forecasts of the predictand are obtained by inserting NWP model forecasts of the predictors into the PP equation (Wilks 2006). Bothwell (2002) used the PP method to develop lightning guidance for the western United States on a 40 km × 40 km grid, using analyses from the NCEP 40-km Rapid Update Cycle (RUC40).

A drawback to the PP scheme is that it assumes a “perfect” forecast of the predictors by the NWP model and thus does not account for model biases. Conversely, a significant advantage is the stability of the equations. Since PP equations are developed without NWP information, any changes to the driving NWP models do not require redevelopment of the PP equations. In fact, improving random or systematic errors in the NWP model should improve the statistical forecasts (Wilks 2006). This advantage makes PP the method of choice for this study.

The main objective of this study is to use the PP method to develop a high-resolution, gridded forecast guidance product for warm-season CG lightning over Florida on a 10 km × 10 km grid at 3-hourly intervals. An archive of analyses from the 20-km RUC model (RUC20) is used to examine relationships between observed atmospheric parameters and spatial and temporal patterns of CG lightning. The most important RUC-derived parameters then are used to develop equations producing 3-hourly forecasts for the probability of one or more CG flashes (PROB ≥ 1), as well as the probability of exceeding various flash count percentile thresholds (PROB ≥ *T*). The equations then are applied to output from three mesoscale models during an independent test period (the 2006 warm season).

Section 2 describes the study domain as well as the lightning and RUC analysis datasets. The model development procedure is described in section 3. A discussion of the parameters composing the models and their physical significance to lightning occurrence is given in section 4. Section 5 evaluates the utility of the lightning forecast scheme when applied to output from several mesoscale models during the 2006 warm season. A summary of the model development and verification results is given in section 6.

## 2. Data

Our lightning guidance was developed for all of Florida (Fig. 1), covering the peninsula and panhandle and extending northward into southeast Alabama and southern Georgia. Figure 1 also shows the 10 km × 10 km grid points used for computing the lightning probabilities. Since CG lightning over heavily populated areas is of most interest, only data for grid points over land were included in the model development. However, the procedure described in this paper also could be used to develop equations for offshore areas.

We utilized CG lightning data from the National Lightning Detection Network (NLDN) (Cummins et al. 1998). This network, in operation since 1989, detects and records CG flashes across the contiguous United States. The NLDN is owned and operated by Vaisala-Global Atmospherics, Inc. (GAI). The network consists of 113 sensors that utilize the Improved Accuracy from Combined Technology (IMPACT) method to detect CG flashes. A more detailed description of sensors and methods of detection is given in Cummins et al. (1998).

Lightning data for the 1995–2005 warm seasons was used to develop climatological predictors. Data for the 2002–05 warm seasons was used to develop the PP equations, while data for the 2006 warm season was used for testing (section 5). A systemwide NLDN upgrade completed in 1995 allowed a greater number of flashes to be detected and provided improved location accuracy. Through 2001, the NLDN had a location accuracy of ∼500 m over most of the United States, and a flash detection efficiency of 80%–90% (Cummins et al. 1998). Beginning in spring 2002, the network underwent its most recent systemwide upgrade with the installation of new IMPACT sensors. CG detection efficiencies now are estimated to be 90%–95% throughout the continental United States, with location accuracy <500 m (Cummins et al. 2006). We did not apply corrections to account for the variations in detection efficiency and location accuracy during the 1995–2006 period. Thus, actual CG flash counts are underestimated. We also employed a quality control procedure recommended by Cummins et al. (1998) to remove possible cloud discharges and duplicate strikes. This procedure is described in Shafer and Fuelberg (2006).

For every land grid point (Fig. 1), flash totals were tabulated for each 3-h period (e.g., 0000–0259, . . . , 2100–2359 UTC) by summing the strikes that occurred within a 10-km radius. The flash totals then were transformed into binary variables; “1” if one or more flashes occurred during the 3-h period or “0” if no lightning occurred. Binary variables also were assigned based on whether the flash total exceeded the 50th, 75th, 90th, or 95th percentiles for a given 3-h period (these percentiles are given in Table 1 for the four most active 3-h periods). The 3-h flash totals and binary indicators served as the predictands for developing the lightning forecast equations.

Our fixed-radius counting method has advantages and disadvantages compared to the more traditional gridbox approach. Figure 2 depicts a superposition of two counting regions with equal areas, one a box and the other a circle, centered on point P. Also shown are two lightning strikes at locations A and B. We assume that a physical process acting at P has some influence on the occurrence of a lightning strike within a certain distance of P. It is clear that a gridbox counting procedure will ignore strike B while strike A will be counted, irrespective of the fact that strike B is closer to P than strike A. Conversely, with a fixed-radius approach, all strikes occurring within a specified radius of P are counted. However, a consequence of this approach is that the counting regions must overlap to assure complete coverage of the study area, and this can lead to a flash being assigned to more than one grid point if it occurs in the overlap between two or more circular regions. This can be advantageous since the degree of data smoothing can be controlled by changing the radius. However, the lightning counts for adjacent grid points are not independent, which is a key assumption in many statistical tests. This issue is addressed in the model development procedure described in section 3.

Observed atmospheric predictors were obtained from an archive of RUC analyses during the 2002–05 warm seasons (∼600 days). The RUC is a mesoscale high-frequency data assimilation and short-range numerical prediction system (Benjamin et al. 2004). A 20-km, 50-level, hourly version of the RUC was implemented during April 2002, replacing the 40-km version (RUC40) implemented in 1998 and containing improvements in the analysis and model physics. Another RUC upgrade was implemented at NCEP on 28 June 2005 with increased resolution [13 km (RUC13)] and additional improvements. The atmospheric parameters calculated from the RUC analyses are described in section 3.

Two statistical software packages were used for the exploratory analysis and model development. They are S-PLUS, version 6.1 for Windows, and the Statistical Package for the Social Sciences (SPSS), version 11.5, for Windows. Both are powerful, state-of-the-art packages with a wide range of analysis and modeling capabilities.

## 3. Model development

### a. Climatological and map-type lightning predictors

Climatological and pattern-type lightning frequencies were developed and used as candidate predictors to capture local enhancements due to interactions between the low-level wind, thermal circulations, and coastline topography (e.g., Pielke 1974; Arritt 1993; Laird et al. 1995; Lericos et al. 2002). These predictors have the potential to add detailed information about local effects that may not be well resolved by NWP models (Reap 1994a).

We used a simple correlation technique described in Lund (1963) and Reap (1994a) to develop the map-type predictors. Based on Reap (1994a), the correlation technique was applied to 3-hourly observed sea level pressure (SLP) fields from RUC analyses spanning the 1998–2005 warm seasons (∼1224 days). SLP patterns imply both the direction and speed of the low-level flow. Although this procedure also was applied to the 950-hPa height fields, SLP produced the most distinct pattern types and the greatest number of maps that could be classified into a type.

The pattern classification was performed over the geographical area shown in Fig. 1. To capture only the regional-scale patterns (i.e., the prevailing wind) and to smooth small-scale variations, the RUC SLP values were interpolated to a coarser grid (100 km) (Fig. 1). Each smoothed SLP map then was correlated with every other map in the sample (9613 available maps). The pressure pattern with the most maps correlated with it at a threshold of 0.70 or greater (e.g., Reap 1994a) was denoted type A and removed from the sample along with all other type A maps. The procedure was repeated to determine subsequent map types until the residual sample contained less than 3% of the maps that were correlated at 0.70 or greater (Lund 1963; Reap 1994a).

Table 2 shows results of the map-type classification. Five map types (A–E) were developed using the 0.70 correlation threshold. Two types (A and B) make up ∼44% of the total sample, while types C–E make up ∼34%. The remaining ∼22% of the sample could not be classified at the 0.70 threshold. This threshold usually is considered the smallest acceptable for pattern classification (Lund 1963; Reap 1994a). Increasing the threshold produces more map types and more detailed map patterns, but at the expense of producing more unclassified cases.

Using the binary indicators for one or more flashes (section 2), relative lightning frequencies (MTFREQ) were calculated for each map type and 3-h period. Similarly, 3-h flash totals were used to calculate an unconditional mean number of flashes for each map type (MTMEAN). Only time periods when the SLP map could be classified at 0.70 or greater were used in preparing the map-type predictors. However, when developing the lightning guidance equations, the unclassified maps were assigned to the type with which they were most correlated. Climatological relative frequencies and unconditional means also were calculated using all warm-season days during 1995–2005. The climatological and map-type lightning frequencies and means were submitted as candidate predictors for the regression analysis described later.

