1.The ZEUS Long Range Detection Network

spiritualblurtedAI and Robotics

Nov 24, 2013 (5 years and 2 months ago)




The ZEUS Long Range Detection Network

In 2002 The
University of Connecticut

was awarded a three year grant from the Water Cycle
Program of NSF
Earth Sciences Directorate, with contributions by the
National Observatory of
Athens (NOA)
, the
University of
Sao Paolo
Brazil (USP)

and the
University of Nevada in Las
Vegas (UNLV)
, to establish an experimental long
range lightning detection system (named ZEUS).
The system consists of a network of Very Low Frequency (7
15 kHz) radio
located in the
opean, African and south American continents.
The receivers detect s


is the radio
signal emitted by lightning
over a broad region of the electromagnetic spectrum, which
in the VLF band can propagate over thousands of kilometers in th
e earth
ionosphere wave
The system takes full advantage of modern computing technology, signal processing algorithms,
GPS, and communications networking capabilities to optimize ground based receiver d
esign for
range detection
As of 2006
S network receivers (shown in Figure 1)

situated in:
Birmingham [UK], Roskilde [Denmark], Iasi [Romania], Larnaka [Cyprus], Evora [Portugal],
Addis Ababa [Ethiopia], Dar e Salaam [Tanzania], Bethlehem [South Africa], Lagos [Nigeria],
Dakar [Senegal], G
uadeloupe [Fr
ance], Fortaleza [Brazil] and Sã
o Pa
lo [Brazil]. Chronologically,
the deployment of the European receivers, contributed by NOA, was completed in June 2001
(Anagnostou et al. 2002; Chronis and Anagnostou 2003), while the installation of the A
was completed
with NSF
WCR support in July 2003 (Chronis and Anagnostou 2006).
Finally, i
July 2006 the

ZEUS network was further augmented with three additional receivers
contributed by USP (Fortaleza

and São Pau
lo stations) and UNLV (Gu
adeloupe station).
ZEUS system

its current configuration of stations can be used to monitor lightning activity on a
continuous basis over Europe, Africa, the Atlantic Ocean and Central
South America.

Figure 1
The ZEUS station locations as
f July 2006
. In red we present stations
proposed for future deployment.

range lightning detection systems based on VLF receivers (like ZEUS) can primarily detect
CG lightning strikes (
Lee 1986a and
Morales 2001). A typical CG lightning strike
excites sferics
that propagates along the ground (ground wave) and bounce off between the Earth’s surface and the
lower part (D
layer) of the ionosphere (sky wave).

Thus the wave type delimits the range coverage,
i.e., ground waves ~ 600 km while sky waves

up 7,000 km. Based on that, the ZEUS network was
configured to sample

the vertical electric field of the propagating sky wave. The

series from all receivers are synchronized using a time stamp from a GPS clock

with 1


pon the synchronized sferics wave forms,
he ZEUS’ locating algorithm
Arrival Time Difference (ATD) technique (Lee 1986a,b). Namely, measured ATD values are


extracted for every receiver pair by maximizing the temporal cross
correlation of the wav
sampled by each receiver. Theoretically, an ATD pair between receivers “i” and “j” is defined as:





are the distances (km) of the lightning source to the




the associate
d wave propagation velocities (km/sec) and


represent the time when the two
receivers recorded the signal. An ATD defines a hyperbola of possible lightning source positions
over the Earth’s surface associated with the same arrival time difference
. Combining the

hyperbolas of all ATD pairs (45

pairs for the 10
receiver network), we can identify a common point
of intersection, which uniquely defines the location of the
. The inverse solution of locating
the intersection point of the hyperbo
las is computationally defined by the minimization of the
following cost function (Lee 1986a):



Subscript “
” denotes the ATD pair index. “Measured” superscript symbolizes ATD values for a
lightning strike recorded by ZEUS
system. “Simulated” superscript indicates the theoretical ATD
values calculated on the basis of the propagation path (from the strike location (
) to the receivers)
and the modeled sferics group velocity along that path (
) (Chronis and Anagnostou 2003).

is the measurement and modeling error variance (given in μsecs
) of each ATD pair. This error
is mostly related to technical specifications of the receiver sensitivity and time resolution. In this

is typically conside
red the same for all ATD pairs (i.e., pairs have
equal weight in determining Eq. 2).

To attain a unique solution from Equation (2)
minimum of four receivers are required to
fix a

event. Furthermore, due to multiple sferic

reports from lightnin
g strikes occurring over
greatly varying distances from each receiver, the sferic report for any particular strike is often
interspersed with reports from other strikes in a non
sequential manner. To address this issue, the
locating algorithm searches for
reports from among the receivers that are likely same
candidates and then confirm their relationship. This process suffers a “combinatorial explosion”
when the network area is large. To handle this problem ZEUS network is divided into two
sized regions (Europe and Africa), containing seven receivers each. The European cell
consists of the five European receivers plus two receivers located in Africa (Ethiopia and Senegal).
The African cell consists of the five African receivers plus the Cy
prus and Portugal receivers of
Europe. The candidate data from those cells are then combined and the final strike location is
optimized according to Eq. 2. This allows a satisfactory compromise in computational load versus
desired accuracy. The locating pr
ocedure described in Equation (2) can be ill posed. This issue
arises particularly when the lightning event is located outside the network’s periphery. Under these
circumstances, the minimization problem suffers from the so
called multiple minima convergen
Namely, solutions for Equation (2) located outside the periphery (main solution domain) tend to be
unique. At the same time,

is no longer defined by a unique point but rather a group of points
defining an area of “multiple local minima”. Below
we present measurements and discuss error
characteristics of the system.


