WHERE ARE THE LIGHTNING

HOTSPOTS ON EARTH?
by R achel I. Albrecht, Steven J. Goodman, Dennis E. Buechler,
Richard J. Blakeslee, and Hugh J. Christian

Earth’s lightning hotspots are revealed with very high-resolution lightning climatology
derived from 16 years of space-based Lightning Imaging Sensor observations.

E
arth’s lightning flash frequency has been an orbiting instruments were used by several authors
object of interest and study for decades. The first to estimate global flash rates [World Meteorological
estimate of the global lightning flash rate was 100 Organization (WMO) 1953; Kotaki 1984; Orville
flashes per second (fl s–1) based on an average number and Henderson 1986; Goodman and Christian 1993;
of lightning flashes per storm and thunderstorm Williams and Heckman 1993; Heckman et al. 1998;
counts recorded by several surface weather stations Mackerras et al. 1998; Boccippio et al. 2000; Williams
(Brooks 1925). In subsequent years, human observers, et al. 2000; Christian et al. 2003; Albrecht et al. 2014].
individual flash counters, regional–continental ground- Ground-based radio frequency measurement systems
based lightning location networks, and finally Earth- provide reliable lightning location but offer limited
spatial coverage, especially over the oceans and the
tropics. These systems detect predominantly cloud-
AFFILIATIONS: A lbrecht—Instituto de Astronomia, Geofísica to-ground lightning relying on land-based receivers
e Ciências Atmosféricas, Universidade de São Paulo, São that are at great distances from where lightning is
Paulo, Brazil, and Cooperative Institute for Climate and Satellites- striking. It was not until the satellite era that consistent
Maryland, University of Maryland, College Park, College Park, monitoring of total lightning (i.e., cloud to ground and
Maryland; Goodman —NOAA/National Environmental Satellite intracloud) was possible at the planetary scale. The first
Data and Information Service, Greenbelt, Maryland; Buechler
lightning observations from space are dated from the
and Christian —University of Alabama in Huntsville, Huntsville,
Alabama; B lakeslee —Marshall Space Flight Center, National
early 1960s using both optical and radio frequency
Aeronautics and Space Administration, Huntsville, Alabama sensors. More than a dozen spacecraft orbiting Earth
CORRESPONDING AUTHOR: Rachel Ifanger Albrecht, have flown instruments that recorded signals from
Universidade de São Paulo, Rua do Matão, 1226, 05508-090 São thunderstorm lightning over the past 45 years. These
Paulo-SP, Brazil instruments generally had low spatial resolution and
E-mail: rachel.albrecht@iag.usp.br low detection efficiency and were unable to provide
The abstract for this article can be found in this issue, following the precise measurements of lightning characteristics
table of contents. (Christian et al. 1992; Goodman and Christian 1993).
DOI:10.1175/BAMS-D-14-00193.1 In 1995, the National Aeronautics and Space
A supplement to this article is available online (10.1175/BAMS-D-14-00193.2) Administration (NASA) launched the Optical
Transient Detector (OTD)—a prototype of the
In final form 2 February 2016
©2016 American Meteorological Society Lightning Imaging Sensor (LIS) instrument—as a
hosted payload on board the Microlab-1 (later renamed

AMERICAN METEOROLOGICAL SOCIETY NOVEMBER 2016 | 2051

Unauthenticated | Downloaded 01/02/23 05:38 PM UTC
 as OrbView-1) small communications satellite. The mountainous regions, regions with frequent extra-
OTD was the first instrument specifically designed to tropical synoptic-scale cyclones, and large-scale
detect lightning from space during both day and night convergence zones [South Atlantic, South Pacific,
with storm-scale resolution (Christian et al. 2003). In and intertropical convergence zone (ITCZ)]. Based
a low-Earth-orbit inclination of 70° and at 740 km of on the smoothed 0.5°-resolution total lightning map
altitude, OTD observed individual storms in a region from OTD, Christian et al. (2003) declared the equa-
1,300 km × 1,300 km for about 3 min during each torial Congo basin as the lightning hotspot of the
satellite overpass between 1995 and 2000. The result planet, more specifically at Kamembe, Rwanda, in
is an annual global flash rate estimate of 44 fl s–1 the top place, and several locations in the Democratic
with a maximum of 55 fl s–1 in the boreal summer Republic of the Congo as well as Nigeria, Gabon,
and a minimum of 35 fl s–1 in the austral summer Madagascar, and Cameroon. In South America, the
(Christian et al. 2003). The LIS on board the NASA principal lightning hotspots were located in northern
Tropical Rainfall Measuring Mission (TRMM) was Argentina extending toward Paraguay and Brazil,
launched in November 1997 into a precessing orbit along one of the regions of the most intense thunder-
inclination of 35° at an altitude of 350 km.1 From storms on Earth (Cecil et al. 2005; Zipser et al. 2006),
this orbit an individual storm within the LIS field of and a few isolated locations in Colombia, Brazil, and
view of 600 km × 600 km was observable for about Peru. In Asia, Kuala Lumpur, Malaysia, stood out
90 s, with no discernible degradation during its time as the place with the most lightning occurrence; in
in orbit (Buechler et al. 2014). The LIS was powered Australia, Fitzroy Crossing stood out, and in North
off and the TRMM mission ended data collection on America, Tampa, Florida stood out. Boccippio et al.
8 April 2015, thus extending what was originally a (2000) also revealed the Congo basin as the region
3-yr mission for a total of 17+ years. with the highest flash rate per thunderstorm cell.
The overall measurement objective of LIS was to A merged global lightning 0.5°-resolution dataset
supply further insights into the nature of tropical composed of individual LIS and OTD orbits has
oceanic and continental convective clouds, providing been updated each year and is freely available online
the basis for the development of a comprehensive (http://lightning.nsstc.nasa.gov/data/data_lis-otd
global thunderstorm and lightning climatology. With -climatology.html; Cecil et al. 2014). This LIS–OTD
this focus, total lightning observed from OTD and climatology is the most accurate depiction of total
LIS have been used to analyze global and regional lightning across the planet to date, named the High
thunderstorm electrical activity (e.g., Boccippio et al. Resolution Annual Climatology (HRAC) database.
2000; Goodman et al. 2000; Williams et al. 2000; Following the methodology of Christian et al. (2003)
Christian et al. 2003; Cecil et al. 2005; Petersen et al. and Boccippio et al. (2000), the HRAC also has spatial
2005; Zipser et al. 2006; Liu et al. 2011; Cecil et al. (2.5° moving average) and temporal (55 days moving
2014). Based on 5 years of OTD and 3 years of LIS average) smoothing and low-pass digital filtering.
measurements, Christian et al. (2003) and Boccippio Cecil et al. (2014) reconfirm the earlier lightning
et al. (2000) presented the first detailed lightning flash rate climatology, with a mean global f lash
climatology maps, which described the frequency rate of 46 fl s–1 (varying from around 35 fl s–1 in
and distribution of lightning over the globe and the austral summer to 60 fl s–1 in boreal summer) with a
tropics, respectively. These authors were pioneers peak annual flash rate of 160 fl km–2 yr–1 over the eastern
in constructing comprehensive total lightning cli- Congo basin.
matologies using a 0.5° resolution for annualized In this paper, we depict a very high-resolution
composite and 2.5° resolution for decomposition total lightning climatology gridded at 0.1° using
of the annual cycle. To overcome limited sampling 16 years (1998–2013) of observations from LIS. We
in places with low flash rates such as oceans, arid used individual LIS orbit data files available from the
regions, and higher latitudes sampled only by OTD, NASA Earth Observing System Data and Information
these climatology maps have substantial spatial System (EOSDIS) Global Hydrology Resource Center
and temporal smoothing as well as low-pass digital (GHRC; http://ghrc.nsstc.nasa.gov/). TRMM started
filtering. Christian et al. (2003) and Boccippio et its descending path to decommissioning in 2014
al. (2000) found that the greatest flash rates occur with several instrument outages during the period,
indeed in the tropical region along coastal areas, which we expect could introduce uncertainties on
the climatology, ranking, and annual and diurnal
1
TRMM was boosted to 402 km of altitude in August 2001 to cycles presented in this study. Therefore, 2014 and
extend the mission lifetime. 2015 data were not included in this study. The focus