Composite SLP patterns associated with each map type are shown in the left panels of Fig. 3, while the right panels show spatial distributions of the mean number of flashes (MTMEAN) for the 1800–2059 UTC period. The five map types represent distinctly different flow patterns, and are similar to those from previous studies (e.g., Reap 1994a; Lericos et al. 2002). The predominant pattern, type A (Figs. 3a,b), is characterized by high pressure northeast of Florida that produces prevailing easterly and southeasterly flow across the state. As a result, most of the lightning is confined to the west coast, with maxima near Tampa Bay, Fort Myers, and east of Lake Okeechobee. Map type B (Figs. 3c,d) contains a surface ridge over south Florida that results in southwesterly flow across the state. This focuses the lightning along the east coast of the peninsula, with coastline interactions evident near the big bend of the panhandle (Camp et al. 1998). Map type C (Figs. 3e,f) represents a transition between types A and B, in which the east–west surface ridge is located over central Florida. This pattern produces southeasterly flow over south Florida and south-southwesterly flow over the northern peninsula. Thus, the lightning patterns are a combination of types A and B, with maxima along both coasts. Map type D (Figs. 3g,h) is characterized by high pressure north of Florida and lower pressure to the southeast, which is most common during May and September after a cold frontal passage. The dry northeasterly flow confines most of the lightning to south Florida. Finally, map type E (Figs. 3i,j) is a variation of type B, exhibiting a lobe of high pressure over the Gulf of Mexico and lower pressure to the northeast. West-northwesterly flow confines most lightning to the east coast and Big Bend, with generally less coverage than observed with type B.

### b. Model-analyzed candidate predictors

A large number of RUC-analyzed predictors were investigated for possible inclusion in the candidate predictor pool, many of which have been found useful in previous studies. The parameters investigated, their abbreviations, and a short description of each are listed in Table 3. The parameters were calculated from the RUC-analyzed temperature, dewpoint, wind, height, and surface pressure fields valid every 3 h (0000, 0300 UTC, . . . , etc.). The fields were interpolated to the 10 km × 10 km grid (Fig. 1) and transformed into the format of a vertical sounding at each grid point (Bothwell 2002). RUC cloud hydrometeor profiles also were investigated; however, a documented error in the RUC cloud analysis procedure through June 2006 (Forecast Systems Laboratory 2006) rendered these fields unusable.

An important assumption is that the model analyses provide the *best estimate* of the state of the atmosphere at the analysis time, and thus can be treated as “observations” for purposes of developing the PP equations. Although it is important to investigate as many relevant predictors as possible, we focused on those that are well handled by today’s NWP models. Nonetheless, some parameters that can be difficult to forecast on small scales (e.g., MFLXC, LCAPE, CCTHGT) were investigated, with the expectation that as the spatial resolution and physics of mesoscale models continue to improve, forecasts of these parameters also will improve.

### c. Generalized linear models

MLR has been used in the majority of previous statistical lightning studies (e.g., Neumann and Nicholson 1972; Reap and Foster 1979; Reap and MacGorman 1989; Reap 1994a; Hughes 2001). However, unless the assumptions of constant variance and Gaussian residuals are met (which is rarely the case with count data), these methods can lead to undesirable and sometimes nonsensical results. Thus, we considered alternative regression methods; namely, the family of generalized linear models (GLMs).

*p*is the predicted probability resulting from the

_{i}*i*th set of predictors (

*x*

_{1},

*x*

_{2}, . . . ,

*x*). The quantity on the left of (1) is the logit link function, which relates the log of the odds ratio (

_{K}*p*/1 −

*p*) to a linear combination of predictors. In BLR, the regression parameters (

*b*

_{0},

*b*

_{1}, . . . ,

*b*) are estimated by maximizing a log-likelihood function using iterative methods [Wilks (2006) gives a thorough description of these methods]. Unlike MLR, (2) guarantees that the probabilities are bounded within the interval (0, 1). BLR does not assume a direct linear relationship between the predictors and the response and accommodates the non-Gaussian (Bernoulli) distributions of the regression residuals (Lehmiller et al. 1997).

_{K}We used BLR to develop equations giving the probability of one or more flashes (PROB ≥ 1) within a 10-km radius of each grid point (Fig. 1) to produce spatial probability forecasts for each 3-h period. BLR has been used successfully in previous lightning forecasting studies (e.g., Bothwell 2002; Mazany et al. 2002; Lambert et al. 2005; Shafer and Fuelberg 2006). The procedure used to develop the equations is described later in section 3d.

*μ*) of the dependent variable (

*y*):

*μ*[

*x*] is the mean response resulting from the

_{i}*i*th set of predictors (

*x*

_{1},

*x*

_{2}, . . . ,

*x*). If one assumes that events occur randomly and at a constant average rate (

_{K}*μ*) with Var(

*y*) =

*μ*, then the events are said to be generated by a Poisson process with the probability model

A histogram of the conditional count distribution for our most active lightning period (1800–2059 UTC) is shown in Fig. 4. It is clear that the counts are strongly skewed, with the majority of cases having 10 or fewer flashes and few cases having 100 or more. The count distributions for the other time periods (not shown) are even more strongly skewed, with more cases in the lower count bins and fewer cases in the higher count bins. Since the variance of the distribution is very large, ∼80 times greater than the mean (*μ* ∼ 23 flashes), the data do not fit the Poisson assumption that Var(*y*) = *μ*. The most likely explanation is that the counts were generated by an inhomogeneous Poisson process (also known as a Cox process), whereby the number of storms over a given region and the number of flashes produced per storm are both approximately Poisson. This “mixed” Poisson process results in the lightning counts having much more dispersion than is accounted for by a homogeneous Poisson model (T. Jagger, The Florida State University, 2006, personal communication).

*μ*) is modeled by (4); however, the variance Var(

*y*), now is a quadratic function of

*μ:*

*θ*

^{−1}is the shape parameter (estimated by maximum likelihood). The resulting probability model for the number of flashes,

*y*, as a function of

*μ*and

*θ*is given by

Figure 5 shows the probability distributions implied by the Poisson [(5)] and NB [(7)] models with only the intercept term (*b*_{0}) included for *μ* = 23.15 and *θ* = 0.342 (estimated from the observed data using the S-PLUS software). Also shown is the observed frequency distribution. It is clear that the Poisson model is a poor choice for representing the count distribution since too little probability is assigned to the smallest lightning counts while too much is assigned to counts near the mean. The NB model is a much better fit to the data, capturing the large number of cases with 10 or fewer flashes and more closely representing the tail of the observed distribution.

*y*= 0. However, if instead we treat

*y*− 1 as having a NB distribution, then (7) can be used to estimate the probability for each

*y*− 1. Since (7) is a probability density function, the individual probabilities (

*y*= 1, ∞) must sum to 1. Thus, the probability of meeting or exceeding any count threshold,

*T*, can be obtained from

### d. Equation development

We determined whether relationships between observed predictors and lightning were generally the same for the entire study area or if they varied significantly from one portion of the state to another. We first subdivided the domain into nine areas (Fig. 6). Then, separate sets of equations were developed for each area (i.e., a regionalized operator approach), with the results compared to those obtained using a model developed for all grid points (i.e., a generalized operator approach). A comparison of verification scores revealed that it would be sufficient to consolidate the nine areas into four larger regions: East Coast, West Coast, Panhandle, and Alabama–Georgia (Fig. 6). To minimize spatial discontinuities at the boundaries, the regions were permitted to overlap, and the probabilities for grid points within the overlapping regions were averaged.

Parameters calculated from the RUC analyses (Table 3), as well as the map type and climatological frequencies for each 3-h period (section 3a), composed the initial set of candidate predictors. However, it is clear (Table 3) that many of the predictors contain redundant information. Including predictors with strong mutual correlation in a prediction equation can lead to poor estimates of the regression parameters (Wilks 2006). This problem was addressed by performing a principal component analysis (PCA) to examine intercorrelations among the predictors (Table 3) and to aid in choosing a smaller subset to retain for the regression analysis. This procedure is described in detail in Shafer and Fuelberg (2006). The principal component groupings, in conjunction with Spearman correlation coefficients between the predictors and the lightning predictands (section 2), were used as an objective method to select a subset of the most physically relevant predictors containing less mutual correlation.

The final list of candidate predictors is given in Table 4. To allow the possibility that one predictor may be more important than another in conjunction with others already in the model, several parameters with mutual correlation were retained for possible selection (KI and PRECPW, several LCAPEs, etc.). Also shown in Table 4 are Spearman rank correlation coefficients between each predictor and the binary indicator for one or more flashes during the 1800–2059 UTC period for the East Coast (EC) region. The correlations are low, meaning that no *single* observed predictor is a good indicator of lightning (Bothwell 2002; Shafer and Fuelberg 2006). Correlations for the amount of lightning (not shown) also were found to be low. To account for possible nonlinear and interaction effects, power terms up to the fourth degree and two-way cross products were calculated for each parameter selected thus far (Table 4) and then included in the final predictor pool. Finally, the 3-h change in each parameter also was calculated and submitted as candidate predictors.

As mentioned in section 2, the fixed-radius approach requires that the counting regions overlap. The main drawback of this approach is that different locations in the domain are not covered by the same number of counting regions. With a counting radius of 10 km and centers on a 10-km grid, a particular location could be covered by as few as two or as many as four regions. As a result, a lightning strike could be used as few as two or as many as four times in the data analysis. While this would not be a problem if each region were modeled separately, the assumption of independence may be violated if data for all grid points compose the developmental sample (as done in this study). As a precaution, only data for nonoverlapping (disjoint) regions were used for each screening regression sample (illustrated in Fig. 7). Each sample of nonoverlapping data contains ∼¼ of the available data points, resulting in four separate data samples for use in the regression analysis. It should be noted, however, that the variables selected for inclusion in each model and their respective coefficients generally were found to be similar even if data for all grid points composed the regression sample.