ZEUS Measurements


ZEUS network has provided long
range data over the past 5 years based on different station
configurations since the system is constantly being upgraded. For exampl
n Figure 2 we present
global distributions of monthly lightning accumulation measured by ZEUS system for the period
July 2004 to February 2005

that is

based on the Europe
Africa receiver network.

Figure 2:

Monthly ZEUS measured lightning accumulation for the period July 2004 to February 2005.

noted from Figure 2

are that ZEUS system with that

configuration could detect lightning
activity over Europe, Africa, Central
South America and the Atlantic Oc
ean. The geographic
regions exhibiting erroneous patterns in the lightning data (e.g., Northern South America, Central
America, Pacific Ocean, South Asia) are associated with poorly defined minima in the

Those are regions where the network rece
pairs’ ATD hyperbolas become almost parallel, thus
significantly decreasing the sensitivity of


surface around the minimum. In such situations
measurement noise and errors in modeling the signal’s wave velocity would make the locating


algorithm to c
onverge to sub
optimal solutions that as mentioned above can be far from the “true”
lightning source location. This problem is discussed in detail in Chronis and Anagnostou (2003)
and (2006) who have conducted experimental and simulation based evaluations

of the location
retrieval error of ZEUS network for different regions. To optimize ZEUS accuracy in the Atlantic
Ocean and South America we augmented the network with the Guadeloupe,
São Paulo
Fortaleza receivers as discussed earlier in the proposal.

ZEUS measurements over the Atlantic and
South America region with the current system configuration show dramatic improvements in the
locating accuracy of the system
over the Atlantic and Central
South America region

et al.

work in progress)

An example of ZEUS measurements after July 2006 is shown in Figure 3

Note the disappearance (in Figure 3) of the elongated solutions over the Golf of Mexico,
South America that is apparent in Figure 2 plots. The lack of lightning measur
ements in
East Africa in Figure 3 is because the solutions in this plot have not included the European
station data. Comparing ZEUS and LIS measurements from Figure 3 we note a strong coincidence
on lightning distribution. ZEUS detected notable li
ghtning activity in the coast of West Africa and
over the

that is not captured well by LIS. Increasing the network stations in the
Americas will further enhance the detection efficiency and locating accuracy of ZEUS in the area.


ZEUS and LIS measured total lightning activity for the period August
September 2006. A point to
note is that ZEUS
network measurements shown

in this figure

not include the European station data,
which is why
the network missed the

in the Central
East Africa

We have


compared ZEUS data against corresponding measurements from another VLF

lightning detection network
(named WWLLN global lightning detection system)

described in Lay et al. (

he comparison (shown in Figure 4
) is made in terms of the
mean flash rate diurnal variability derived from measured ZEUS and WWLLN data over Africa,
Atlantic and the Amazon basin

during August
September 2004
A point to note from this
figure is that

ZEUS has significantly higher (~10 times) detection

than WWLLN,
and provides


much better definition of the diurnal

variability of convection over those regions.
We would like
to point out that

both networks are undergoing changes in both the numbe
r of stations,
configurations and locating software

so this plot

does not constitute a
network performance
comparison, but, should be viewed as an indication of the ZEUS potential i
n capturing lightning
activity over

a, Atlantic Ocean, and
South America.



Current Status of ZEUS Network

December 2006

the ZEUS network has been renamed to STARNET
Sferics Timi
ng And
Ranging NETwork (
) and operated by the University
of Sao Paolo in collaboration with the University of Connecticut. The currently operational receiver
locations are shown in
, while data from the European
ZEUS network are provided off
line on project need basis. Real
time STARNET data are available
from the official web site covering the Atlantic Ocean, Central and South America a
nd part of the
Eastern Pacific Ocean. The Africa and European continents can be covered on the basis of
integrated European ZEUS network observations.


Contributions to Science

Over the years visible (VIS) and infrared (IR) imagery from geo
stationary satellites has been
practically the sole sensor for near
continuous global
scale cloud observations.
As described herein

has now advanced to the point where lightning can be continuously monitored over
large regions by means of the electromagnetic

that is emitted in the Very Low Frequency
(VLF) range of 5 to 15 kHz. Whereas VIS/IR imagery yields information on the heig
ht of cloud
tops, lightning provides information about the internal structure of clouds, which
can be used in a
number of applications ranging from
predicting cyclone genesis and hurricane intensification to
advancing understanding and the predictability o
f global/regional water cycle
. Those are
summarized below.