2052 | NOVEMBER 2016

Unauthenticated | Downloaded 01/02/23 05:38 PM UTC
of this paper is to identify Earth’s lightning hotspots be clearly observed by intense convection occurring
and other regional features revealed by a detailed more frequently over the continents than the oceans,
very high-resolution satellite-derived climatology. mainly due to the lower heat capacity of land, where
The targets of our analysis are the places where most the ground can warm the air more effectively,
lightning occurs, that is, the lightning hotspots on increasing convective updrafts (Williams and Stanfill
Earth. Therefore, no temporal or spatial smoothing 2002; Williams 2004; Futyan and Del Genio 2007).
is applied to the lightning climatology to best reveal In general, the local standard time (LST) and month
the specific location of lightning hotspots tied to of maximum lightning activity (Figs. 1b and 1c,
orographic features. The methodology to derive respectively) over land occur during the afternoon
the climatology in this study is similar to that used (1200 to 1800 LST) and the hemisphere summer tri-
in the previous satellite depictions (Christian et al. mester (December–February in the Southern Hemi-
2003; Cecil et al. 2014) but at higher resolution and sphere and June–August in Northern Hemisphere).
no spatial smoothing to the full climatology dataset However, some coastal–oceanic regions present
(besides the intrinsic smoothing of gridding data). moderate FRD (up to 30 fl km–2 yr–1) associated with
This climatology is called the LIS very high-resolution frequent synoptic-scale extratropical cyclones and
climatology and is composed of five datasets (http:// cold fronts (such as south-southeast coasts of Brazil,
dx.doi.org/10.5067/LIS/LIS/DATA306): very high- South Africa, and Australia) that occur throughout
resolution full climatology (VHRFC), very high- the year and at all times of the day (i.e., there is a
resolution diurnal climatology (VHRDC), very negligible or nonexistent annual and diurnal cycle).
high-resolution monthly climatology (VHRMC), Moderate lightning activity is also observed within
very high-resolution seasonal climatology (VHRSC), the Gulf of Mexico (with a peak in winter/spring) and
and very high-resolution annual cycle (VHRAC). over the warm Gulf Stream at the Southeastern U.S.
Each dataset has the flash rate density (FRD) and coast (summertime peak). During the winter months,
viewtime (VT) gridded fields, all compiled with LIS the Gulf of Mexico region has twice the number of
data from 1998 to 2013. A full description of these lightning flashes, days, and hours during El Niño
fields and the details on the computation of this years than non–El Niño years (Goodman et al. 2000).
0.1°-resolution climatology, as well as its uncertain- Monsoonal regions such as the Amazon and Indian
ties, can be found in the online supplemental mate- subcontinent have the most lightning activity during
rial (OSM; see online at http://dx.doi.org/10.1175 spring, considered the premonsoon phase (Williams
/BAMS-D-14-00193.2) accompanying this article. It et al. 2002; Kandalgaonkar 2003; Kodama et al. 2005;
is imperative that potential users of this dataset read Petersen et al. 2006; Lal and Pawar 2009; Albrecht
the OSM to better understand the insights of the total et al. 2011b). The Indian subcontinent as a whole
climatology computations. The datasets and figure has a lightning maximum during the premonsoon
maps from this manuscript are publicly available season and a secondary peak in late August to October
and hosted by the GHRC online (at https://ghrc.nsstc (Kandalgaonkar 2005; Lal and Pawar 2009). This second
.nasa.gov/pub/lis/climatology/LIS/). The reader can peak is related to thunderstorms during the postmon-
view detailed close-ups of the features described soon season in the central region (also shown in Fig. 1c).
in the next sections using Google Earth API at Regions where thunderstorm activity is dominated
http://lightning.nsstc.nasa.gov/data/data_lis-vhr by land–sea breeze interactions show high lightning
-climatology.html. activity (30–50 fl km–2 yr–1) over land near the coast
(Fig. 1a), where the maximum thunderstorm activity
TRMM LIS VERY HIGH-RESOLUTION FULL occurs during afternoon (1300–1700 LST; Fig. 1b)
CLIMATOLOGY. The TRMM LIS total lightning and during the summer (Fig. 1c). At night, offshore
FRD (fl km–2 yr–1) climatology map from 16 years wind flow develops thunderstorms over the oceans
(1998–2013) of observations, and in very high hori- near coastal regions with low (6–10 fl km–2 yr–1)
zontal (0.1°) resolution, is shown in Fig. 1. Figure 1 to moderate (10–30 fl km–2 yr–1) lightning activity.
also shows the local time of maximum FRD and the Several pronounced examples of this are found in
month of maximum FRD, based on the VHRDC and the coastal regions of the continents and islands in
VHMC datasets, respectively. As expected, the large- Central America and the Maritime Continent. One
scale features of tropical thunderstorm activity found noticeable example is the Strait of Malacca, a narrow
by Boccippio et al. (2000), Christian et al. (2003), and stretch of water between peninsular Malaysia and
Cecil et al. (2014) are still captured here at 0.1° reso- Sumatra Island (Indonesia). During the day, onshore
lution. The difference between land and ocean can breezes foment convergence and thunderstorm

AMERICAN METEOROLOGICAL SOCIETY NOVEMBER 2016 | 2053

Unauthenticated | Downloaded 01/02/23 05:38 PM UTC
 2054 |
NOVEMBER 2016
Fig . 1. Very high horizontal (0.1°) resolution total lightning climatology from 16 years (1998–2013) of TRMM LIS total lightning observations: (a) FRD
(fl km−2 yr −1), (b) local hour of maximum FRD, and (c) month of maximum FRD. Data shown for the latitude band of ±38°.

Unauthenticated | Downloaded 01/02/23 05:38 PM UTC

development over the island, while during the night, complete list of the first 500 hotspots are shown in the
the offshore breeze blows from both landmasses online supplemental material. In Table 1, we show the
toward the Strait of Malacca, converging over warm top 10 lightning hotspots for each major continental
seawater and driving nocturnal thunderstorms (more landmass covered by LIS (North America, South
details in the next section). The presence of enhanced America, Africa, Asia, and Oceania). Figure 2 shows
nocturnal ice scattering from clouds at the Strait of the geographical distribution of the top 10 lightning
Malacca has been reported by Nesbitt and Zipser hotspots of each major tropical continental landmass
(2003), and Virts et al. (2013) reports high cloud- as well as a zoomed plot of the total lightning climatol-
to-ground lightning activity during the night in this ogy (refer to Fig. 1) at each first continental landmass
same region. Nocturnal lightning is also observed at hotspot.2 Note that LIS measurements are limited to
regions with complex terrain, such as in the Andes approximately ±38°, such that continents are not ob-
(especially in Colombia) and the Himalayas (Virts served in their entirety. The rankings are based on the
et al. 2013). Moreover, nocturnal and early morning portions of the continents observed by LIS.
moderate-to-high lightning activity is found in Not surprisingly, Africa is represented by 283 of
the central-southern U.S. Great Plains, northern the top 500 places with the highest lightning fre-
Argentina, Southeast Asia, and West Africa. These quency, followed by Asia with 87, South America
regions produce large mesoscale convective systems with 67, North America with 53, and Oceania with 10
(MCS) predominantly during spring time (Fig. 1c) (Table ES3 of OSM). However, one surprising discov-
with maximum convective activity during the night ery is that Earth’s top-ranked total lightning hotspot
(Fig. 1b) (Laing and Fritsch 1997; Kodama et al. 2005; is not in Africa as reported previously but instead
Rasmussen and Houze 2011; Rasmussen et al. 2014). directly over Lake Maracaibo in Venezuela (9.75°,
Places with ver y high FRD (higher than –71.65°) where 233 fl km–2 yr–1 occur, leaving the
50 fl km –2 yr–1) are extensively located over the previously identified Congo basin hotspot in Africa
Congo basin; the coastal zones of Cuba, Saudi Arabia, (Christian et al. 2003; Goodman et al. 2007; Cecil et al.
and Yemen; and near several mountain regions, such 2014) (more details in the OSM). in second place with
as the Andes, Sierra Madre Mountains (at the west 205 fl km–2 yr–1. The closest populated place to this
coast from northern Mexico to Guatemala), western hotspot over Lake Maracaibo is Lagunillas in the state
highlands of the Cameroon line, Mitumba Mountains, of Zulia on the east side of the lake, about 60 km away.
Himalayas, and mountain ranges in Indonesia and Figure 2 shows that Earth’s top-ranked maximum over
Papua New Guinea. It is over these complex terrain Lake Maracaibo is very localized, covering an area
regions that most of Earth’s lightning hotspots are of approximately 5,500 km2 with FRD greater than
located. These hotspots are ranked and identified in 50 fl km–2 yr–1. This corresponds to only 45 pixels in
the following section. 0.1° resolution and less than 2 pixels in 0.50° resolu-
tion and explains why it was not similarly identified in
LI G HTN I N G F L A S H R ATE D E N S IT Y the previous lower-resolution (0.5°) LIS climatologies
R A N K I N G A N D C O N T I N E N TA L of Christian et al. (2003) and Cecil et al. (2014; more
LANDMASSES’ HOTSPOTS. The individual details in the OSM). The diurnal and annual cycles
0.1° total lightning flash rate density grid boxes from of lightning activity over Lake Maracaibo and other
Fig. 1 were ranked from highest to lowest values landmass hotspots are shown in Fig. 3. Lightning activ-
and associated with a populated place name with ity over Lake Maracaibo presents two peaks during the
more than 1,000 inhabitants from the geographi- year, where most of the lightning occurs from August
cal database GeoNames (www.geonames.org/). To to November (>65 fl day–1), drastically decreasing in
minimize an excessive number of identical place December with almost no lightning in January and
names within clusters of very high flash rate density February. The second peak is an approximately flat
(>50 fl km–2 yr–1), such as Africa and the northern flash rate (47 fl day–1) in May and June and is related
region of South America, a criterion of a minimum to the migration of the ITCZ (Pulwarty et al. 1998).
distance of 100 km between maxima is applied. The strong diurnal cycle of lightning frequency reveals
Therefore, the identified lightning hotspots should almost no lightning during the day and a nocturnal
be interpreted as a cluster of hotspots around a refer-
ence location. Also, it may exclude popular and well- Note that all 500 hotspots can be navigated (with zoom capa-
2

known locations as it chooses the closest populated bility) using Google Earth API at the LIS Very High Resolu-
place and not the place with the most inhabitants. tion Climatology web page hosted by NASA GHRC at http://
More details on the ranking methodology and the lightning.nsstc.nasa.gov/data/data_lis-vhr-climatology.html.