A combination of forward stepwise selection and cross validation was used to develop the BLR equations for each region (Fig. 6) and 3-h period using the SPSS software. This procedure is similar to that described in Shafer and Fuelberg (2006). Each database of nonoverlapping data first was subdivided into two samples of ∼300 days, one containing even years (2002 and 2004) and the other odd years (2003 and 2005). Data for even years were used as a “learning” sample for screening the variables for selection, while the odd years were used as an “evaluation” sample to test the model each time a variable was added or removed during the stepwise selection process. Thorough discussions of stepwise selection procedures are given in Hosmer and Lemeshow (1989) and Wilks (2006). The predictors comprising the model at the step with the highest percentage of correctly classified events for the evaluation sample were noted. Only parameters for which the sign of the coefficient made physical sense were retained in the model in any screening sample. The stepwise selection procedure was repeated for each sample of nonoverlapping data, and the predictors chosen for each sample were noted. This procedure identified the combination of predictors that is most likely to generalize to independent data and not overfit the dependent sample. The set of “best” predictors from this process then was re-entered using data for all grid points and all years to determine the final coefficients for each model.

The NB models for PROB ≥ *T* were developed using S-PLUS. We found that the overnight and early morning periods did not contain a sufficient number of events in the upper percentiles to allow stable, reliable models to be developed. Thus, NB models were developed only for the four most active periods (1500–1759, 1800–2059, 2100–2359, and 0000–0259 UTC). The same sampling procedure (i.e., even and odd years) described above was used to develop the models. However, since the S-PLUS software does not permit stepwise selection for NB regression, the predictors were entered simultaneously into the model. Those predictors contributing the greatest reduction of variance (usually 8 or 10) then were reentered, and the resulting model was tested on the evaluation sample (odd years). Predictors were removed from the model one at a time until the optimal set of predictors that produced the greatest reduction of variance for the evaluation sample was identified. As with the BLR models, the set of “best” predictors then was reentered using all data to determine the final coefficients for each region and 3-h period. The predictors comprising the models and their physical relationships to lightning are described in section 4a.

We used a model containing only climatology and persistence (denoted L-CLIPER) as a benchmark for assessing forecast skill. Climatology consisted of the lightning frequencies and unconditional means for each 3-h period (section 3a), as well as the sine of the day number. Persistence consisted of a binary indicator for whether one or more flashes occurred during the same 3-h period the previous day, as well as the previous-day flash count. Separate L-CLIPER models were developed for each region and 3-h period.

Bothwell (2002, 2005) appears to be the first to use the PP method to develop probabilistic guidance for CG lightning (over the western United States). However, our use of pattern-type predictors differs from Bothwell who included climatological predictors for different pentads. In addition, our NB approach for forecasting the amount of lightning requires only one model to calculate PROB ≥ *T* for any count threshold, whereas Bothwell used separate BLR models for each threshold. Finally, our guidance is produced on a higher resolution grid (10 km × 10 km versus Bothwell’s 40 km × 40 km). Other than the 12 km × 12 km grid used by Reap (1994a) for 12-h forecasts, our scheme appears to provide the highest-resolution guidance currently available for lightning. It is specifically designed for use with high-resolution models.

## 4. Results

### a. Discussion of model parameters

This section describes the parameters selected for the BLR and NB models as well as their relationships to lightning occurrence. Since the equations for each 3-h period are variations on a similar theme, the physical reasoning presented here can be extended to all other times. Therefore, this discussion focuses on the most active lightning period (1800–2059 UTC).

The BLR models giving PROB ≥ 1 and the NB models for PROB ≥ *T* during the 1800–2059 UTC period are shown in Tables 5 and 6, respectively, for the four study regions (Fig. 6). The predictors and standardized coefficients are indicated. A series of diagrams displaying the frequency of one or more flashes (FREQ ≥ 1) and the unconditional mean number of flashes (MEANNF) as a function of several important predictors is shown in Figs. 8 –11. The acronyms used to describe the predictors are defined in Tables 3 and 4.

PRECPW is the most important predictor for one or more flashes (Table 5; Fig. 8a), while KI, a measure of 850–700-hPa moisture as well as stability, was selected in 3 out of the 4 NB models estimating the amount of lightning (Table 6; Fig. 8b). This finding agrees with numerous studies indicating that deep-layer moisture provides the most favorable large-scale environment for warm-season thunderstorms over Florida (e.g., López et al. 1984; Reap and MacGorman 1989; Watson et al. 1995; Mazany et al. 2002). The inclusion of a second-order term (Table 5) implies that this relationship is nonlinear (Fig. 8a), with a peak in FREQ ≥ 1 for PRECPW ∼ 5.5 cm, followed by a decline for even greater values. Largest values of PRECPW usually are associated with widespread shallow convection from tropical systems, which tends to lack the vigorous updrafts and ice processes necessary for lightning formation (Price and Rind 1992; Zipser 1994).

BESTLI was selected as the second most important parameter in the BLR models (Table 5; Fig. 9a) and also is important for predicting the amount of lightning (Table 6; Fig. 9b). The negative coefficients and the relationships depicted in Fig. 9 imply that FREQ ≥ 1 and MEANNF increase with increasing instability (i.e., as BESTLI becomes more negative). Many studies have shown that sufficient instability leading to a persistent and strong updraft is necessary for charge generation (e.g., Price and Rind 1992; Solomon and Baker 1994; Zipser 1994; Petersen and Rutledge 1998). Other stability parameters such as LCAPE and SSI were selected for other time periods.

Coincident areas of abundant moisture (PRECPW) and instability (BESTLI) are expected to be regions of high thunderstorm probability; however, storms will not develop without a source of lift. The selection of MFLXC2 in the BLR and NB models (Tables 5 and 6) indicates that boundary layer forcing is important for lightning formation (e.g., Watson et al. 1987; Reap and MacGorman 1989; Watson et al. 1991). Large MFLXC2 usually is associated with low-level convergence due to the sea breeze and other boundaries (e.g., lake/river breezes, outflows, etc.). The relationships in Figs. 10a,b show that FREQ ≥ 1 and MEANNF generally increase with greater MFLXC2. However, a nonlinear effect is evident for large negative values, possibly due to lightning occurring in the divergent stratiform regions of decaying storms.

First- and second-order terms of 1000–700-hPa mean wind (MEANU3, MEANV3) were selected in several equations (Tables 5 and 6). This relationship is nonlinear for the EC region (Figs. 11a,b), with peak lightning for offshore speeds between 2 and 4 m s^{−1}, and a decline for increasing MEANU3. Weak offshore flow produces a better developed sea breeze and greater convergence, while strong offshore flow may prevent the sea breeze from penetrating inland (McPherson 1970; Pielke 1974; Arritt 1993). Interaction terms involving MEANU3 and the distance from the coast (DISTEC, DISTWC) also were selected, implying that this relationship is modulated by proximity to the coast.

Finally, the pattern-type predictors (MTFREQ and MTMEAN) enter all of the equations (Tables 5 and 6). Although MTFREQ does not rank highly in the BLR models during the 1800–2059 UTC period, it usually is among the first selected for other time periods. Conversely, MTMEAN consistently is the most important predictor in the NB models, implying that the prevailing wind greatly influences locations where storms are most likely to persist over an area and produce large lightning counts (López and Holle 1987; Lericos et al. 2002).

Forecast maps of PROB ≥ 1 are shown in Fig. 12 for 4–5 June 2004, that is, two days of the dependent dataset. The left panels indicate the 3-hourly PROB ≥ 1 based on the RUC analyses valid at the beginning of each 3-h period, while the right panels plot the lightning strikes that occurred during each period. This example, which begins at 1200 UTC on 4 June and ends at 0300 UTC on 5 June, was a very active day with over 36 000 flashes observed. The sequence of probability maps shows the expected diurnal trend in lightning, peaking during the afternoon and then diminishing. More importantly, the agreement between the forecasts and the verification is good, with most of the observed lightning contained within the higher probability contours. A forecast example also is presented in section 5 for the 2006 independent test period.

### b. Reliability

*y*≥

*T*), were calculated by

*y*≥

*T*|

*y*≥ 1) is PROB ≥

*T*conditional on one or more flashes occurring (7), and Pr(

*y*≥ 1) is PROB ≥ 1 obtained from (2). Figures 13 and 14 show that the forecasts exhibit good reliability and are well calibrated, meaning that the event-relative frequencies are nearly identical to the forecast probabilities. Reliability for other time periods (not shown) also is very good. Probabilistic verification and skill scores relative to climatology and persistence are presented next for the 2006 independent test period.

## 5. Results for independent test period

### a. Description of model data

We applied the lightning guidance equations (section 4) to forecast output from two mesoscale models run by NCEP during the 2006 warm season and to output from local high-resolution runs of the Weather Research and Forecasting (WRF) Model for a domain over south Florida. The two NCEP models were the 1500 UTC run of the 13-km RUC (RUC13) and the 1200 UTC run of the 12-km North American Mesoscale (NAM12). The high-resolution WRF runs were initialized at 1500 UTC with NCEP 1/12° sea surface temperatures (SST) and data from the Local Analysis and Prediction System (LAPS).