Figure 4:
Mean flash rate
) versus hour of the
day (in UTC) determined from ZEUS
olid lines) and WWLLN (dashed lines)
measurements over Africa, South
America (Amazon basin) and ITCZ
(Atlantic Ocean).

T h e
s e p l o t s w e r e
d e r i v e d f r o m A u g
S e p
2 0 0 4 d a t a
r e c o r d s

t h u s a r e b a s e d o n E u r o p e a n d
A f r i c a n
Z E U S n e t w o r k c o n f i g u r a t i o n


A critical aspect in investigating continental
scale water cycle is the need to resolve the land
surface hydrological processes at high temporal (less than 3
hourly) and spatial (less than 5 km)
esolutions. This is important because the physical processes that dictate land
atmosphere interactions occur at these fine space
time scales. Although, Land Surface Models
(LSMs) are capable of modeling land
atmosphere interactions at resolutio
ns down to 1
km, the
most current attempts to estimate rainfall near those scales have relied on proxy VIS/IR data merged
with that from infrequent but more physically
based passive microwave estimates (e.g., Todd et al.
2000; Huffman et al. 2003; Turk et
al. 2002). Unfortunately, the desired progression to finer scales
in satellite rain retrieval is counter
balanced by the increasing dimensionality of error, which has a
consequentially complex effect on land surface
atmosphere interaction simulations (Ana
2005; Lee and Anagnostou 2004; Hossain and Anagnostou, 2005). In essence, this represents a
competing trade off between lowering the rain retrieval error and modeling land
atmosphere processes at the finest scale possible. Our studies
based on ZEUS data have shown that
the inclusion of continuous lightning observations (proxy to convective precipitation) in satellite
observations improves the physical basis of the retrieval facilitating higher resolution rain estimates
(<0.1 deg) and im
proved accuracy in the estimation of (convective) rain rates (Morales and
Anagnostou 2003; Chronis et al. 2004). This advancement in high
frequency satellite rain
estimation was further corroborated by Anagnostou (2005) and Anagnostou et al. (2004a) to hav
consequential improvements on the accuracy of LSM simulations.

Quantitative precipitation forecasts (QPF) that support flood forecasting and other severe
weather management systems rely on outputs from regional numerical weather prediction (NWP)

At the scales typically used by regional NWP, numerical models exhibit low skill at
forecasting highly variable convective precipitation events. Improved initialization of the local
environment in NWP models (particularly those of moisture and temperatu
re profiles) is an avenue
for potentially improving QPF as it relates to simulated convective precipitation. To this end,
continuous lightning observations over large regions (such as those available from ZEUS network)
provide useful information about the

growth, location, lifecycle and ice microphysics of convection
in mesoscale convective systems. This is because charge separation leading to lightning is a
physical process that takes place in regions of a thunderstorm associated wi
th rigorous vertical
Arguably, developing a way to modify model
generated variables (e.g., humidity and
temperature profiles) on the basis of lightning location and intensity could result in more accurate
parameterizations of the convective environment. Recently, Pap
adopoulos et al. (2005a) made use
of the sound link between lightning occurrence and convection to develop a technique to amend the
initialization data inadequacies and limitations in formulating sub
grid scale processes in NWP
using continuous regional li
ghtning observations from ZEUS network. They showed that nudging
the model humidity profiles to empirical convective profiles relative to the flash rates, leads to more
realistic model soundings and consequential improvements in convective precipitation f
This study was recently expanded to demonstrate the technique’s capabilities in different regions
(Europe, US, and Africa

Papadopoulos et al. 2005b). Example from application of the
assimilation technique using ZEUS data over a three
day period

in West Africa is presented in
. In this example, control (CTRL) and ZEUS
assimilated (CASE) mesoscale
model runs were performed using NCEP reanalysis initial and boundary condition fields. Model
predictions are compared against correspo
nding gridded (0.2
deg resolution) rain rate fields derived
from satellite passive microwave (SSM/I) observations (Dinku and Anagnostou 2005a,b). The
following points are noted: The intense rainfall observed by SSM/I is well correlated with regions
of lig
htning activity detected by ZEUS. Significant improvement is demonstrated comparing CTRL
and CASE rainfall field error statistics (Bias Score

BS, Equitable Threat Score

ETS, Hiedle

HS, and Root
Square Error

RMSE). Evidence from the Papadopoulo
s et al.


(2005a,b) studies indicates the capability of lightning
data in triggering and enhancing convection
in NWP models. Future extension of the aforementioned studies is to strengthen the physical basis
of lightning data assimilation through simulation
s from a cloud
resolving model with lightning


SSM/I retrieved rainfall (mm/h) overlaid by contours delineating ZEUS lightning activity occurring
within a ±½
hour time window centered at the SSM/I overpass time;

error statistics (SSM/I vs. model) of CASE
(green lines) and CTRL (red lines) model runs evaluated based on 11 SSM/I orbits in the period Aug 13
15 2004.