Unauthenticated | Downloaded 01/02/23 05:38 PM UTC

 Table 1. Top 10 FRD (fl km –2 yr –1) for each continental landmass, indicating its position in the global
ranking, latitude (Lat) and longitude (Lon) position on TRMM LIS 0.1° climatology grid, as well as the
name of the nearest populated place name (PPL), position (PPL lat, PPL lon), and distance from grid point
(Dist), according to the GeoNames database (www.geonames.org/). A minimum distance of 100 km from a
previous ranked grid point was applied.

Global Lat Lon PPL lat PPL lon Dist

rank FRD (o) (o) PPL Country (o) (o) (km)
South America
1 232.52 9.75 –71.65 Lake Maracaibo 10.13 –71.26 60.1
Venezuela
(Lagunillas)
4 172.29 7.55 –75.35 Cáceres Colombia 7.58 –75.35 3.4
7 138.61 8.85 –73.05 El Tarra Colombia 8.58 –73.09 30.9
11 124.26 5.75 –74.95 Norcasia Colombia 5.58 –74.89 20.4
18 114.19 8.45 –74.55 Majagual Colombia 8.54 –74.62 12.6
25 105.73 8.15 –76.85 Turbo Colombia 8.09 –76.73 14.8
46 95.38 11.15 –72.95 Barrancas Colombia 10.96 –72.79 27.8
74 87.96 –17.25 –65.05 Chimoré Bolivia –16.95 –65.14 34.9
78 87.61 10.35 –70.95 El Corozo Venezuela 10.12 –71.04 27.5
136 77.02 10.45 –75.35 Santa Rosa Colombia 10.44 –75.37 2.2
Africa
2 205.31 –1.85 27.75 Democratic Republic –2.50 28.79 136.2
Kabare
of the Congo
3 176.71 –3.05 27.65 Democratic Republic –3.60 26.67 124.9
Kampene
of the Congo
5 143.21 –0.95 27.95 Democratic Republic –1.57 29.04 140.0
Sake
of the Congo
8 129.58 5.25 9.35 Nguti Cameroon 5.33 9.42 11.7
9 129.50 0.25 28.45 Democratic Republic 0.14 29.29 94.3
Butembo
of the Congo
10 127.52 –1.55 20.95 Democratic Republic –0.28 20.88 141.2
Boende
of the Congo
14 117.98 0.55 20.35 Democratic Republic –0.28 20.88 109.7
Boende
of the Congo
15 117.19 –2.45 26.95 Democratic Republic –2.94 25.92 126.7
Kindu
of the Congo
16 116.78 6.95 10.45 Baissa Nigeria 7.23 10.63 36.6
19 112.17 0.35 26.65 Democratic Republic 0.52 25.19 163.3
Kisangani
of the Congo
Asia
6 143.11 34.45 72.35 Daggar Pakistan 34.51 72.48 14.0
12 121.41 33.35 74.55 Rājauri India 33.38 74.31 22.6
13 118.81 33.75 70.75 Doāba Pakistan 33.42 70.74 36.2
22 108.03 14.55 43.45 Al Ḩadīyah Yemen 14.53 43.57 13.2
28 104.59 33.85 73.25 Murree Pakistan 33.91 73.39 14.5
31 101.79 25.25 91.95 Cherrapunji India 25.30 91.70 26.1
42 97.02 4.75 103.05 Paka Malaysia 4.64 103.44 44.7
45 95.92 1.95 103.85 Kota Tinggi Malaysia 1.74 103.90 24.2
50 94.64 3.75 98.05 Tenggulun Indonesia 3.99 98.01 27.3
52 93.96 3.15 101.65 Kuala Lumpur Malaysia 3.14 101.69 4.2

2056 | NOVEMBER 2016

Unauthenticated | Downloaded 01/02/23 05:38 PM UTC
Table 1. Continued.

Global Lon PPL lat PPL lon Dist

rank FRD Lat (o) (o) PPL Country (o) (o) (km)
North America
17 116.76 14.35 –91.15 Patulul Guatemala 14.42 –91.17 7.6
29 103.23 14.85 –92.05 Catarina Guatemala 14.85 –92.08 2.8
33 100.63 22.35 –83.95 San Luis Cuba 22.29 –83.77 20.1
34 100.24 18.55 –74.35 Chambellan Haiti 18.57 –74.32 4.0
37 99.39 13.15 –87.25 San Jerónimo Honduras 13.18 –87.14 12.7
39 98.22 22.35 –80.65 Rodas Cuba 22.34 –80.56 9.8
40 98.06 21.75 –78.85 Venezuela Cuba 21.74 –78.80 5.8
47 95.32 22.85 –82.15 Mañalich Cuba 22.81 –82.15 4.3
82 86.96 22.25 –105.25 Rosamorada Mexico 22.12 –105.21 14.9
90 85.78 18.15 –77.65 Balaclava Jamaica 18.17 –77.64 2.6
Oceania
61 92.15 –15.35 125.35 Derby Australia –17.30 123.63 284.4
83 86.75 –14.45 126.55 Kununurra Australia –15.78 128.74 278.0
228 65.11 –16.65 124.75 Derby Australia –17.30 123.63 139.6
308 59.69 –15.65 128.45 Kununurra Australia –15.78 128.74 34.6
316 59.19 –4.75 142.95 Ambunti Papua New Guinea –4.22 142.82 61.0
327 58.57 –15.25 129.45 Kununurra Australia –15.78 128.74 95.8
355 57.13 –13.15 131.05 McMinns Lagoon Australia –12.55 131.11 66.6
381 55.57 –13.95 129.95 Darwin Australia –12.46 130.84 191.6
471 51.35 –19.15 137.85 Mount Isa Australia –20.73 139.50 245.6
477 51.22 –17.45 126.05 Halls Creek Australia –18.22 127.67 191.6

maximum (>2.5 fl h–1) from 0000 to 0500 LST abruptly and mountain–valley breezes) that provide low-level
peaking (5.4 fl h–1) at 0300 LST (Fig. 3a). convergence inside the valleys and offshore (Gulf
of Panama) as well as convergence of gravity waves
South America. Of South America’s top 10 hotspots from diurnally varying heat sources (differential
(including Earth’s top hotspot at Lake Maracaibo), heating between land, lake, mountain, and valleys;
7 are located near or at the valleys or foothills of the Mapes et al. 2003a). Lightning activity in the Bolivian
northern Andes Mountains in Colombia, Venezuela, hotspot (Chimoré) is from austral spring to summer
and Bolivia (Fig. 2). All these hotspots, except the one (>5 fl day–1, peaking in November, 30 fl day–1; Fig. 3b)
in Bolivia, have similar annual and diurnal cycles of without a well-defined diurnal cycle (0.5–1.0 fl h–1
lightning activity (Figs. 1c, 3a), with most lightning from 1300 to 0400 LST; Fig. 3a). This hotspot is also
activity (>15 fl day–1) from austral fall [March–May at the eastern side of the Andes’ foothills, at the end
(MAM)] to austral spring [September–November (SON)] of the northwest–southeast-orientated range, inside
with a peak in August or September (37–88 fl day–1) Carrasco National Park. In this configuration, the
and most lightning occurring during the night mountains are a natural barrier to the moist South
(>2 fl h–1). Nocturnal maxima of deep convection were American Low-Level Jet winds from the Amazon
also observed by other authors over the valleys along basin (Vera et al. 2006), forcing it uphill and favoring
the most northern portion of the Andes Mountains, thunderstorm development during spring and
including Lake Maracaibo (Garreaud and Wallace summer. Inferred peaks of hailstorm frequency, based
1997; Negri et al. 2000; Mapes et al. 2003a; Poveda on passive microwave observations from satellite,
et al. 2005; Bürgesser et al. 2012; Virts et al. 2013). were also found by Cecil and Blankenship (2012)
These authors attributed the nocturnal convection to at this same spot during spring and summer. Four
local thermally driven circulations (land–sea breezes hotspots (sixth, seventh, ninth, and tenth) are located

AMERICAN METEOROLOGICAL SOCIETY NOVEMBER 2016 | 2057

Unauthenticated | Downloaded 01/02/23 05:38 PM UTC
Fig. 2. (left) Location (numbered symbols) of the top 10 FRD hotspots per major continental landmasses (Africa,
Asia, North America, Oceania, and South America), and (right) 2.5° zoom boxes on FRD climatology of each
landmass first hotspot (Table 1), using the same FRD color coding as in Fig. 1a. Numbered symbols indicate the
continent landmass ranking order (first to tenth) from Table 1, and colors indicate the continent (purple indicates
Africa, magenta indicates Asia, blue indicates North America, light blue indicates Oceania, and red indicates
South America). Shaded color scale in the left and central figures is terrain elevation.