A description of the RUC model was given in section 2. More detailed descriptions of RUC’s model physics and data assimilation methods can be found in Benjamin et al. (2002, 2004). We also used two versions of the WRF model to evaluate the lightning guidance equations—the 1200 UTC run of the 12-km NCEP operational NAM–WRF and a 4-km LAPS-initialized WRF (WRF–LAPS) that is run locally at the National Weather Service (NWS) Weather Forecast Office (WFO) in Miami, Florida. WRF is a state-of-the-art, regional atmospheric model for operational numerical weather prediction and atmospheric research (additional information is available online at http://www.wrf-model.org/index.php). It has two dynamical cores, the WRF–Nonhydrostatic Mesoscale Model (NMM) version developed by the Environmental Modeling Center (EMC) at NCEP and the Advanced Research WRF (ARW) developed by NCAR. Both contain a variety of options for physics packages, including cloud microphysics, boundary layer and surface processes, convective parameterizations, and shortwave and longwave radiation. These options provide sufficient sophistication so that WRF can be used for a broad range of research and operational applications (http://wrf-model.org/index.php). Janjic et al. (2004) and Skamarock et al. (2005) describe WRF’s model physics and parameterization options. A list of the various dynamics and physics options used in the 12-km NCEP NAM–WRF and the 4-km WRF–LAPS run at WFO Miami is shown in Table 7.

WRF–LAPS is part of the WRF Environmental Modeling System (WRF–EMS) that is distributed by R. Rozumalski at the University Corporation for Atmospheric Research (UCAR). WRF–EMS is a complete, full-physics, numerical weather prediction package that incorporates the dynamical cores of both the NCAR ARW and the NCEP NMM into a single end-to-end forecasting system (http://strc.comet.ucar.edu). Unlike the NCEP operational NAM–WRF, the version run at WFO Miami uses high-resolution LAPS data for model initialization (Table 7). LAPS is a diagnostic component of the Advanced Weather Information Processing System (AWIPS). LAPS produces a high-resolution three-dimensional analysis of the atmosphere by combining a background field (obtained from the 1-h forecast of the AWIPS 40-km RUC) with local data from a variety of observing systems. LAPS input data include surface observing systems, Doppler radars, satellites, wind and temperature profilers, and data from aircraft (Hiemstra et al. 2006). The LAPS analysis produced at Miami has a horizontal resolution of 5 km, with 39 vertical levels at 25-hPa intervals from 1000 to 50 hPa. It uses satellite data and level 3 reflectivity data from the Miami and Key West NWS radars to create three-dimensional diabatic analysis grids to initialize the WRF model (“hot start” initialization). The inclusion of data from local mesonetworks enhances the analysis of inland and coastal gradients, and better depicts the effects of Lake Okeechobee on surface fields (Etherton and Santos 2008).

WRF–LAPS forecasts (initialized at 1500 UTC) for the period 19–30 September 2006 were provided by P. Santos (NWS Miami), while runs for the 1 August–18 September 2006 period were produced locally at The Florida State University (FSU) using the WRF–EMS package. The LAPS, 1/12° SST, and NAM12 lateral boundary condition files that were required to produce the FSU runs were provided by P. Santos. The WRF–LAPS domain is centered on the Miami WFO county warning area (Fig. 1). The model configuration used to create the FSU runs for 1 August–18 September was identical to that used at Miami (Table 7). We did not compare results using different physics options or cumulus schemes. An examination of the sensitivity of the lightning forecasts to different model configurations is beyond the scope of this study.

Forecasts from RUC13 encompass the entire 2006 warm season (1 May–30 September), while forecasts from the NCEP operational NAM–WRF span 21 June–30 September (the NAM–WRF became operational on 20 June). We used forecasts valid every 3 h out to 12 h (i.e., the 0-, 3-, 6-, 9-, and 12-h projections). Forecast parameters needed for the lightning guidance equations were calculated from the model forecast temperature, dewpoint, wind, height, and surface pressure fields and interpolated to the 10 km × 10 km forecast grid (Fig. 1).

One should note that the 1200 UTC cycle time of the NAM–WRF hinders the forecasting of afternoon and evening lightning compared to the 1500 UTC runs of the RUC13 and WRF–LAPS. The time periods for which forecasts were available and the verification regions also differ among the three models. Thus, our intent is not to scrutinize differences in model performance. Rather, our objective simply is to describe the results that would have been achieved operationally if the most recently available run of each model had been used to generate the lightning forecasts valid at noon of each day.

### b. Forecast verification

*N*is the number of forecast–observation pairs (i.e., the number of grid points),

*f*is the forecast probability, and

_{i}*o*is the observation (set to 1 if the event occurred, or 0 if the event did not occur). The Brier score essentially is the mean of the squared differences between the forecast probabilities and the binary (0 or 1) observations. Perfect forecasts exhibit BS = 0, while less accurate forecasts have 0 < BS ≤ 1. We also calculated the Brier skill score, given by

_{i}_{MODEL}is the Brier score for the model and BS

_{REF}is the Brier score for a reference forecast. As discussed in section 3d, we developed models based on climatology and persistence (L-CLIPER) as well as persistence alone for the reference forecasts. It is clear from (11) that forecasts with a lower Brier score than the reference will have BSS > 0 (or positive skill), while forecasts with higher Brier scores than the reference forecasts have BSS < 0 (or negative skill).

Brier scores for forecasting the probability of one or more CG flashes during each 3-h period are shown in Table 8 for the 1200 UTC NCEP NAM12 (top), the 1500 UTC NCEP RUC13 (middle), and the 1500 UTC WRF–LAPS (bottom). The rightmost column shows the Brier skill score with respect to L-CLIPER (left) and persistence alone (right). Results show that Brier scores are smallest for the less active time periods, and greatest for the more active periods (i.e., 1800–2359 UTC). This occurs because Brier scores tend to be lower when the variance of the forecasts and observations is small (i.e., little lightning and small probabilities), and vice versa when the variance of forecasts and observations is large (Burrows et al. 2005). More importantly, Brier scores for all three models generally are an improvement over L-CLIPER and an even greater improvement over persistence alone through the 2100–2359 UTC period. The reader is reminded that results for each model should not be compared since they are from different initialization times (for the NAM12) and represent different time periods and different regions of the state.

Forecasting the amount of lightning is much more difficult than forecasting whether or not at least one flash will occur. Table 9 shows Brier scores and Brier skill scores for forecasting events in the 75th percentile of flashes or greater (Table 1), conditional on the occurrence of one or more flashes. It is evident that Brier scores generally are higher, and skill scores relative to L-CLIPER and persistence generally are lower compared to those for merely forecasting one or more flashes (Table 8). Nonetheless, skill scores generally are positive through the 2100–2359 UTC period. For events in the 90th percentile or greater (not shown), Brier skill scores are slightly positive or near zero at most time periods, with only the RUC13 producing positive skill through the 2100–2359 UTC period. This finding suggests that events in the 90th or greater percentiles are near the threshold of predictability, at least for forecasts longer than 3–6 h. This is likely due to an inherent weakness in the PP method, namely, the tendency to forecast extreme events unreliably when the accuracy of the predictors (i.e., the NWP model forecasts) decreases to the point where it becomes inadvisable to attempt to predict the extremes of the distribution (Glahn et al. 1991).

The Brier score computations (Tables 8 and 9) require that lightning events occur within a 10-km radius of a grid point. This is a very strict verification criterion that does not consider forecasts at neighboring grid points. A more relaxed approach is to use the maximum probability or the average probability within a certain radius of each grid point (e.g., Burrows et al. 2005). The maximum or average probability then is used in the Brier score computations instead of the gridpoint specific values. We determined that maximum probabilities within 20 km of each grid point generally give the best improvement relative to L-CLIPER and persistence. The results using this new verification approach are shown in Tables 10 and 11 for forecasting one or more flashes and forecasting events in the 75th percentile or greater, respectively. In most cases, magnitudes of the Brier scores remain relatively unchanged; however, Brier skill scores generally are higher when using the more relaxed verification criteria, especially during the first 6 h of the forecast. For reasons that are not clear, the relaxed criteria most benefit L-CLIPER and persistence during the later forecast periods (as evident by the lower skill scores for these periods in Tables 10 and 11).

As one would expect, skill scores deteriorate beyond the 6–9-h projections (Tables 8 –11) as errors in the position and magnitude of predicted convection and synoptic-scale features increase with time in the driving NWP models. Nonetheless, the results shown in Tables 8 –11 are encouraging, especially considering that L-CLIPER alone generally produces very good forecasts during Florida’s warm season. In fact, L-CLIPER is the most difficult standard of reference to beat since it represents an optimal linear combination of both climatology and previous-day persistence. L-CLIPER becomes particularly difficult to beat in situations when the synoptic pattern on a particular day is similar to that of the previous day, which often is the case during Florida’s warm season.

Figures 15 and 16 contain reliability diagrams for forecasting the probability of one or more flashes and the probability of ≥75th percentile events, respectively, for the most active lightning period (1800–2059 UTC). Results for the three mesoscale models are plotted on the same graph. Forecasts from the RUC13 lie reasonably close to the 1:1 line in both plots. However, forecasts for the NAM12 and WRF–LAPS show an underforecasting bias in the lower half of the probability range, and then bend back toward the 1:1 line for higher forecast probabilities. The reason for this behavior is not entirely clear, but most likely is due to inherent biases in the model moisture, temperature, and/or wind field forecasts. Reliability plots for other forecast projections (not shown) indicate similar biases. Reliability when forecasting the 90th percentile or greater (not shown) is similar to that for the 75th percentile (Fig. 16), but tend to deviate a bit more from the 1:1 line for higher forecast probabilities.