Complementary approaches to the direct assimilation of lightning data are to assimilate rainfall and

heating parameters retrieved from combined satellite Infrared and lightning observations
(Alexander et al. 1999; and Chang et al. 2001). Those authors had demonstrated that assimilation of
convective rainfall and adjusted latent heating diagnosed
from lightning data could improve the
term forecasting of convective storms (Goodman 2003). The physical basis for this approach
relies on well
established links between convective updraft strength and the initiation of robust
mixed phase precipitat
ion processes, electric charge separation, and ultimately lightning production.

Studying the lightning formation in tropical storms off the coast of Africa

could provide the
additional information needed to improve remote sensing and modeling of those systems. As
indicated by Boccippio et al. (2000) and Williams et al. (2000), oceanic thunderstorms may produce
similar lightning flash rates to those observed
in continental storms. The main difference between
the two systems is in the frequency of occurrence of thunderstorms (continental thunderstorms are
significantly more frequent than oceanic). Consequently, the occurrence of lightning in oceanic
storms is

very important microphysical information (probably as important as in continental
systems), as it indicates regions of strong vertical velocity and ice concentrations above freezing

level (Zipser and Lutz 1994).
To understand, though, the complex relation

of lightning flash rates to
meteorological properties controlling charge separation (vertical motion, ice particle concentration,
precipitation structure etc.) it is desirable to have continuous data of lightning flash rate distribution
(from cloud
und discharges) jointly with other
in situ

or remotely sensed (space
based and
airborne) measurements of storm dynamics, ice microphysics (concentration and particle size
distribution) and precipitation water content profiles.

Besides the direct associati
on of location and timing lightning information with the
abovementioned micro/macro
physical variables, the polarity of the lightning discharge is
associated with other intense and consequential electrical discharges occurring in the atmosphere.


Such pheno
mena that only recently the scientific community has focused on are the sprites and
elves (Inan et al. 1991). Scientific hypothesis relates these phenomena to molecular excitation due
to lightning electromagnetic pulse heating that increases the electron p
opulation in the lower
ionosphere and creating an elongated luminous discharge. Recent studies have shown that it is
mostly the positive CG lightning discharges that generate these vertically elongated structures right
above active thunderstorms. These dis
charges follow the +CG lightning that present charge moment
greater than 600 Coulomb/km (Hu et al., 2002; Williams 1998). Recently, satellite and ground
based observations (Smith et al., 2005; Cummer et al., 2005) related to the generation of terrestrial
amma rays escaping into space and occurring above active thunderstorms. Williams et al. (2006)
using combined ground
based VLF observations, among which the ZEUS network data, were able
to identify and collocate gamma rays observed by the RHESSI satellite

Moreover, the knowledge
of polarity can give some insight on the cloud charge mechanism.
For instance Willians et al.

and Cary and Buffalo

found the
high cloud base heights may provide larger cloud
water in the mixed phase, which is favorab
le for the positive charging of large ice particles that may
result in storms with

reversed polarity of its main

Another important contribution of long
range lightning monitoring is in the study of tropical
storms and hurricanes, and in particular the aspects of the
ir genesis and evolution (Goodman 1990).
Hurricanes are associated with relatively infrequent lightning occurrences, but under the influence
of strong vertical forcing can exhibit robust ice processes (i.e., interactions between ice and graupel)
leading to

lightning (Black and Hallett, 1999). Few studies exist on the association of lightning
with the initiation (genesis) and intensification of a hurricane (Molinari et al. 1999, 1994; Samsury
et al. 1994; Lyons et al. 1994; Williams 1995). Observations of
those studies have shown that
lightning activity in the vicinity of a hurricane’s eyewall region occurs during the intensification
phase of a hurricane, and that the most frequent lightning activity is at convective centers of the
outer rain
bands (>100km)
. This is evident in the snapshot image (
) of hurricane Linda
lightning activity taken by the Optical Transient Detector (OTD) during intensification period. On
the other hand, only weak experimental evidence exists on the association of lightnin
g intensity in
the pre
hurricane tropical thunderstorms to the genesis of the hurricane (Molinari et al. 1999). This
is due to the lack of continuous lightning observations over remote oceanic regions (Atlantic and
Pacific) where hurricanes originate. Lim
ited experimental studies by Molinari et al. (1999) based on
the National Lightning Detection Network data indicate that a positive impact can be expected from
the use of continuous lightning observation in predicting the timing and location of cyclone gen
The Eastern Atlantic is a typical example where tropical disturbances over West Africa grow into
tropical depressions, tropical storms and eventually hurricanes (Berry and Thorncroft, 2005).
During boreal summer AEW are formed as the result of mixed

baroclinic and barotropic
instabilities originating from Africa and propagate westward with varying periods (3
6 days) and
typical wavelengths ranging from 2,000
3,000 km. A typical AEW activity produces 30
60 waves
per boreal warm season (Avila and Pasch
, 1992) or an average of 6
10 waves per month (May

Snapshot of hurricane Linda eyewall lightning observed
by the OTD during a period of changing intensity on September 12,
1997. Lightning is also observed at the outer rain
bands of the