2058 | NOVEMBER 2016

Unauthenticated | Downloaded 01/02/23 05:38 PM UTC
near the Colombia and Venezuela coasts, presenting cycle and sea breeze near elevated terrain (Fig. 2;
the same annual cycle as most of the top 10 South South America zoom at Lake Maracaibo).
American hotspots. However, those hotspots farther
to the west (seventh and ninth) have afternoon peaks Africa. Earth’s second greatest lightning hotspot
(>2 fl h–1) at 1700 LST due to a well-defined diurnal is at Kahuzi-Biéga National Park (–1.85°, 27.75°),

Fig. 3. (a) Hourly and (b) daily flash rate density distribution of the top 10 lightning hotspots of each major
continental landmass from Table 1. These values are calculated over a 1° box centered at the hotspot.

AMERICAN METEOROLOGICAL SOCIETY NOVEMBER 2016 | 2059

Unauthenticated | Downloaded 01/02/23 05:38 PM UTC
 136.2 km northwest of Kabare in the Democratic (Nicholson 1996; Jackson et al. 2009). This is reflected
Republic of the Congo near the border of Rwanda in the lightning maximum ranking (Table 1), where 8
(on the west side of the Mitumba Mountains; Fig. 2). of the 10 most active lightning places in Africa occur
The Mitumba Mountains at the eastern edge of the within the Democratic Republic of the Congo, all
Congo basin mark the beginning of the Western Rift having FRD greater than 110 fl km–2 yr–1. The two
Valley in East Africa, along a north–south orienta- African hotspots at the Cameroon border (Nguti,
tion at ~29°E and with peaks ranging from 2,000 to Cameroon, and Baissa, Nigeria) have their diurnal
5,200 m. Total lightning flash rate densities greater cycle influenced by local afternoon convection driven
than 50 fl km–2 yr–1 are observed along a continuous by sea-breeze and moisture flux convergence from the
large area (~320,000 km2) on the western foothills of Gulf of Guinea (Nicholson 2009).
these mountains from 4°S to 5°N, having 6 out of the
10 African hotspots. The FRD decreases westward Asia. High to very high FRD (>30 fl km–2 yr–1) can be
of 25°E to a narrow region of 40–50 fl km–2 yr–1 and found at the high elevation envelope along the length
increases again to over 50 fl km–2 yr–1 (Fig. 1), having of the Himalayas in Asia. Asia’s top-ranked lightning
two other African hotspots. All the hotspots over the hotspot, Daggar, Pakistan (34.45°, 72.35°), is located in
Mitumba Mountains exhibit higher mean diurnal the westernmost portion of the Himalayas foothills at
cycle flash rates (Fig. 3b) during the afternoon (peak the Indus’ Plain, more specifically at the Hindu Kush
of 5 fl h–1 from 1500 to 1700 LST) than in the central range in the northernmost region of Pakistan (Fig. 2).
Congo (~3 fl h–1 at 1500 LST), with some activity Three other Asian hotspots (the second, third, and
during the night (~1 fl h–1) probably associated with fifth highest) are also located in this region, whereas
the maximum extent of MCSs (Laing and Fritsch the sixth Asian hotspot is at the most eastern portion
1993; Jackson et al. 2009). of Himalayas, at the Kashi Hills (on the border between
Thunderstorms occur year-round in central and India and Bangladesh). The western Himalayas were
western Africa, but these regions are electrically more also identified as a region with the most intense
active from September to May, with flash rates over thunderstorms on Earth by Cecil et al. (2005), Zipser
15 fl day–1 occurring on most days, with less light- et al. (2006), and Houze et al. (2007). Houze et al.
ning activity observed in July (Fig. 3b). The spatial (2007) found that the deepest intense convective storms
and temporal distributions of African rainfall, and occur upwind or at the foothills of the mountains,
therefore also lightning activity, are tied to a com- where the moist southwesterly monsoon flow from
bination of large- and local-scale factors, such as the the Arabian Sea meets the descending dry air from the
seasonal migration of the ITCZ, the position of the Afghan or Tibetan Plateaus, suggesting a similarity to
African easterly jets (which migrate seasonally with dryline convection. All these hotspots have more light-
the ITCZ and are evident from August to November), ning activity (>15 fl day–1) during the monsoon months
long nocturnal life cycle of MCSs, afternoon bound- (May–October, peaking in September; Fig. 3) and less
ary layer destabilization, local effects of sea breezes from November to March, with the most lightning
and topography, as well as the sea surface tempera- activity extending from late afternoon (1600 LST) to
tures (Nicholson 1996; Yin and Nicholson 2000; early morning (0600 LST), with the exception of the
Nicholson and Grist 2003; Balas and Nicholson 2007; second- and third-ranked hotspots. The nocturnal
Jackson et al. 2009; Nicholson 2009). Particularly, lightning is controlled by a few MCSs and local con-
from August to November, the African easterly jet of vection induced by the interaction between the large-
the Northern Hemisphere (AEJ-N) and the African scale synoptic flow and the complex terrain (Barros
easterly jet of the Southern Hemisphere (AEJ-S) are and Lang 2003; Barros et al. 2004; Zipser et al. 2006;
well developed and located, respectively, above and Houze et al. 2007). Barros et al. (2004) found that two
below the edges of the Mitumba mountain range. In types of orographic terrain control cloudiness in the
addition, there is a strong low-level moisture conver- Himalayan range: the first type is associated with the
gence on its lee side, coming from the Atlantic and major river valleys (~300 km wide) and the overall
central Congo rain forest (Jackson et al. 2009). This mountain range that connects India and the Tibetan
mesoscale configuration results in a convergence Plateau, while the second type is associated with the
region with strong vertical motions on the right succession of ridges and small valleys (5–150 km wide).
entrance of the AEJ-S, which, along with the deep In fact, Fig. 1a shows a long, nearly contiguous line
low-level moisture source, promotes an explosive and a broken line of FRD greater than 30 km–2 yr–1
convection scenario for thunderstorm and lightning within the Himalayan river valleys and the narrow
development over the whole of the Congo basin valleys, respectively, in northern India and Nepal, with