### c. Forecast example

An example lightning probability forecast for 16–17 August 2006 using the 1500 UTC WRF–LAPS model (Fig. 17) is described next. A map containing county names and labeled geographical features is shown in Fig. 18. Results using the RUC13 and NAM–WRF on this day (not shown) compare favorably with those from WRF–LAPS (Fig. 17). The figure panels show the probability of one or more flashes (left panels), the unconditional probability of ≥90th percentile events (center), and the CG strike verification (right panels) for four 3-h time periods. The flow pattern on this day is type A (section 3), with prevailing southeasterly low-level flow and no synoptic or tropical influences. Between 1500 and 1759 UTC, the greatest probability of one or more flashes (between 30% and 40%) is forecast over eastern Broward and northern Miami–Dade (MD) counties, with probabilities of 10% or greater for areas south of Lake Okeechobee (LOK) (Fig. 17a). Forecast probabilities beginning at 1800 UTC (Figs. 17d,e) are considerably greater than those at 1500 UTC across south Florida (Figs. 17a,b), with the greatest values concentrated along the west coast as well as eastern Palm Beach (PB), Broward, and MD counties. The verification (Fig. 17f) reveals a significant increase in activity (over 7000 flashes) along the west coast and over Broward and MD counties. With the exception of the activity south of LOK, this verification agrees well with the forecast probabilities (Figs. 17d,e). Forecast probabilities for the 2100–2359 UTC period (Figs. 17g,h) have increased south of LOK, and lightning occurs just east of this area over central and western PB and Broward counties (Fig. 17i). Although the area of enhanced probabilities north of LOK does not verify during this period (Fig. 17i), lightning does occur there only one hour earlier, that is, between 2000 and 2030 UTC (Fig. 17f). Finally, forecast probabilities for the 0000–0259 UTC period (Figs. 17j,k) show a diminishing lightning threat, and indeed little activity occurs during this period (Fig. 17l).

The example in Fig. 17 is typical of many others during the 1 August–30 September 2006 period. That is, the sequence of probability maps shows the expected diurnal trend in lightning that peaks during the afternoon and then diminishes. The lightning forecasts generally show good agreement with the verification, with most of the observed lightning occurring within the higher forecast probability contours. However, as observed on 16–17 August, the timing and placement of lightning is not perfect. Nonetheless, the forecasts do capture the general spatial and temporal trends in observed lightning at a level of detail that, to our knowledge, has not been reported previously.

## 6. Summary and conclusions

The major objective of this study was to develop a high-resolution, gridded forecast guidance product for warm season (May–September) CG lightning over Florida using the PP method. Four warm seasons of NLDN data and an archive of RUC20 analyses were used to develop equations for PROB ≥ 1, as well as PROB ≥ *T*, for each 3-h period on a 10 km × 10 km grid. RUC analyses of sea level pressure and a map-typing technique were used to develop lightning frequencies for five dominant flow types to capture small-scale enhancements due to local forcing that are not well resolved by NWP models. The map-type frequencies as well as RUC-analyzed parameters describing temperature, moisture, stability, and wind were used as candidate predictors in the PP equations.

BLR was used to develop the equations for PROB ≥ 1, while NB regression was used to model the number of flashes conditional on one or more flashes occurring. The NB distribution provided the best fit to the underlying count distribution, enabling PROB ≥ *T* for any count threshold (*T*) to be calculated. A principal component analysis was used to select a subset of the most physically relevant predictors that contain less mutual correlation. A combination of stepwise screening and cross validation was used to select the best combination of predictors that was most likely to generalize to independent data. Five variables were found to have the greatest influence on the likelihood of one or more flashes and the amount of lightning. They are deep-layer moisture (PRECPW), instability (BESTLI, LCAPE), boundary layer forcing (MFLXC2), pattern type (MTFREQ, MTMEAN), and the low-level wind (MEANU3, MEANV3).

The equations were applied to forecast output from three mesoscale models during an independent test period (1 May–30 September 2006): the 1500 UTC NCEP RUC13, the 1200 UTC NCEP NAM–WRF, and high-resolution runs of the WRF initialized at 1500 UTC with LAPS and NCEP 1/12° SST data (WRF–LAPS). When forecasting one or more flashes, all three mesoscale models generally showed positive Brier skill scores relative to L-CLIPER and persistence alone through the 2100–2359 UTC period. Skill was found to deteriorate beyond 6–9 h as errors in the model forecasts increased with time in the driving NWP models. The models also showed some improvement over L-CLIPER and persistence when forecasting events in the 75th or greater percentiles, with limited skill for events in the 90th percentile or greater.

An example probability forecast was shown for 16–17 August 2006 using the 0-, 3-, 6-, and 9-h forecasts from the 1500 UTC WRF–LAPS. The sequence of probability maps for each 3-h period exhibited the expected diurnal trend in lightning. The exact timing and placement of lightning was not perfect; however, there generally was good agreement between the forecasts and the verification, with most of the observed lightning occurring within the higher forecast probability contours.

The positive skill demonstrated by the RUC13, NAM–WRF, and WRF–LAPS (through 2100–2359 UTC) during the 1 May–30 September 2006 test period provides some evidence that the PP scheme is model independent. Testing on a larger independent sample (i.e., multiple warm seasons) is needed to further support this hypothesis. The results also demonstrate that a model initialized with high-resolution LAPS and SST data is capable of producing skillful lightning forecasts.

The guidance that we have developed represents an important step toward more precise and timely lightning forecasts. Our results indicate that skillful lightning forecasts out to 6–9 h are possible using high-resolution models. In addition, a major strength of the PP method is that the inevitable changes that will occur in the NWP models will not require redevelopment of the equations, and in fact, should only improve the forecasts (Wilks 2006). Conversely, the main drawback is that the PP scheme assumes a “perfect” forecast/analysis of the predictors by the NWP model and thus does not account for, or correct, any type of NWP forecast error. Thus, it appears that some kind of MOS procedure would be needed to produce skillful forecasts beyond 9–12 h. Nonetheless, as model resolution, physics, and data assimilation methods continue to improve, better lightning forecasts are expected to result.

Our methodology is an enhancement to schemes already in use (e.g., Bothwell 2002, 2005). Further improvements will be achieved through future work. For example, temporal resolution can be increased by developing separate PP equations for each hour, which then could be applied to hourly forecasts from the RUC13. Plans already are in place to incorporate the lightning guidance into the Interactive Forecast Preparation System (IFPS) Graphical Forecast Editor (GFE) at the Tallahassee NWS office. A forecaster then can use output from one NWP model or a blend of two or more models to generate lightning probabilities using a “smart tool” in GFE. The resulting lightning forecasts then could be accessed by the public through NWS Web sites and used by the NWS in forecast products.

Future work also will seek to expand the scheme to other parts of the country. Since some of the statistical assumptions made for Florida may not be applicable to other areas, appropriate modifications will be needed. In addition, future efforts should determine whether a Bayesian framework produces better results than those achieved in this study using more classical methods. In the longer term, the availability of higher-resolution analyses (e.g., ≤10 km) and a larger developmental sample should produce a greater reduction of variance and more robust estimates of the model parameters. Accurate cloud analyses and improved NWP model forecasts of cloud hydrometeor profiles will permit cloud microphysical parameters to be included in the equations.

## Acknowledgments

This research was funded by Florida Power & Light Corporation (FP&L). We appreciate the assistance of Drs. James Elsner and Thomas Jagger of the Department of Geography at The Florida State University for their helpful suggestions and advice on many statistical aspects of the research. Appreciation also is extended to Dr. Pablo Santos at the NWS in Miami for providing the initialization datasets required to produce the WRF runs. We also thank Paul Hebert from FP&L for providing many suggestions based on his extensive knowledge of summertime sea-breeze weather patterns in Florida. Last, we thank the three anonymous reviewers for their very helpful comments and suggestions for improving the manuscript.

## REFERENCES

Arritt, R. W., 1993: Effects of the large-scale flow on characteristic features of the sea breeze.

,*J. Appl. Meteor.***32****,**116–125.Benjamin, S. G., and Coauthors, 2002: RUC20—The 20-km version of the Rapid Update Cycle. NWS Technical Procedures Bulletin 490, 30 pp.

Benjamin, S. G., G. A. Grell, J. M. Brown, and T. G. Smirnova, 2004: Mesoscale weather prediction with the RUC hybrid isentropic-terrain-following coordinate model.

,*Mon. Wea. Rev.***132****,**473–494.Bothwell, P. D., 2002: Prediction of cloud-to-ground lightning in the western United States. Ph.D. thesis, University of Oklahoma, 178 pp.

Bothwell, P. D., 2005: Development of an operational statistical scheme to predict the location and intensity of lightning. Preprints,

*Conf. on Meteorological Applications of Lightning Data,*San Diego, CA, Amer. Meteor. Soc., 4.2. [Available online at http://ams.confex.com/ams/pdfpapers/85013.pdf.].Brier, G. W., 1950: Verification of forecasts expressed in terms of probability.

,*Mon. Wea. Rev.***78****,**1–3.Brunet, N., R. Verret, and N. Yacowar, 1988: An objective comparison of model output statistics and “perfect prog” in producing numerical weather element forecasts.