The image was extracted from MSFC web page on LIS and OTD

accomplishments to Science


November). Roughly 10% of these disturbances become “seeds” for the development of TC and
potential hurricane formation off the West African coast, especially between the latitudes of 10o
15o N (Carlson, 1
969a,b). A key ingredient for cyclogenesis is vertical air motion that concentrates
angular momentum. This vertical motion, also associated with deep convective cloud systems
embedded in the AEW, is ever present in the tropical westward
moving air masses
AEW. However, the electrification activity within these large
scale features lacks thorough
understanding due to lack of continuous and spatially distributed CG lightning data over the main
chimneys (Africa
S. America). A very recent study based

on ZEUS network observations from the
2004 hurricane season showed that AEW disturbances exhibiting sustained lightning in the Atlantic
are far more likely to produce tropical cyclones (TC) than those for which lightning ceases at the
coast (

et al
. 2007
Along the same line
, Price et al.


evidence that most
of the Atlantic hurricanes initiated as African easterly waves over the African continent were
associated with lightning activity in the beginning of its development in the eas
tern part of Africa.
Similarly, observations from the Los Alamos National Laboratory on hurricanes Katrina and Rita
have shown that lightning can signal the intensification stages of hurricanes (Shao et al. 2005).

Attempts to retrieve robust geophysical s
ignals from long term lightning time series, were
motivated by earlier works on the behavior of global lightning activity (e.g., Williams 1992,
Goodman et al. 2000; Christian et al. 2003) and trends between periodic, quasi

periodic phenomena
or free mode K
elvin waves. A number of these studies revealed that global lightning activity shows
sensitivity to a number of different sets of variables (temperature, ENSO index, etc) as well as
magnitude modulation by long range propagating Rossby waves. Castro (2000
) and Patel (2001)
using Schumann resonance showed a clear lightning activity and precipitation modulation over
continental Africa by the so
called 5
day wave phase initially documented by Madden and Julian
(1972) with further supporting evidence by and Ma
dden (1978). Recently, observations from ZEUS
range lightning detection network in Africa revealed
linkage between tropospheric convective
forcing and
he intensification of the wave (Chronis et al. 2007).
This feature exhibits a
1 structur
e and amplitude variation of 0.5
1 mb and its phase variation show high
agreement with accumulated lightning activity and rainfall over the African continent.
information would be very important to study the
causes of water cycle variability as studie
s show
that the Africa chimney can change by as much as a factor of two in rainfall and more than a factor
of two in lightning on a five
day time scale. Comparable studies have targeted other similar periodic
signals as the Madden
Julian (Madden and Julian
, 1972) 40
50 day oscillation as means to
improved weather forecasts in the tropical atmosphere (Cadet and Daniel 1988; Waliser et al.,
2003). Future research may be extended towards the potential link between the potential harmonic
modulation of AEW and
larger scale features like the aforementioned Rossby 5
day wave and their
implication for cyclogenesis on
Eastern Atlantic


Contributions to


ZEUS data may contribute to a number of research and applied projects. Some of the
centers or
programs that can make use of ZEUS data are:

African Monsoon Multidisciplinary Analyses (AMMA, http://www.joss.ucar.edu/amma/)

Global Precipitation Measurement (GPM) Mission (http://gpm.gsfc.nasa.gov/)

The Nile Basin Initiative (NBI) project

National, Regional and/or International Weather Forecast Offices and Water Resources
Management Agencies in Africa.


US and International agencies responsible for flood and drought monitoring in Africa (e.g.,
USAID, World Bank, EU).



Alexander, G. D., J. A. Weinman, J. A, Karyampudi, V. M, Olson, W. S, Lee, A. C., 1999: The
effects of assimilating rain rates derived from satellite and lightning on forecasts of the 1993
superstorm. Mon. Wea. Rev., 1
27, 1433

Anagnostou, E.N., 2005: Assessment of satellite rain retrieval error propagation in the prediction of
land surface hydrologic variables, Book Chapter in Measuring Precipitation from Space:
EURAINSAT and the Future, (eds) V. Levizzani, P. Ba
uer and F.J. Turk. Kluwer Academic

Anagnostou, E.N., K.H. Lee, T. Dinku, D. Wang, and A. Tadese, 2004a: Assessment of satellite
rainfall estimation in simulating land surface processes, 18th Conference on Hydrology, 84th
AMS Annual Meeting, 11
15 January 2004 Seattle, WA, USA.

Anagnostou, E.N., M.A. Anagnostou, W.F. Krajewski, A. Kruger, and B. Miriovsky, 2004b: High
Resolution Rainfall Estimation from X
Band Polarimetric Radar Measurements,” Journal of
Hydrometeorology, Vol. 5, No. 1, 110
, 2004.

Anagnostou, E.N., and C. Morales, 2002: Rainfall Estimation from TOGA Radar Observations
during LBA Field Campaign, Journal of Geophysical Research
Atmospheres, 107(D18).