2060 | NOVEMBER 2016

Unauthenticated | Downloaded 01/02/23 05:38 PM UTC
an indistinct local hour of
maximum lightning activi-
ty (Fig. 1b) at night and early
morning. At high elevations
over the Tibetan Plateau, the
local hour of maximum is
during the day.
The remaining Asian
hotspots, except for the one
in Yemen, lie over the coasts
of Malaysia and Sumatra
Island (Indonesia) and
are caused by a sea breeze
resulting in the afternoon
(peak at 1500 LST; Fig. 3a) Fig. 4. Daytime (0700–1800 LST) and nighttime (1900–0600) flash rate density
(fl km−2 yr −1) at the Strait of Malacca, Indonesia. White lines are elevation
in FRD. Figure 4 shows
from 500 to 3,000 m in 500-m intervals, and gray lines are country physical
day time and nighttime boundaries.
FRD over Malaysia and
Sumatra Island. High FRD
greater than 50 fl km–2 yr–1 is found inland, alongside Court and Griffiths (1982). These authors compiled
the coast and high elevated terrain. During the night, the number of thunderstorm days recorded by surface
convergent offshore breezes from the peninsular weather stations around the globe from 1951 to 1975
Malaysia and the Sumatra Island over the Strait of and found that Fort Meyers, Florida, is the place
Malacca leads to increased FRD over the ocean after with the most thunderstorm days per year (96) in the
midnight (Fig. 1b), also seen in cloud-to-ground United States, with a peak in July.
lighting activity (Virts et al. 2013). The annual cycle
shows two peaks during the year associated with the Oceania. Oceania hotspots all are found at the
ITCZ migration from the Northern to the Southern northern coast of Australia, except the fifth most
Hemispheres (Fig. 3b). The hotspot at Yemen [Al active locale that is found in Papua New Guinea.
ῌadīyah (14.55°, 43.45°)] also presents a maximum Most of the FRD across Australia and Papua New
of activity in the afternoon because of the sea breeze, Guinea is less than 10 fl km–2 yr–1 and is attribut-
but its annual cycle shows lesser lightning activity able to local convection in the afternoon, mostly
(<30 fl day–1) peaking in August. during austral summer monsoon with two peaks
(December and mid-January to March; Fig. 3b;
North America. In North America, the first and second Jayaratne and Kuleshov 2006; Kuleshov et al. 2006),
hotspots with FRD > 100 fl km–2 yr–1 (respectively, in contrast to the other two monsoonal regimes
Patalul and Catarina, in Guatemala) as well as the of the Amazon and India that exhibit lightning
ninth hotspot (Rosamorada, Mexico) are also at the peaks during premonsoon. The diurnal cycle of the
foothills of a mountain range, the Sierra Madre, in a hotspot at Ambunti, Papua New Guinea, also exhibits
very narrow low-level strip of land between the Pacific lightning maximum during the afternoon (Fig. 3a),
Ocean and the mountains. These hotspots show an but its annual cycle shows much lower FRD than
afternoon lightning peak with most of the lightning Australia, peaking 2 months earlier (October) than
activity occurring during boreal spring and summer. Australia hotspots. It is worthwhile to note that North
All remaining North American hotspots are over the America’s and Oceania’s top hotspots have an FRD
Central America islands (Cuba, Honduras, Haiti, and that is barely on the order of Africa’s tenth-ranked
Jamaica) that also exhibit afternoon lightning activity hotspot (~100 fl km–2 yr–1). This reflects the greater
occurring mostly during spring and summer. occurrence of convection and lightning activity in
The top U.S. hotspot with an FRD of 79 fl km–2 yr–1 tropical belts of Africa, South America, and Asia.
lies over the Everglades near Orangetree, Florida,
about 37 km from Fort Meyers. The top U.S. hotspot is EARTH’S TOP LIGHTNING HOTSPOT.
ranked only 14th in North America (122nd globally). Nocturnal thunderstorms over Lake Maracaibo are
This same location coincides with that found in the so frequent that their lightning activity was used as a
first thunderstorm day climatology constructed by lighthouse by Caribbean navigators in colonial times

AMERICAN METEOROLOGICAL SOCIETY NOVEMBER 2016 | 2061

Unauthenticated | Downloaded 01/02/23 05:38 PM UTC
 (Codazzi 1841), and they are also mentioned in the La Andes are branched out, defining several large valleys
Dragontea (1598) poem, from the Spanish poet Felix over Colombia, and form a large valley in which lies
Lope De Vega, as what prevented an English pirate Lake Maracaibo. The lake is connected to the Gulf of
from attacking Maracaibo. These thunderstorms are Venezuela at its northern end, near Maracaibo city,
locally known as the “Lighthouse of Catatumbo,” and the very warm water temperatures over the lake
the “Never-Ending Storm of Catatumbo,” or simply and at the gulf throughout the year (28°–31°C), with
Catatumbo lightning. Catatumbo is a river that ends its maximum in September (Muñoz et al. 2008), serve
southwest of Maracaibo Lake, and nowadays tour- as a great source of heat and moisture for convection.
ists can take nighttime boat tours to observe these All these geographical features combined with
beautiful storms. convergent wind flow (mountain–valley, lake, and
Lake Maracaibo has a unique set of features that land–sea breezes) over this warm, humid lake within
contributes to the development of deep convection at the ITCZ make an ideal spot for thunderstorm devel-
the same place 297 days per year on average (Fig. 3b) opment. During the day, radiative heating by the sun
with a maximum at night (Fig. 3a). This lake is the causes the land surrounding the lake to become hotter
largest in South America, with an area of 13,210 km2, than the lake, inducing a lake-breeze circulation with
and is located between the most northern ridges of the subsidence over Lake Maracaibo. In addition, the
Andes Mountains (Fig. 5). At its northern portion, the sides of the mountains warm up more rapidly than

Fig. 5. (middle) Northwestern Venezuela elevation map as well as (left) daytime and (right) nighttime wind
roses from three locations around Lake Maracaibo: Zulia, Mene Grande, and Maracaibo. Elevation color scale
is the same as in Fig. 2.

2062 | NOVEMBER 2016

Unauthenticated | Downloaded 01/02/23 05:38 PM UTC
the valley, thereby inducing
a valley breeze accompa-
nied by subsidence over
the lake, inhibiting con-
vection. This can be seen
in the diurnal distribution
of FRD over this region in
Fig. 6a. Lightning activ-
ity is only observed over a
small area of the lake dur-
ing the late afternoon. Most
of the daytime lightning
occurs near the coast and
is driven by a sea-breeze
circulation that is forced
uphill to higher elevations.
On the other hand, during
the night, radiative cooling
causes the land to become
cooler than the lake with
the surrounding mountain-
sides cooling more rapidly
than the valley, yielding a
low-level wind convergence
over the hot and humid en-
vironment of the lake.
Figure 5 shows wind
Fig. 6. As in Fig. 4, but for (top) Lake Maracaibo, Venezuela, and (bottom)
roses (frequency of wind
Lake Victoria, Africa.
speed by wind direction)
of surface weather station
data during daytime (0700–1800 LST) and nighttime a maximum at the central Caribbean Sea between
(1900–0600 LST) at three locations around Lake 12° and 16°N and 71° and 76°W. This low-level jet
Maracaibo (Zulia, Mene Grande, and Maracaibo) increases vertical shear that reduces convection in
for the period 1998–2014. This dataset is freely avail- June–July and December–March (Amador 1998;
able online through NOAA/National Climatic Data Wang 2007; Amador 2008; Muñoz et al. 2008, 2016),
Center’s Integrated Surface Database (ISD; www.ncdc coinciding with the two lightning minima observed
.noaa.gov/isd). ISD consists of surface observations over Lake Maracaibo (Fig. 3b). The weakening of CLLJ
from over 35,000 stations worldwide, compiled from in September–November and May (as a response from
numerous sources into a single common data format the smaller gradient in sea level pressure and sea
(Lott et al. 2008). It can be seen that, in general, the surface temperature in the Caribbean Sea) favors
wind indeed blows toward the lake during the night convection and is responsible for the rainy season
and away from the lake during the day. This results over northern South America and southern Central
in a nighttime convergent land–lake breeze and a America (Mapes et al. 2003b; Muñoz et al. 2008,
daytime divergent lake–land breeze over the lake. This 2016). This decrease in vertical shear increases air
is a perfect combination for nocturnal thunderstorm mass thunderstorm development in Lake Maracaibo,
development and very high flash rate occurrence, contributing to the frequent lightning activity in
where FRD above 50 fl km–2 yr–1 covers most of the lake August–October and April–May.
area, the southwestern portion of the valley and also The persistent low-level convergence over the
several valleys in Colombia during nighttime (Fig. 6a). warm waters of Lake Maracaibo almost all year long
The two flash rate peaks from August to October is a perfect scenario for development of thunder-
and May to July coincide with the weakening of the storms and lightning production that contributes to
Caribbean low-level jet (CLLJ). The CLLJ is char- the observed high FRD. Nocturnal convection and
acterized by a predominantly easterly wind with lightning activity over Lake Maracaibo have been