,*Wea. Forecasting***3****,**273–283.Burrows, W. R., C. Price, and L. J. Wilson, 2005: Warm season lightning probability prediction for Canada and the northern United States.

,*Wea. Forecasting***20****,**971–988.Camp, J. P., A. I. Watson, and H. E. Fuelberg, 1998: The diurnal distribution of lightning over north Florida and its relation to the prevailing low-level flow.

,*Wea. Forecasting***13****,**729–739.Crawley, M. J., 2002:

*Statistical Computing: An Introduction to Data Analysis in S-Plus*. John Wiley and Sons, 772 pp.Cummins, K. L., M. J. Murphy, E. A. Bardo, W. L. Hiscox, R. B. Pyle, and A. E. Pifer, 1998: A combined TOA/MDF technology upgrade of the U.S. National Lightning Detection Network.

,*J. Geophys. Res.***103****,**9035–9044.Cummins, K. L., J. A. Cramer, C. J. Biagi, E. P. Krider, J. Jerauld, M. A. Uman, and V. A. Rakov, 2006: The U.S. National Lightning Detection Network: Post-upgrade status. Preprints,

*Conf. on Meteorological Applications of Lightning Data,*Atlanta, GA, Amer. Meteor. Soc., 6.1. [Available online at http://ams.confex.com/ams/pdfpapers/105142.pdf.].Curran, E. B., R. L. Holle, and R. E. López, 2000: Lightning casualties and damages in the United States from 1959 to 1994.

,*J. Climate***13****,**3448–3464.Elsner, J. B., and C. P. Schmertmann, 1993: Improving extended-range seasonal predictions of intense Atlantic hurricane activity.

,*Wea. Forecasting***8****,**345–351.Elsner, J. B., and T. H. Jagger, 2004: A hierarchical Bayesian approach to seasonal hurricane modeling.

,*J. Climate***17****,**2813–2827.Estoque, M. A., 1962: The sea breeze as a function of the prevailing synoptic situation.

,*J. Atmos. Sci.***19****,**244–250.Etherton, B., and P. Santos, 2008: Sensitivity of WRF forecasts for south Florida to initial conditions.

, in press.*Wea. Forecasting*Falls, L. W., W. O. Williford, and M. C. Carter, 1971: Probability distributions for thunderstorm activity at Cape Kennedy, Florida.

,*J. Appl. Meteor.***10****,**97–104.Forecast Systems Laboratory, cited. 2006: RUC-testing. [Available online at http://maps.fsl.noaa.gov/ruc13_docs/RUC-testing-Jun06.htm.].

Gardner, W., E. P. Mulvey, and E. C. Shaw, 1995: Regression analysis of counts and rates: Poisson, overdispersed Poisson, and negative binomial models.

,*Psychol. Bull.***118****,**3. 392–404.Glahn, H. R., and D. A. Lowry, 1972: The use of model output statistics (MOS) in objective weather forecasting.

,*J. Appl. Meteor.***11****,**1203–1211.Glahn, H. R., A. H. Murphy, L. J. Wilson, and J. S. Jensenius Jr., 1991: Lectures presented at the WMO training workshop on the interpretation of NWP products in terms of local weather phenomena and their verification. WMO Tech. Doc. 421, 340 pp.

Glickman, T., 2000:

*Glossary of Meteorology*. 2nd ed. Amer. Meteor. Soc., 855 pp.Gremillion, M. S., and R. E. Orville, 1999: Thunderstorm characteristics of cloud-to-ground lightning at the Kennedy Space Center, Florida: A study of lightning initiation signatures as indicated by the WSR-88D.

,*Wea. Forecasting***12****,**640–649.Hiemstra, C. A., G. E. Liston, R. A. Pielke, D. L. Birkenheuer, and S. C. Albers, 2006: Comparing Local Analysis and Prediction System (LAPS) assimilations with independent observations.

,*Wea. Forecasting***21****,**1024–1040.Hodanish, S., D. Sharp, W. Collins, C. Paxton, and R. E. Orville, 1997: A 10-year monthly lightning climatology of Florida: 1986–95.

,*Wea. Forecasting***12****,**439–448.Holle, R. L., R. E. López, and E. B. Curran, 1999: Distributions of lightning caused casualties and damages since 1959 in the United States. Preprints,

*11th Conf. on Applied Climatology,*Dallas, TX, Amer. Meteor. Soc., 363–370.Hosmer, D. W., and S. Lemeshow, 1989:

*Applied Logistic Regression*. John Wiley and Sons, 307 pp.Hughes, K. K., 2001: Development of MOS thunderstorm and severe thunderstorm forecast equations with multiple data sources. Preprints,

*18th Conf. on Weather Analysis and Forecasting,*Fort Lauderdale, FL, Amer. Meteor. Soc., 191–195.Hughes, K. K., 2002: Automated gridded forecast guidance for thunderstorms and severe local storms based on the Eta model. Preprints,

*19th Conf. on Weather Analysis and Forecasting,*San Antonio, TX, Amer. Meteor. Soc., J19–J22. [Available online at http://ams.confex.com/ams/pdfpapers/47497.pdf.].Janjic, Z., T. Black, M. Pyle, H. Chuang, E. Rogers, and G. DiMego, cited. 2004: An evolutionary approach to non-hydrostatic modeling. [Available online at http://www.ncep.noaa.gov/nwp50/Presentations/.].

Klein, W. H., 1971: Computer prediction of precipitation probability in the United States.

,*J. Appl. Meteor.***10****,**903–915.Klein, W. H., and H. R. Glahn, 1974: Forecasting local weather by means of model output statistics.

,*Bull. Amer. Meteor. Soc.***55****,**1217–1227.Klein, W. H., B. M. Lewis, and I. Enger, 1959: Objective prediction of five day mean temperature during winter.

,*J. Meteor.***16****,**672–682.Laird, N. F., D. A. R. Kristovich, R. M. Rauber, H. T. Ochs III, and L. J. Miller, 1995: The Cape Canaveral sea and river breezes: Kinematic structure and convective initiation.

,*Mon. Wea. Rev.***123****,**2942–2956.Lambert, W. C., M. Wheeler, and W. Roeder, 2005: Objective lightning forecasting at Kennedy Space Center and Cape Canaveral Air Force Station using cloud-to-ground lightning surveillance system data. Preprints,

*Conf. on Meteorological Applications of Lightning Data,*San Diego, CA, Amer. Meteor. Soc., 4.1. [Available online at http://ams.confex.com/ams/pdfpapers/85944.pdf.].Lehmiller, G. S., T. B. Kimberlain, and J. B. Elsner, 1997: Seasonal prediction models for North Atlantic basin hurricane location.

,*Mon. Wea. Rev.***125****,**1780–1791.Lericos, T. P., H. E. Fuelberg, A. I. Watson, and R. L. Holle, 2002: Warm season lightning distributions over the Florida peninsula as related to synoptic patterns.

,*Wea. Forecasting***17****,**83–98.Livingston, E. S., J. W. Nielson-Gammon, and R. E. Orville, 1996: A climatology, synoptic assessment, and thermodynamic evaluation for cloud-to-ground lightning in Georgia: A study for the 1996 Summer Olympics.

,*Bull. Amer. Meteor. Soc.***77****,**1483–1495.López, R. E., and R. L. Holle, 1987: The distribution of summertime lightning as a function of low-level wind flow in central Florida. NOAA Tech. Memo. ERL ESG-28, National Severe Storms Laboratory, Norman, OK, 43 pp.

López, R. E., P. T. Gannon Sr., D. O. Blanchard, and C. C. Balch, 1984: Synoptic and regional circulation parameters associated with the degree of convective shower activity in south Florida.

,*Mon. Wea. Rev.***112****,**686–703.Lund, I. A., 1963: Map-pattern classification by statistical methods.

,*J. Appl. Meteor.***2****,**56–66.MacGorman, D. R., and W. D. Rust, 1998:

*The Electrical Nature of Storms*. Oxford University Press, 422 pp.Mazany, R. A., S. Businger, S. I. Gutman, and W. Roeder, 2002: A lightning prediction index that utilizes GPS integrated precipitable water vapor.

,*Wea. Forecasting***17****,**1034–1047.McPherson, R. D., 1970: A numerical study of the effect of a coastal irregularity on the sea breeze.

,*J. Appl. Meteor.***9****,**767–777.Neumann, C. J., and J. R. Nicholson, 1972: Multivariate regression techniques applied to thunderstorm forecasting at the Kennedy Space Center. Preprints,

*Int. Conf. on Aerospace and Aeronautical Meteorology,*Washington, DC, Amer. Meteor. Soc., 6–13.Orville, R. E., 1994: Cloud-to-ground lightning flash characteristics in the contiguous United States: 1989-1991.

,*J. Geophys. Res.***99****,**10833–10841.Orville, R. E., and G. R. Huffines, 2001: Cloud-to-ground lightning in the United States: NLDN results in the first decade, 1989–98.

,*Mon. Wea. Rev.***129****,**1179–1193.Orville, R. E., G. R. Huffines, W. R. Burrows, R. L. Holle, and K. L. Cummins, 2002: The North American Lightning Detection Network (NALDN)—First results: 1998–2002.

,*Mon. Wea. Rev.***130****,**2098–2109.Petersen, W. A., and S. A. Rutledge, 1998: On the relationship between cloud-to-ground lightning and convective rainfall.