Anagnostou, E.N., T. Chronis, and D.P. Lalas, 2002: New Receiver Network Ad
vances Long
Range Lightning Monitoring. EOS Feature Article, AGU, 83(50), 589, 594

Anagnostou, E.N., and W.F. Krajewski, 1999a: Real
Time Radar Rainfall Estimation: 1. Algorithm
Formulation, Journal of Atmospheric and Oceanic Technology, 16(2), 189

Anagnostou, E.N., and W.F. Krajewski, 1999b: Real
Time Radar Rainfall Estimation: 2. Case
Study, Journal of Atmospheric and Oceanic Technology, 16(2), 198

Anagnostou, E.N., W.F. Krajewski, and J. Smith, 1999: Uncertainty Quantification of Mean
Rainfall Estimates, Journal of Atmospheric and Oceanic Technology, 16(2), 206

Avila L.A. and R.J. Pasch. 1992: Atlantic Tropical Systems of 1991. Monthly Weather Review:
Vol. 120, No. 11, pp. 2688


Berry, G. J. and C. Thorncroft,
2005: Case study of an intense African Easterly Wave. Mon. Wea.
Rev. 133, 752

Black, R.A. and J. Hallett, 1999: Electrification of the Hurricane. J. Atmos. Sci., 56, 2004


Boccippio, S.J. Goodman, and S. Heckman, 2000: Regional Differences In
Tropical Lightning
J. Appl. Met. (TRMM Special Issue)
, Vol 39, 2231

Boccippio, D.J., Driscoll, K., Hall, J., Buechler, D., 1998: LIS/OTD Software Guide, Global
Hydrology and Climate Center.

Burpee, R.W., 1972: The origin and structu
re of easterly waves in the lower troposphere in North
Africa, J. Atmos. Sci., 29, 77

Budden, K.G., 1951: The Propagation of a Radio
Atmospheric. Phil. Mag., Ser 7, 42, 1

Cadet, D.L. and P. Daniel, 1988: Long range forecast of the break and acti
ve summer monsoons,
Tellus, 40A, 133

Carlson, T.N, 1969: Some remarks on African disturbances and their progress over the tropical
Atlantic, Mon. Wea. Rev., 97, 716

Castro, D.S., 2000: The relationship between precipitation and electromagnetic
signals in Schumann
resonances, M. Eng, Thesis, Massachusetts Institute of Technology, Cambridge MA.

Christian H., R.J. Blakeslee, D.J. Boccippio, W.L. Boeck, D.E. Buechler, K.T. Driscoll, S.J.
Goodman, J.M. Hall, W.J. Koshak, D.M. Mach and M.F. Stewart,
2003: Global frequency and
distribution of lightning as observed from space by the Optical Transient Detector. J. Geophys.
Res., 108:doi:10.1029/2002JD002347.

Christian, H.J., R.J. Blakeslee, S.J. Goodman, D.A. Mach, M.F. Stewart, D.E. Buechler, W.J.
ak, J.M. Hall, W.L. Boeck, K.T. Driscoll, and D.J. Boccippio, 1999: The Lightning
Imaging Sensor. Proccedings of the 11th Int. Conf. on Atmos. Electricity, Guntersville,
Alabama, 746

Chronis T., and Anagnostou E., 2005: Evaluation of a Long
Range Lig
htning Detection Network
with Receivers in Europe and Africa, IEEE Transaction on Geosciences and Remote Sensing

Chronis, T.G., E. Williams, and E.N. Anagnostou, 2005: African Lightning activity and rainfall
modulation by periodic pressure wav
es. IAMAS Proceedings, Beijing, China 2005.

Chronis T., Anagnostou E and Dinku T., 2004: High frequency estimation of thunderstorms via
Satellite Infrared and a long
range lightning network in Europe, Quarterly Royal Meteorological
Society Vol.130, No.

Chronis T. and Anagnostou E., 2003: Long Range Validation of the European Long Range
Lightning Detection Network, J. Geophys. Res., Vol. 108, No. D24, 4779.

Chang, D.
E., J. A. Weinman, C .A. Morales, and W. S. Olson, 2001: The effect of spaceborne
microwave and ground
based continuous lightning measurements on forecasts of the 1998
Groundhog Day Storm. Mon. Wea. Rev., 129, 1809



Cummer, S.A., Y. Zhai, W. Hu, D.M. Smith, L.I. Lopez, and M.A. Stanley, 2005: Measurements
and implications of the r
elationship between lightning and terrestrial gamma
ray flashes,
Geophys. Res. Lett., 32, NO.8, L08811.

Del Genio, A. D., and W. Kovari, 2002: Climatic properties of tropical precipitation convection
under varying environmental conditions. J. Climate, 15,


Diedhiou, A., S. Janicot, A. Viltard, P. de Felice, and H. Laurent, 1999: Easterly wave regimes and
associated convection over West Africa and tropical Atlantic: results from the NCEP/NCAR
and ECMWF reanalyses, Clim. Dyn., 15, 795


T., and E.N. Anagnostou, 2005a: Regional Differences in Overland Rainfall Estimation from
Calibrated TMI Algorithm, Journal of Applied Meteorology, Vol. 44, No. 2, pp. 189


Dinku, T., and E.N. Anagnostou, 2005b: TRMM Calibration of SSM/I Algorithm

for Overland
Rainfall Estimation, Journal of Applied Meteorology (accepted).