AMERICAN METEOROLOGICAL SOCIETY NOVEMBER 2016 | 2063

Unauthenticated | Downloaded 01/02/23 05:38 PM UTC
 reported in the literature by previous authors (Mapes circulations (such as mountain–valley and land–lake
et al. 2003b; Martinez et al. 2003; Albrecht et al. 2011a; breezes) and nocturnal development of MCSs. The
Falcon 2011; Bürgesser et al. 2012). Several other month of maximum flash rate density shows most
inland lakes with similar conditions, that is, deep noc- lightning occurrence during summer months, except
turnal convection driven by locally forced convergent for strong monsoonal regions, as with India and the
flow over a warm lake surface, were identified in the Amazon where lightning maximum activity occurs
LIS climatology. Lake Victoria, in Africa, is one of during springtime (premonsoon months) and in the
them. It is the largest tropical lake in the world, with case of India also in the postmonsoon months.
an area of approximately 68,800 km2. This lake is Finally, we identify Earth’s lightning hotspots by
located east of the Mitumba Mountains, in between ranking the climatological flash rate density, as well
the two branches of Rift Valley, and is divided among as the top 10 hotspots of each continental landmass,
Tanzania, Uganda, and Kenya. As with Lake Maracaibo, to highlight details of their annual and diurnal cycle
Lake Victoria lightning activity is more pronounced of lightning activity. Earth’s top-ranked lightning
during the night. Maximum convection and rainfall hotspot occurs over Lake Maracaibo in Venezuela
occurring during the night to early morning has long where lightning occurs 297 days per year (with a
been recognized by studies dating from the 1960s. This peak in September) due to convergent land breeze
nighttime maximum has been attributed to convergent over this hot lake. Lake Victoria (as well as other
nocturnal land breezes (e.g., Flohn and Fraedrich lakes along the East African Rift Valley) exhibits
1966; Ba and Nicholson 1998; Yin and Nicholson 2000; deep nocturnal convective activity driven by locally
Nicholson and Yin 2002). Ba and Nicholson (1998) forced land breezes. The second-ranked lightning
found that maximum convection occurs during the hotspot on Earth is at Kahuzi-Biéga National Park,
morning (0500–0800 LST) with a second peak in the near the city of Kabare in Democratic Republic
afternoon, while over the surrounding land, maximum of the Congo on the west side of the Mitumba
convection is observed in the afternoon and evening. Mountains. Several African lightning maxima are
These findings are corroborated by the daytime and located along this mountain range, with strong
nighttime FRD over Lake Victoria (shown in Fig. 6b). afternoon lightning activity all year long with peaks
According to Flohn and Fraedrich (1966) and Nicholson in March and September. Asia’s lightning hotspot is
and Yin (2002), because of prevailing easterly low-level at the northwestern ridges of the Himalayas, near
winds, convection generally moves from the eastern Daggar, Pakistan. Its annual and diurnal cycles show
surrounding land in the late afternoon toward the maximum activity, respectively, in the summer and
northeastern portion and center of the lake during late afternoon to early evening. North America and
the night and then toward the western portion of the Oceania’s top lightning hotspots are, respectively,
lake during late night and early afternoon. In addition, in Patulul, Guatemala, and Derby, Australia, with
other African Rift Valley lakes such as Tanganyika, afternoon and summer maximums. In the United
Malawi, and Albert show nocturnal convergent land States, the lightning hotspot is in the Everglades, near
breeze (Nicholson and Yin 2002) with the LIS light- Orangetree, Florida, which is ranked in 14th position
ning activity peaks during the late night and early in North America and 122nd globally.
morning (Fig. 1b). Court and Griffiths (1982) found that the place
in the world with more thunderstorms per day
SUMMARY AND CONCLUSIONS. The 16 was in Uganda (242), followed by the Democratic
years of TRMM LIS lightning measurements are used Republic of the Congo (formerly Zaire; 228) and
to construct a very high-resolution (0.1°) total (cloud Rwanda (221). These last two locations are within
to ground and intracloud) lightning climatology. The 200 km from Kabare, Democratic Republic of the
derived lightning flash rate density maps reconfirm Congo, the second hotspot of Earth found in the
the previous general observations of continental deep current study. Unfortunately, they did not have data
convection over land and reduced lightning activity from Venezuela. Court and Griffiths (1982, p. 33)
over the ocean with greater detail on mountainous also observed that most thunderstorm days were
regions such as the Andes, Himalayas, Sierra Madre, associated with complex terrain and coastal lines
and Mitumba Mountains in Africa, all of which are and stated, “There is a relation, although not perfect,
marked by the frequent occurrence of thunderstorms at with topography, land-sea configuration, airmass
their foothills. The diurnal cycle over most continental movements and airflow on all scales.” In fact, from
regions presents the lightning maximum during the the first 30 lightning hotspots only 6 are not near
afternoon, with exceptions driven by locally forced mountainous regions. For example, the strongest

2064 | NOVEMBER 2016

Unauthenticated | Downloaded 01/02/23 05:38 PM UTC
hotspots from each continental landmass are near or NOAA geostationary satellites, the GOES-R series,
related to mountainous regions: Lake Maracaibo in will be launched with the first Geostationary Light-
South America, Kabare in Africa, Daggar (Pakistan) ning Mapper (GLM) that will provide unprecedented
in Asia, and Patulul (Guatemala) in North America, continuous day and night total lightning observa-
with the exception being that Oceania’s top hotspot tions from geostationary orbit and extend the climate
occurs in Derby (Australia) within a flat coast region. dataset begun with OTD and TRMM LIS for another
These features are so localized that they are only rec- 20–30 years. Therefore, the present study provides
ognized when looking at the fine scales of the interac- insight and context to forecasters and researchers who
tion between the wind and complex terrain that leads will use total lightning activity to better understand
to cloud development. Nesbitt and Anders (2009) and the earth around us.
Biasutti et al. (2012) explored TRMM Precipitation
Radar climatologies at very high resolutions (0.1° ACKNOWLEDGMENTS. We would like to acknowl-
and 0.05°, respectively) to investigate the role of edge the support of the NASA Earth Science Division and
topography and coastlines affecting tropical rainfall the NOAA GOES-R Science Office in addition to the work
distributions. Nesbitt and Anders (2009) found high of all LIS Science Team members. All past and current
precipitation accumulations over narrow bands from members have contributed in some degree to the success
500 to 2,000 m in elevation at the eastward foothills of this mission. A special thanks is offered to Dennis
of the Andes as well as a distinctly drier region along Boccippio, who developed the source code for OTD and LIS
high (>2,000 m) slopes of the northern portion and gridded climatologies, which were modified here to obtain
Altiplano to the south. Here, lightning activity in the very high-resolution grid (0.1°) results. We also thank
very high terrain is also very low (<2 fl km–2 yr–1) the editor, Edward Zipser, and three anonymous reviewers
or absent, distinctly marking the Andes mountain for their helpful comments and suggestions. Douglas Mach
ranges along the South American west coast, the (USRA), thank you for the discussions on LIS algorithm
Sierra Madre in Mexico, and the Himalayas in Asia. and flash length. LIS data used in this work and the gridded
One noticeable exception is the northern range climatologies produced here are distributed by the NASA
over Colombia and Venezuela. Whereas most of the EOSDIS Global Hydrology Resource Center Data Analysis
accumulated precipitation is found west of the Andes and Archive Center (DAAC), Huntsville, Alabama, (http://
toward the ocean (Mapes et al. 2003b; Nesbitt and thunder.nsstc.nasa.gov). The views, opinions, and findings
Anders 2009; Biasutti et al. 2012), lightning activity contained in this report are those of the authors.
is concentrated within the several valleys between
the mountain ranges, marking two distinct regimes
of rain yield per flash: high precipitation rates and REFERENCES
moderate lightning activity toward the ocean (high Albrecht, R. I., S. J. Goodman, W. A. Petersen, D. E.
rain yield per flash) and high precipitation rates and Buechler, E. C. Bruning, R. J. Blakeslee, and H. J.
high FRD along the valleys (low rain yield per flash; Christian, 2011a: The 13 years of TRMM Lightning
Takayabu 2006). Imaging Sensor: From individual flash characteris-
After over 17 years of operational service, the tics to decadal tendencies. Extended Abstracts, 14th
TRMM mission came to a close as the satellite used Int. Conf. on Atmospheric Electricity, Rio de Janeiro,
up its fuel and began a slow orbital descent that began Brazil, ICAE, 1–4. [Available online at http://ntrs.nasa
in October 2014 and ended when the LIS instrument .gov/archive/nasa/casi.ntrs.nasa.gov/20110015779
was powered off on 8 April 2015 when TRMM reached .pdf.]
its decommissioning altitude of 335 km. This and all —, C. A. Morales, and M. A. F. Silva Dias, 2011b:
previous work related to TRMM LIS and OTD have Electrification of precipitating systems over the
proven the excellent low-Earth-orbit lightning mea- Amazon: Physical processes of thunderstorm
surements. To extend the continuity of these measure- development. J. Geophys. Res., 116, D08209,
ments, a space-qualified LIS, built as the flight spare for doi:10.1029/2010JD014756.
TRMM, will be launched on a science mission to the —, D. J. Cecil, and S. J. Goodman, 2014: Lightning.
International Space Station (ISS) in an orbit altitude Encyclopedia of Remote Sensing, E. G. Njoku, Ed.,
(426 km) similar to TRMM LIS (402 km) (Blakeslee Encyclopedia of Earth Sciences Series, Springer,
and Koshak 2016). This LIS is expected to be launched 339–344.
and placed in operation aboard the ISS in early-to mid- Amador, J. A., 1998: A climatic feature of the tropical
2017 (R. Blakeslee 2016, personal communication). Americas: The trade wind easterly jet. Top. Meteor.
Moreover, in November 2016 the next generation of Oceanogr., 5, 91–102.