,*J. Geophys. Res.***103****,**14025–14040.Pielke, R. A., 1974: A three-dimensional numerical model of the sea breezes over south Florida.

,*Mon. Wea. Rev.***102****,**115–139.Price, C., and D. Rind, 1992: A simple lightning parameterization for calculating global lightning distributions.

,*J. Geophys. Res.***97****,**9919–9933.Price, C., and D. Rind, 1993: Modeling global lightning distributions in a general circulation model.

,*Mon. Wea. Rev.***122****,**1930–1939.Randell, S. C., S. A. Rutledge, R. D. Farley, and J. H. Helsdon Jr., 1994: A modeling study on the early electrical development of tropical convection: Continental and oceanic (monsoon) storms.

,*Mon. Wea. Rev.***122****,**1852–1877.Reap, R. M., 1994a: Analysis and prediction of lightning strike distributions associated with synoptic map types over Florida.

,*Mon. Wea. Rev.***122****,**1698–1715.Reap, R. M., 1994b: 24-h NGM based probability and categorical forecasts of thunderstorms and severe local storms for the contiguous U.S. NWS Technical Procedures Bulletin 419, 14 pp.

Reap, R. M., and D. S. Foster, 1979: Automated 12-36 hour probability forecasts of thunderstorms and severe local storms.

,*J. Appl. Meteor.***18****,**1304–1315.Reap, R. M., and D. R. MacGorman, 1989: Cloud-to-ground lightning: Climatological characteristics and relationships to model fields, radar observations, and severe local storms.

,*Mon. Wea. Rev.***117****,**518–535.Reynolds, S. E., M. Brook, and M. F. Gourley, 1957: Thunderstorm charge separation.

,*J. Meteor.***14****,**426–436.Sakamoto, C. M., 1973: Application of the Poisson and negative binomial models to thunderstorm and hail days probabilities in Nevada.

,*Mon. Wea. Rev.***101****,**350–355.Shafer, P. E., and H. E. Fuelberg, 2006: A statistical procedure to forecast warm season lightning over portions of the Florida peninsula.

,*Wea. Forecasting***21****,**851–868.Simpson, J. E., 1994:

*Sea Breeze and Local Wind*. Cambridge University Press, 234 pp.Skamarock, W. C., J. B. Klemp, J. Dudhia, D. O. Gill, D. M. Barker, W. Wang, and J. G. Powers, 2005: A description of the Advanced Research WRF Version 2. NCAR Tech. Note NCAR/TN-468+STR, 88 pp. [Available online at http://www.mmm.ucar.edu/wrf/users/docs/arw_v2.pdf.].

Solomon, R., and M. Baker, 1994: Electrification of New Mexico thunderstorms.

,*Mon. Wea. Rev.***122****,**1878–1886.Solomon, R., and M. Baker, 1998: Lightning flash rate and type in convective storms.

,*J. Geophys. Res.***103****,**14041–14057.Steiger, S. M., R. E. Orville, and G. Huffines, 2002: Cloud-to-ground lightning characteristics over Houston, Texas: 1989-2000.

,*J. Geophys. Res.***107****.**4117, doi:10.1029/2001JD001142.Ushio, T., S. J. Heckman, D. J. Boccippio, and H. J. Christian, 2001: A survey of thunderstorm flash rates compared to cloud top height using TRMM satellite data.

,*J. Geophys. Res.***106****,**24089–24095.Vonnegut, B., 1963: Some facts and speculations concerning the origin and role of thunderstorm electricity.

*Severe Local Storms, Meteor. Monogr.,*No. 27, Amer. Meteor. Soc., 224–241.Watson, A. I., R. E. López, R. L. Holle, and J. R. Daughetry, 1987: The relationship of lightning to surface convergence at Kennedy Space Center: A preliminary study.

,*Wea. Forecasting***2****,**140–157.Watson, A. I., R. L. Holle, R. E. López, R. Ortiz, and J. R. Nicholson, 1991: Surface wind convergence as a short-term predictor of cloud-to-ground lightning at Kennedy Space Center.

,*Wea. Forecasting***6****,**49–64.Watson, A. I., R. L. Holle, and R. E. López, 1995: Lightning from two national detection networks related to vertically integrated liquid and echo-top information from WSR-88D radar.

,*Wea. Forecasting***10****,**592–605.Westcott, N. E., 1995: Summertime cloud-to-ground lightning activity around major Midwestern urban areas.

,*J. Appl. Meteor.***34****,**1633–1642.Wexler, R., 1946: Theory and observations of land and sea breezes.

,*Bull. Amer. Meteor. Soc.***27****,**272–287.Wilks, D. S., 2006:

*Statistical Methods in the Atmospheric Sciences*. International Geophysics Series, Vol. 91, Academic Press, 627 pp.Williams, E. R., 1985: Large-scale charge separation in thunderclouds.

,*J. Geophys. Res.***90****,**6013–6025.Williams, E. R., M. E. Weber, and R. E. Orville, 1989: The relationship between lightning type and convective state of thunderclouds.

,*J. Geophys. Res.***94****,**13213–13220.Williford, W. O., M. C. Carter, and P. Hsieh, 1974: A Bayesian analysis of two probability models describing thunderstorm activity at Cape Kennedy, Florida.

,*J. Appl. Meteor.***13****,**718–725.Zipser, E. J., 1994: Deep cumulonimbus cloud systems in the tropics with and without lightning.

,*Mon. Wea. Rev.***122****,**1837–1851.

Illustration of two counting regions (box and circle) with equal areas centered on a point of influence P, with two lightning strike locations A and B.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Illustration of two counting regions (box and circle) with equal areas centered on a point of influence P, with two lightning strike locations A and B.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Illustration of two counting regions (box and circle) with equal areas centered on a point of influence P, with two lightning strike locations A and B.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

(left) Composite sea level pressure and (right) spatial distribution of the unconditional mean number of flashes for the 1800–2059 UTC period for: (a),(b) type A; (c),(d) type B; (e),(f) type C; (g),(h) type D; and (i),(j) type E. The inferred low-level wind is indicated by arrows on each map.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

(left) Composite sea level pressure and (right) spatial distribution of the unconditional mean number of flashes for the 1800–2059 UTC period for: (a),(b) type A; (c),(d) type B; (e),(f) type C; (g),(h) type D; and (i),(j) type E. The inferred low-level wind is indicated by arrows on each map.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

(left) Composite sea level pressure and (right) spatial distribution of the unconditional mean number of flashes for the 1800–2059 UTC period for: (a),(b) type A; (c),(d) type B; (e),(f) type C; (g),(h) type D; and (i),(j) type E. The inferred low-level wind is indicated by arrows on each map.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Histogram of the distribution of flash counts during the 1800–2059 UTC period for all cases when one or more flashes occurred. The histogram has been truncated at 200 flashes to emphasize the lower part of the distribution. Histogram bins are five flashes.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Histogram of the distribution of flash counts during the 1800–2059 UTC period for all cases when one or more flashes occurred. The histogram has been truncated at 200 flashes to emphasize the lower part of the distribution. Histogram bins are five flashes.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Histogram of the distribution of flash counts during the 1800–2059 UTC period for all cases when one or more flashes occurred. The histogram has been truncated at 200 flashes to emphasize the lower part of the distribution. Histogram bins are five flashes.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Flash count probability distribution for the 1800–2059 UTC period implied from the Poisson and negative binomial regression models. The observed frequency also is shown for comparison.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Flash count probability distribution for the 1800–2059 UTC period implied from the Poisson and negative binomial regression models. The observed frequency also is shown for comparison.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Flash count probability distribution for the 1800–2059 UTC period implied from the Poisson and negative binomial regression models. The observed frequency also is shown for comparison.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Map of the original nine areas for which equations first were developed. The dark shaded areas represent the final four regions used to develop equations.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Map of the original nine areas for which equations first were developed. The dark shaded areas represent the final four regions used to develop equations.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Map of the original nine areas for which equations first were developed. The dark shaded areas represent the final four regions used to develop equations.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Illustration of the 10 km × 10 km grid with disjoint circular counting regions used in the regression analysis. RUC-derived predictors are defined at each 10-km grid point.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Illustration of the 10 km × 10 km grid with disjoint circular counting regions used in the regression analysis. RUC-derived predictors are defined at each 10-km grid point.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Illustration of the 10 km × 10 km grid with disjoint circular counting regions used in the regression analysis. RUC-derived predictors are defined at each 10-km grid point.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Plots for (a) the frequency of one or more flashes as a function of precipitable water, and (b) the unconditional mean number of flashes as a function of K index for all regions combined during the 1800–2059 UTC period.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Plots for (a) the frequency of one or more flashes as a function of precipitable water, and (b) the unconditional mean number of flashes as a function of K index for all regions combined during the 1800–2059 UTC period.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Plots for (a) the frequency of one or more flashes as a function of precipitable water, and (b) the unconditional mean number of flashes as a function of K index for all regions combined during the 1800–2059 UTC period.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Plots for (a) the frequency of one or more flashes and (b) the unconditional mean number of flashes as a function of best lifted index for all regions combined during the 1800–2059 UTC period.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Plots for (a) the frequency of one or more flashes and (b) the unconditional mean number of flashes as a function of best lifted index for all regions combined during the 1800–2059 UTC period.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Plots for (a) the frequency of one or more flashes and (b) the unconditional mean number of flashes as a function of best lifted index for all regions combined during the 1800–2059 UTC period.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Plots for (a) the frequency of one or more flashes and (b) the unconditional mean number of flashes as a function of 1000-hPa moisture flux convergence for all regions combined during the 1800–2059 UTC period.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Plots for (a) the frequency of one or more flashes and (b) the unconditional mean number of flashes as a function of 1000-hPa moisture flux convergence for all regions combined during the 1800–2059 UTC period.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Plots for (a) the frequency of one or more flashes and (b) the unconditional mean number of flashes as a function of 1000-hPa moisture flux convergence for all regions combined during the 1800–2059 UTC period.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Plots for (a) the frequency of one or more flashes, and (b) the unconditional mean number of flashes, as a function of 1000–700-hPa mean *u*-wind component for the East Coast region (1800–2059 UTC period).