Goodman, S. J., 2003: Atmospheric electrical activity and the prospects for improving short
weather forecasting. Reprints of the 12th International Conference on Atmospheri
c Electricity,
13 June 2003, Versailles, France.

Goodman S.J., D.E. Buechler, K. Knupp, K. Driscoll and E.W. McCaul Jr., 2000: The 1997
El Nino event and related wintertime lightning variations in the Southeastern United States,
Geophys. Res. Lett.
, 27:541

Goodman, J.S., 1990: Predicting thunderstorm evolution using ground based lightning detection
networks. NASA Tech Memo. TM

Hossain, F., and E.N. Anagnostou, 2005 (in press): Numerical Investigation of the Impact of
Uncertainties in
Satellite Rainfall and Land Surface Parameters on Simulation of Soil Moisture,
Advances in Water Resources.

Hu, W., S.A. Cummer, W.A. Lyons, and T. Nelson, 2002: Lightning charge moment changes for
the initiation of sprites, Geophys. Res. Lett, 29(8), 127

Huffman, G.J., R.F. Adler, E.F. Stocker, D.T. Bolvin, E.J. Nelkin, 2003: Analysis of TRMM 3
Hourly Multi
Satellite Precipitation Estimates Computed in Both Real and Post
Real Time.

Conf. on Sat. Meteor., 2 and Oceanog., 9
13 Feb. 2003, Long Beach
, CA.

Inan, U., T.

Bell, and J.

Rodriguez, 1991: Heating and ionization of the lower ionosphere by
lightning, Geophysical Research Letters, 18, 705

Koshak, W. J., Solakiewicz, R. J., 2001: TOA Lightning Location Retrieval on Spherical and
Oblate Spher
oidal Earth Geometries, J. Atmos. and Ocean. Tech., 18, 187


Lee, K.H. and E.N. Anagnostou, 2004: Investigation of the nonlinear hydrologic response to
precipitation forcing in physically based land surface modeling, Canadian Journal of Remote
g, Vol. 30, No. 5, 706

Laurent H., L. A. T. Machado, C. A. Morales, and L. Durieux, 2002: Characteristics of the
Amazonian mesoscale convective systems observed from satellite and radar during the
WETAMC/LBA experiment, J. Geophys. Res., 107 (D20), 8
054, doi:10.1029/2001JD000337.

Lyons, Walter A., Keen, Cecil S. 1994: Observations of Lightning in Convective Supercells within
Tropical Storms and Hurricanes. Monthly Weather Review: Vol. 122, No. 8, pp. 1897


Machado L.A.T. and H. Laurent, 2004: T
he convective system area expansion over Amazonia and
its relationships with convective system life duration and high
level wind divergence. Monthly
Weather Review, 132, 714

Madden R.A., 1978: Further Evidence of Traveling Planetary Waves. Journal of

the Atmospheric
Sciences: Vol. 35, No. 9, pp. 1605


Madden R.A. and P.R. Julian. 1972: Description of Global
Scale Circulation Cells in the Tropics
with a 40

50 Day Period. Journal of the Atmospheric Sciences: Vol. 29, No. 6, pp. 1109


i, John, Moore, Paul, Idone, Vincent. 1999: Convective Structure of Hurricanes as Revealed
by Lightning Locations. Monthly Weather Review: Vol. 127, No. 4, pp. 520


Molinari, J., P.K. Moore, V.P. Idone, R.W. Henderson, and A.B. Saljoughy, 1994: Cloud
lightning in Hurricane Andrew, J. Geophys. Res., 16665

Morales, C., and Anagnostou, E.N., 2003: Extending the Capabilities of High
Frequency Rainfall
Estimation from Geostationary
Based Satellite Infrared via a Network of Long
Range Light
Observations, Journal of Hydrometeorology, 4(2), 141

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

2000. Monthly
Weather Review: Vol. 130,

No. 8, pp. 2098


Papadopoulos, A., T.G. Chronis and E. N. Anagnostou, 2005a: Improving Convective Precipitation
Forecasting Through Assimilation of Regional Lightning Measurements in a Mesoscale Model,
Monthly Weather Review, Vol. 133, 1961


Papadopoulos A., Th. Chronis, and E. Anagnostou, 2005b: Assimilation of Continuous Cloud
Ground Lightning in Mesoscale Models”. Book of Abstracts of the 4rd EuroGoos Conference,
9 June 2005, Brest, France, 245.

Pasch R.J. and L.A. Avila. 1992: Atlant
ic Hurricane Season of 1991. Monthly Weather Review:
Vol. 120, No. 11, pp. 2671


Patel, C.A., 2001: Modulation of African lightning rainfall by the global five day wave, B.Sc,
Thesis, Massachusetts Institute of Technology, Cambridge MA.