AMERICAN METEOROLOGICAL SOCIETY NOVEMBER 2016 | 2065

Unauthenticated | Downloaded 01/02/23 05:38 PM UTC
 —, 2008: The intra-Americas sea low-level jet: Over- —, S. J. Goodman, D. J. Boccippio, E. J. Zipser, and S. W.
view and future research. Ann. N. Y. Acad. Sci., 1146, Nesbitt, 2005: Three years of TRMM precipitation
153–188, doi:10.1196/annals.1446.012. features. Part I: Radar, radiometric, and lightning char-
Ba, M. B., and S. E. Nicholson, 1998: Analysis of convec- acteristics. Mon. Wea. Rev., 133, 543–566, doi:10.1175
tive activity and its relationship to the rainfall over /MWR-2876.1.
the Rift Valley Lakes of East Africa during 1983–90 —, D. E. Buechler, and R. J. Blakeslee, 2014: Gridded
using the Meteosat infrared channel. J. Appl. Meteor., lightning climatology from TRMM-LIS and OTD:
37, 1250–1264, doi:10.1175/1520-0450(1998)037 Dataset description. Atmos. Res., 135–136, 404–414,
<1250:AOCAAI>2.0.CO;2. doi:10.1016/j.atmosres.2012.06.028.
Balas, N., S. E. Nicholson, and D. Klotter, 2007: The Christian, H. J., R. J. Blakeslee, and S. J. Goodman,
relationship of rainfall variability in west central 1992: Lightning Imaging Sensor (LIS) for the Earth
Africa to sea–surface temperature fluctuations. Int. Observing System. NASA Tech. Memo. 4350, 36 pp.
J. Climatol., 27, 1335–1349, doi:10.1002/joc.1456. —, and Coauthors, 2003: Global frequency and dis-
Barros, A. P., and T. J. Lang, 2003: Monitoring the tribution of lightning as observed from space by the
monsoon in the Himalayas: Observations in central Optical Transient Detector. J. Geophys. Res., 108,
Nepal, June 2001. Mon. Wea. Rev., 131, 1408–1427, 4005, doi:10.1029/2002JD002347.
doi:10.1175/1520-0493(2003)131<1408:MTMITH Codazzi, A., 1841: Resuìmen de la Geografiìa de Venezuela.
>2.0.CO;2. Impr. de H. Fournier y compia, 648 pp.
—, G. Kim, E. Williams, and S. W. Nesbitt, 2004: Court, A., and J. F. Griffiths, 1982: Thunderstorm cli-
Probing orographic controls in the Himalayas matology. Thunderstorm Morphology and Dynamics
during the monsoon using satellite imagery. Nat. of Thunderstorms: A Social, Scientific, and Techno-
Hazards Earth Syst. Sci., 4, 29–51, doi:10.5194/nhess logical Documentary, E. Kessler, Ed., University of
-4-29-2004. Oklahoma Press, 11–52.
Biasutti, M., S. E. Yuter, C. D. Burleyson, and A. H. Sobel, Falcon, N., 2011: Phenomenology and microphysics of
2012: Very high resolution rainfall patterns measured lightning flash of the Catatumbo River (Venezuela).
by TRMM precipitation radar: Seasonal and diurnal Proc. 14th Int. Conf. on Atmospheric Electricity, Rio
cycles. Climate Dyn., 39, 239–258, doi:10.1007 de Janeiro, Brazil, ICAE, 1–4.
/s00382-011-1146-6. Flohn, H., and K. Fraedrich, 1966: Tagesperiodische
Blakeslee, R, and W. Koshak, 2016: LIS on ISS: Expanded zirkulation und niederschlagsverteilung am Victoria-
global coverage and enhanced applications, Earth See (Ostafrika) [The daily periodic circulation and
Obs., 28, 4–14. [Available online at http://eospso.nasa distribution of rainfall over Lake Victoria (East
.gov/sites/default/files/eo_pdfs/May_June_2016 Africa)]. Meteor. Rundsch., 19, 157–165.
_color%20508.pdf.] Futyan, J. M., and A. D. Del Genio, 2007: Relation-
Boccippio, D. J., S. J. Goodman, and S. J. Heckman, ships between lightning and properties of convec-
2000: Regional differences in tropical lightning tive cloud clusters. Geophys. Res. Lett., 34, L15705,
distributions. J. Appl. Meteor., 39, 2231–2248, doi:10.1029/2007GL030227.
doi:10.1175/1520-0450(2001)040<2231:RDITLD Garreaud, R., and J. M. Wallace, 1997: The diurnal
>2.0.CO;2. march of convective cloudiness over the Americas.
Brooks, C. E. P., 1925: The distribution of thunderstorms Mon. Wea. Rev., 125, 3157–3171, doi:10.1175/1520-
over the globe. Geophys. Mem., 24, 147–164. 0493(1997)125<3157:TDMOCC>2.0.CO;2.
Buechler, D. E., W. J. Koshak, H. J. Christian, and Goodman, S. J., and H. J. Christian, 1993: Global obser-
S. J. Goodman, 2014: Assessing the performance vations of lightning. Atlas of Satellite Observations
of the Lightning Imaging Sensor (LIS) using deep Related to Global Change, R. J. Gurney, J. L. Foster,
convective clouds. Atmos. Res., 135–136, 397–403, and C. L. Parkinson, Eds., Cambridge University
doi:10.1016/j.atmosres.2012.09.008. Press, 191–219.
Bürgesser, R. E., M. G. Nicora, and E. E. Ávila, —, D. E. Buechler, K. Knupp, K. Driscoll, and E. W.
2012: Characterization of the lightning activity of McCaul Jr., 2000: The 1997–98 El Nino event and
“Relámpago del Catatumbo.” J. Atmos. Sol. Terr. related wintertime lightning variations in the
Phys., 77, 241–247, doi:10.1016/j.jastp.2012.01.013. southeastern United States. Geophys. Res. Lett., 27,
Cecil, D. J., and C. B. Blankenship, 2012: Toward a global 541–544, doi:10.1029/1999GL010808.
climatology of severe hailstorms as estimated by —, —, and E. W. McCaul Jr., 2007: Lightning. Our
satellite passive microwave imagers. J. Climate, 25, Changing Planet: The View from Space, M. King et al.,
687–703, doi:10.1175/JCLI-D-11-00130.1. Eds., Cambridge University Press, 44–52.