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Plots for (a) the frequency of one or more flashes, and (b) the unconditional mean number of flashes, as a function of 1000–700-hPa mean *u*-wind component for the East Coast region (1800–2059 UTC period).

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Plots for (a) the frequency of one or more flashes, and (b) the unconditional mean number of flashes, as a function of 1000–700-hPa mean *u*-wind component for the East Coast region (1800–2059 UTC period).

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

(left) Probability of one or more flashes based on RUC analyses and (right) lightning strike verification for 4–5 Jun 2004: (a),(b) 1200–1459; (c),(d) 1500–1759; (e),(f) 1800–209; (g),(h) 2100–2359; and (i),(j) 0000–0259 UTC.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

(left) Probability of one or more flashes based on RUC analyses and (right) lightning strike verification for 4–5 Jun 2004: (a),(b) 1200–1459; (c),(d) 1500–1759; (e),(f) 1800–209; (g),(h) 2100–2359; and (i),(j) 0000–0259 UTC.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

(left) Probability of one or more flashes based on RUC analyses and (right) lightning strike verification for 4–5 Jun 2004: (a),(b) 1200–1459; (c),(d) 1500–1759; (e),(f) 1800–209; (g),(h) 2100–2359; and (i),(j) 0000–0259 UTC.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Reliability diagram for the logistic models predicting the probability of one or more flashes. The results are for the dependent data sample and all regions combined during the 1800–2059 UTC period.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Reliability diagram for the logistic models predicting the probability of one or more flashes. The results are for the dependent data sample and all regions combined during the 1800–2059 UTC period.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Reliability diagram for the logistic models predicting the probability of one or more flashes. The results are for the dependent data sample and all regions combined during the 1800–2059 UTC period.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Reliability diagrams for the unconditional probability of (a) ≥50th, (b) ≥75th, (c) ≥90th, and (d) ≥95th percentiles of flash count. The results are for the dependent data sample and all regions combined during the 1800–2059 UTC period.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Reliability diagrams for the unconditional probability of (a) ≥50th, (b) ≥75th, (c) ≥90th, and (d) ≥95th percentiles of flash count. The results are for the dependent data sample and all regions combined during the 1800–2059 UTC period.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Reliability diagrams for the unconditional probability of (a) ≥50th, (b) ≥75th, (c) ≥90th, and (d) ≥95th percentiles of flash count. The results are for the dependent data sample and all regions combined during the 1800–2059 UTC period.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Reliability diagram for the logistic models predicting the probability of one or more flashes during the 2006 independent test period. Results for the 1200 UTC NCEP NAM12, the 1500 UTC NCEP RUC13, and the 1500 UTC 4-km WRF–LAPS are shown for the 1800–2059 UTC period.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Reliability diagram for the logistic models predicting the probability of one or more flashes during the 2006 independent test period. Results for the 1200 UTC NCEP NAM12, the 1500 UTC NCEP RUC13, and the 1500 UTC 4-km WRF–LAPS are shown for the 1800–2059 UTC period.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Reliability diagram for the logistic models predicting the probability of one or more flashes during the 2006 independent test period. Results for the 1200 UTC NCEP NAM12, the 1500 UTC NCEP RUC13, and the 1500 UTC 4-km WRF–LAPS are shown for the 1800–2059 UTC period.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Reliability diagram for the negative binomial models predicting the probability of ≥75th percentile of flash count during the 2006 independent test period. Results for the 1200 UTC NCEP NAM12, the 1500 UTC NCEP RUC13, and the 1500 UTC 4-km WRF–LAPS are shown for the 1800–2059 UTC period.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Reliability diagram for the negative binomial models predicting the probability of ≥75th percentile of flash count during the 2006 independent test period. Results for the 1200 UTC NCEP NAM12, the 1500 UTC NCEP RUC13, and the 1500 UTC 4-km WRF–LAPS are shown for the 1800–2059 UTC period.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Reliability diagram for the negative binomial models predicting the probability of ≥75th percentile of flash count during the 2006 independent test period. Results for the 1200 UTC NCEP NAM12, the 1500 UTC NCEP RUC13, and the 1500 UTC 4-km WRF–LAPS are shown for the 1800–2059 UTC period.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

(left) Probability of one or more flashes, (center) unconditional probability of ≥90th percentile, and (right) CG strike verification superimposed on the one or more flash probabilities for 16–17 Aug 2006 based on 1500 UTC WRF–LAPS: (a)–(c) 0-, (d)–(f) 3-, (g)–(i) 6-, and (j)–(l) 9-h forecast projections. Valid time periods are shown above each plot. Note the different color scales for the left and center panels. Grid points that received ≥90th percentile of flash count are indicated by the “x” symbols on the center panels.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

(left) Probability of one or more flashes, (center) unconditional probability of ≥90th percentile, and (right) CG strike verification superimposed on the one or more flash probabilities for 16–17 Aug 2006 based on 1500 UTC WRF–LAPS: (a)–(c) 0-, (d)–(f) 3-, (g)–(i) 6-, and (j)–(l) 9-h forecast projections. Valid time periods are shown above each plot. Note the different color scales for the left and center panels. Grid points that received ≥90th percentile of flash count are indicated by the “x” symbols on the center panels.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

(left) Probability of one or more flashes, (center) unconditional probability of ≥90th percentile, and (right) CG strike verification superimposed on the one or more flash probabilities for 16–17 Aug 2006 based on 1500 UTC WRF–LAPS: (a)–(c) 0-, (d)–(f) 3-, (g)–(i) 6-, and (j)–(l) 9-h forecast projections. Valid time periods are shown above each plot. Note the different color scales for the left and center panels. Grid points that received ≥90th percentile of flash count are indicated by the “x” symbols on the center panels.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Map of south Florida with county names and geographical features labeled.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Map of south Florida with county names and geographical features labeled.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Map of south Florida with county names and geographical features labeled.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR2222.1

Conditional percentiles of CG flash count for the four most active 3-h periods. The percentiles are based on CG flash counts for all land grid points in the domain during 11 warm-season periods (1995–2005).

Number of RUC-analyzed sea level pressure maps classified into each type at a correlation threshold of 0.70, using 3-hourly data from the 1998–2005 warm seasons (9613 available maps). For purposes of developing the equations, maps that could not be classified into a type were assigned the type with which they were most correlated.

RUC-analyzed parameters investigated for inclusion in the candidate predictor pool. The name of each variable, a description (where needed), and abbreviations are included.

Final list of RUC-derived candidate predictors used to develop the regression models. Power terms and two-way cross products for each parameter also were included in the final predictor pool (not listed). Spearman rank correlations with the binary (yes/no) lightning predictands for the 1800–2059 UTC period also are shown for the East Coast region.

Logistic regression models for the probability of one or more flashes during the 1800–2059 UTC period. The regression coefficients for each of the four regions are shown. Parameters not selected for inclusion are indicated by a dash.

Negative binomial regression models for the number of flashes during the 1800–2059 UTC period. The regression coefficients and the estimated shape parameter for each of the four regions are shown. Parameters not selected for inclusion are indicated by a dash.

Comparison of NCEP NAM–WRF and WFO Miami WRF–LAPS configurations.

Verification scores for the 2006 independent test period for forecasting the probability of one or more CG flashes. Brier scores for the model, L-CLIPER, and persistence alone are shown for (top) the 1200 UTC NCEP NAM12, (middle) the 1500 UTC NCEP RUC13, and (bottom) the 1500 UTC WRF–LAPS. The rightmost column shows the percent improvement in Brier score with respect to (left) L-CLIPER and (right) persistence alone.

Verification scores for the 2006 independent test period for forecasting the probability of ≥75th percentile of flash count. Brier scores for the model, L-CLIPER, and persistence alone are shown for (top) the 1200 UTC NCEP NAM12, (middle) the 1500 UTC NCEP RUC13, and (bottom) the 1500 UTC WRF–LAPS. The rightmost column shows the percent improvement in Brier score with respect to (left) L-CLIPER and (right) persistence alone.

Verification scores for the 2006 independent test period for forecasting the probability of one or more CG flashes. This is similar to Table 8, except using relaxed verification criteria. In this case the forecast probability is the maximum probability within a 20-km radius of each grid point.

Verification scores for the 2006 independent test period for forecasting the probability of ≥75th percentile of flash count. This is similar to Table 9, except using relaxed verification criteria. In this case the forecast probability is the maximum probability within a 20-km radius of each grid point.