Petersen, W.
A., R. Cifelli, D.J. Boccippio, S.A. Rutledge, and C. Fairall, 2003: Convection and
Easterly Wave Structure Observed in the Eastern Pacific Warm
Pool during EPIC
2001. Journal
of Atmospheric Science, 60, 1754

Petersen, W.A., R. Cifelli, D.J. Boccip
pio, and S.A. Rutledge, 2002: Convection and Easterly
Waves Observed in the Eastern Pacific ITCZ During EPIC 2001. Abstract, 25th Conference on
Hurricanes and Tropical Meteorology, American Meteorological Society, 29 April

3 May,
2002, San Diego, CA.

eed, R. J., and K. D. Jaffe, 1981: Diurnal variation of summer convection over West Africa and
the tropical eastern Atlantic during 1974 and 1978. Mon. Wea. Rev., 109, 2527

Reed, R.J., D.C. Norquist and E.E. Recker. 1977: The Structure and Propertie
s of African Wave
Disturbances as Observed During Phase III of GATE. Monthly Weather Review: Vol. 105, No.
3, pp. 317


Samsury, Christopher E., Orville, Richard E. 1994: Cloud
Ground Lightning in Tropical
Cyclones: A Study of Hurricanes Hugo (1989)

and Jerry (1989). Monthly Weather Review: Vol.
122, No. 8, pp. 1887


Smith, D.M., L.I. Lopez, R.P. Lin, and C.P. Barrington
Leigh, 2005: Terrestrial gamma
ray flashes
observed up to 20MeV, Science, 307, 1085

Thorncroft, C.D., and K. Hodges, 2
001: African easterly wave variability and its relationship to
Atlantic tropical cyclone activity, J. Climate.

Thorncroft, C.D., 1995: An idealized study of African Easterly Waves. III: More realistic basic
states. Q.J.Roy.Meteorol.Soc., 121, 1589

Thorncroft, C.D., and Hoskins, B.J., 1994a: An idealized study of African Easterly Waves. I: A
linear view. Q.J.Roy.Meteorol.Soc., 120, 953

Thorncroft, C.D., and Hoskins, B.J., 1994b: An idealized study of African Easterly Waves. II: A
nonlinear view
. Q.J.Roy.Meteorol.Soc., 120, 983

Todd, M.C., Kidd, C., Kniveton, D., Bellerby, T.J., 2000: A Combined Satellite Infrared and
Passive Microwave Technique for Estimation of Small Scale Rainfall, Journal of Atmospheric
and Oceanic Technology, Vol. 18,

pp, 742

Toracinta, E. R., K. I. Mohr, E. J. Zipser, and R. E. Orville, 1996: A comparison of WSR
reflectivities, SSM/I brightness temperatures, and lightning for mesoscale convective systems in
Texas. Part I: Radar reflectivity and ightning. J.
Appl. Meteor., 35, 902

Toracinta E.R. and E. J. Zipser. 2001: Lightning and SSM/I
Scattering Mesoscale Convective
Systems in the Global Tropics. Journal of Applied Meteorology: Vol. 40, No. 6, pp. 983



Turk, J., E. Ebert, H.
J. Oh, B.
J. So
hn, V. Levizzani, E. Smith and R. Ferraro, 2002: Verification of
an operational global precipitation analysis at short time scales. 1st Intl. Precipitation Working
Group (IPWG) Workshop, Madrid, Spain, 23
27 September 2002.

Waliser, D.E., K.M. Lau, W. Ste
rn and C. Jones, 2003: Potential predictability of the Madden
Julian Oscillation, Bull. Am. Met. Soc., 84, 33

Williams E., Boldi R., Bor J., Satori G., Price C., Greengurg E., Takashashi Y., Yamamoto K.,
Chronis T., Anagnostou E., Smith D., Lopez L.,
2005: Lightning flashes conductive to
production and escape of gamma radiation to space, J. of Geophysical Research

(in review).

Williams, E., V. Mushtak , D. Rosenfeld, S. Goodman and D. Boccippio, 2005 (in press):
Thermodynamic conditions f
avorable to superlative thunderstorm updraft, mixed phase
microphysics and lightning flash rate, Atmospheric Research.

Williams, E., Rothkin, K., Stevenson, D., Boccippio, D., 2000: Global lightning variations caused
by changes in thunderstorm flash rate
and by changes in the number of thunderstorms. Journal
of Applied Meteorology, V 39, 2223, 2230.

Williams, E.R., 1998: The positive charge reservoir for sprite
producing lightning, J. Atmos. Sol.
Terr. Phys., 60, 689

Williams, E., 1995: “Meteorologi
cal aspects of thunderstorms”, in CRC Handbook on Atmospheric
Electrodynamics, Vol. I, 27
60, ed. H. Volland, CRC Press, Boca Raton, FL.

Williams, E.R., 1992: The Schumann Resonance: A global tropical thermometer, Science, 256,

Zipser, E.J., a
nd K. Lutz, 1994: The vertical profile of radar reflectivity of convective cells: A
strong indicator of storm intensity and lightning probability? Mon. Wea. Rev., 122, 1751