2066 | NOVEMBER 2016

Unauthenticated | Downloaded 01/02/23 05:38 PM UTC
Heckman, S. J., E. Williams, and B. Boldi, 1998: Total brightness temperatures at 85 and 37 GHz over the
global lightning inferred from Schumann resonance tropics and subtropics. J. Geophys. Res., 116, D23108,
measurements. J. Geophys. Res., 103, 31 775–31 779, doi:10.1029/2011JD016463.
doi:10.1029/98JD02648. Lott, J. N., R. Vose, S. A. Del Greco, T. F. Ross, S. Worley,
Houze, R. A., D. C. Wilton, and B. F. Smull, 2007: and J. L. Comeaux, 2008: The integrated surface
Monsoon convection in the Himalayan region as database: Partnerships and progress. Proc. 24th
seen by the TRMM Precipitation Radar. Quart. J. Conf. on IIPS, New Orleans, LA, Amer. Meteor. Soc.,
Roy. Meteor. Soc., 133, 1389–1411, doi:10.1002/qj.106. 3B.5. [Available online at https://ams.confex.com
Jackson, B., S. E. Nicholson, and D. Klotter, 2009: /ams/88Annual/techprogram/paper_131387.htm.]
Mesoscale convective systems over western equa- Mackerras, D., M. Darveniza, R. E. Orville, E. R.
torial Africa and their relationship to large-scale Williams, and S. J. Goodman, 1998: Global lightning:
circulation. Mon. Wea. Rev., 137, 1272–1294, Total, cloud and ground flash estimates. J. Geophys.
doi:10.1175/2008MWR2525.1. Res., 103, 19 791–19 809, doi:10.1029/98JD01461.
Jayaratne, E. R., and Y. Kuleshov, 2006: Geographi- Mapes, B. E., T. T. Warner, and M. Xu, 2003a: Diurnal
cal and seasonal characteristics of the relationship patterns of rainfall in northwestern South America.
between lightning ground flash density and rainfall Part III: Diurnal gravity waves and nocturnal con-
within the continent of Australia. Atmos. Res., 79, vection offshore. Mon. Wea. Rev., 131, 830–844,
1–14, doi:10.1016/j.atmosres.2005.03.004. doi:10.1175/1520-0493(2003)131<0830:DPORIN
Kandalgaonkar, S. S., 2003: Diurnal variation of light- >2.0.CO;2.
ning activity over the Indian region. Geophys. Res. —, T. T. Warner, M. Xu, and A. J. Negri, 2003b: Diur-
Lett., 30, 2022, doi:10.1029/2003GL018005. nal patterns of rainfall in northwestern South Amer-
—, 2005: Spatio-temporal variability of lightning ica. Part I: Observations and context. Mon. Wea.
activity over the Indian region. J. Geophys. Res., 110, Rev., 131, 799–812, doi:10.1175/1520-0493(2003)131
D11108, doi:10.1029/2004JD005631. <0799:DPORIN>2.0.CO;2.
Kodama, Y.-M., A. Ohta, M. Katsumata, S. Mori, Martinez, M., J. Ramirez, and R. Montano, 2003:
S. Satoh, and H. Ueda, 2005: Seasonal transition Actividad de rayos en Venezuela, utilizando la data
of predominant precipitation type and lightning del sensor óptico (LIS) del proyecto TRMM de la
activity over tropical monsoon areas derived from NASA. Rev. Téc. Fac. Ing., Univ. Zulia, 26, 127–139.
TRMM observations. Geophys. Res. Lett., 32, L14710, Muñoz, E., A. J. Busalacchi, S. Nigam, and A. Ruiz-
doi:10.1029/2005GL022986. Barradas, 2008: Winter and summer structure of the
Kotaki, M., 1984: Global distribution of atmospheric Caribbean low-level jet. J. Climate, 21, 1260–1276,
radio noise derived from thunderstorm activity. J. doi:10.1175/2007JCLI1855.1.
Atmos. Terr. Phys., 46, 867–877, doi:10.1016/0021 Muñoz, Á. G., J. Daz-Lobatón, X. Chourio, and M. J.
-9169(84)90026-6. Stock, 2016: Seasonal prediction of lightning activity
Kuleshov, Y., D. Mackerras, and M. Darveniza, 2006: in north western Venezuela: Large-scale versus local
Spatial distribution and frequency of lightning drivers. Atmos. Res., 172-173, 147–162, doi:10.1016/j
activity and lightning f lash density maps for .atmosres.2015.12.018.
Australia. J. Geophys. Res., 111, D19105, doi:10.1029 Negri, A. J., E. N. Anagnostou, and R. F. Adler,
/2005JD006982. 2000: A 10-yr climatology of Amazonian rainfall
Laing, A. G., and J. M. Fritsch, 1993: Mesoscale convec- derived from passive microwave satellite observa-
tive complexes in Africa. Mon. Wea. Rev., 121, 2254– tions. J. Appl. Meteor., 39, 42–56, doi:10.1175/1520
2263, doi:10.1175/1520-0493(1993)121<2254:MCCIA -0450(2000)039<0042:AYCOAR>2.0.CO;2.
>2.0.CO;2. Nesbitt, S. W., and E. J. Zipser, 2003: The diurnal cycle
—, and J. M. Fritsch, 1997: The global popula- of rainfall and convective intensity according to
tion of mesoscale convective complexes. Quart. three years of TRMM measurements. J. Climate, 16,
J. Roy. Meteor. Soc., 123, 389–405, doi:10.1002/qj 1456–1475, doi:10.1175/1520-0442-16.10.1456.
.49712353807. —, and A. M. Anders, 2009: Very high resolution pre-
Lal, D. M., and S. D. Pawar, 2009: Relationship between cipitation climatologies from the Tropical Rainfall
rainfall and lightning over central Indian region in Measuring Mission precipitation radar. Geophys.
monsoon and premonsoon seasons. Atmos. Res., 92, Res. Lett., 36, L15815, doi:10.1029/2009GL038026.
402–410, doi:10.1016/j.atmosres.2008.12.009. Nicholson, S. E., 1996: A review of climate dynamics
Liu, C., D. Cecil, and E. J. Zipser, 2011: Relationships and climate variability in eastern Africa. Limnology,
between lightning flash rates and passive microwave Climatology and Paleoclimatology of the East African

AMERICAN METEOROLOGICAL SOCIETY NOVEMBER 2016 | 2067

Unauthenticated | Downloaded 01/02/23 05:38 PM UTC
 Lakes, T. C. Johnson and E. O. Odada, Eds., CRC Takayabu, Y. N., 2006: Rain-yield per flash calculated
Press, 25–56. from TRMM PR and LIS data and its relationship to
—, 2009: A revised picture of the structure of the the contribution of tall convective rain. Geophys. Res.
“monsoon” and land ITCZ over West Africa. Climate Lett., 33, L18705, doi:10.1029/2006GL027531.
Dyn., 32, 1155–1171, doi:10.1007/s00382-008- Vera, C., and Coauthors, 2006: The South American
0514-3. Low-Level Jet Experiment. Bull. Amer. Meteor. Soc., 87,
—, and X. Yin, 2002: Mesoscale patterns of rainfall, 63–77, doi: 10.1175/BAMS-87-1-63.
cloudiness and evaporation over the Great Lakes of Virts, K. S., J. M. Wallace, M. L. Hutchins, and R. H.
East Africa. The East African Great Lakes: Limnology, Holzworth, 2013: Highlights of a new ground-based,
Palaeolimnology and Biodiversity, E. O. Odada and D. hourly global lightning climatology. Bull. Amer. Meteor.
O. Olago, Eds., Advances in Global Change Research Soc., 94, 1381–1391, doi:10.1175/BAMS-D-12-00082.1.
Series, Vol. 12, Springer, 93–119. Wang, C., 2007: Variability of the Caribbean low-level
—, and J. P. Grist, 2003: The seasonal evolution jet and its relations to climate. Climate Dyn., 29,
of the atmospheric circulation over West Africa 411–422, doi:10.1007/s00382-007-0243-z.
and equatorial Africa. J. Climate, 16, 1013–1030, Williams, E., 2004: Islands as miniature continents:
doi:10.1175/1520-0442(2003)016<1013:TSEOTA Another look at the land-ocean lightning con-
>2.0.CO;2. trast. J. Geophys. Res., 109, D16206, doi:10.1029
Orville, R. E., and R. W. Henderson, 1986: Global /2003JD003833.
distribution of midnight lightning: September 1977 —, and S. J. Heckman, 1993: The local diurnal varia-
to August 1978. Mon. Wea. Rev., 114, 2640–2653, tion of cloud electrification and the global diurnal
doi:10.1175/1520-0493(1986)114<2640:GDOMLS variation of negative charge on the earth. J. Geophys.
>2.0.CO;2. Res., 98, 5221, doi:10.1029/92JD02642.
Petersen, W. A., H. J. Christian, and S. A. Rutledge, —, and S. Stanfill, 2002: The physical origin of the
2005: TRMM observations of the global relationship land–ocean contrast in lightning activity. C. R. Phys.,
between ice water content and lightning. Geophys. 3, 1277–1292, doi:10.1016/S1631-0705(02)01407-X.
Res. Lett., 32, L14819, doi:10.1029/2005GL023236. —, K. Rothkin, D. Stevenson, and D. Boccippio, 2000:
— , R . Fu, M. Chen, a nd R . Bla keslee, 20 06: Global lightning variations caused by changes in
Intraseasonal forcing of convection and lightning thunderstorm flash rate and by changes in the number
activity in the southern Amazon as a function of of thunderstorms. J. Appl. Meteor., 39, 2223–2230,
cross-equatorial f low. J. Climate, 19, 3180–3196, doi:10.1175/1520-0450(2001)040<2223:GLVCBC
doi:10.1175/JCLI3788.1. >2.0.CO;2.
Poveda, G., and Coauthors, 2005: The diurnal cycle — , and Coauthors, 2002: Contrasting convec-
of precipitation in the tropical Andes of Colombia. tive regimes over the Amazon: Implications for
Mon. Wea. Rev., 133, 228–240, doi:10.1175/MWR cloud electrification. J. Geophys. Res., 107, 8082,
-2853.1. doi:10.1029/2001JD000380.
Pulwarty, R. S., R. G. Barry, C. M. Hurst, K. Sellinger, and WMO, 1953: World Distribution of thunderstorm
L. F. Mogollon, 1998: Precipitation in the Venezuelan days. Part I: Tables. WMO Publication 21, TP 6,
Andes in the context of regional climate. Meteor. xxx pp.
Atmos. Phys., 67, 217–237, doi:10.1007/BF01277512. Yin, X., S. E. Nicholson, and M. B. Ba, 2000: On the
Rasmussen, K. L., and R. A. Houze, 2011: Orogenic diurnal cycle of cloudiness over Lake Victoria and its
convection in subtropical South America as seen by influence on evaporation from the lake. Hydrol. Sci.
the TRMM satellite. Mon. Wea. Rev., 139, 2399–2420, J., 45, 407–424, doi:10.1080/02626660009492338.
doi:10.1175/MWR-D-10-05006.1. Zipser, E. J., C. Liu, D. J. Cecil, S. W. Nesbitt, and D. P.
—, M. D. Zuluaga, and R. A. Houze, 2014: Severe con- Yorty, 2006: Where are the most intense thunder-
vection and lightning in subtropical South America. storms on Earth? Bull. Amer. Meteor. Soc., 87,
Geophys. Res. Lett., 41, 7359–7366, doi:10.1002 1057–1071, doi:10.1175/BAMS-87-8-1057.
/2014GL061767.

2068 | NOVEMBER 2016

Unauthenticated | Downloaded 01/02/23 05:38 PM UTC