3.3 Use as a tool for GLM calibration and validation
A high-priority effort at NASA MSFC has been to properly validate the
GLM instruments on GOES-16 and GOES-17. This includes examining the GLM
flash detection efficiency (DE), the flash false alarm rate, the
lightning location/timing accuracy, maximum data rate capability, and
long-term instrument degradation. Specific GLM instrument requirements
have to be validated to confirm that GLM performance is acceptable for
critical operations and decision making. GLM validation often makes use
of several different ground-based lightning detection networks. These
are high-quality data sources, but they are all land-based. A large part
of the GLM FOV, especially GLM-17, is over the ocean where the quality
of ground-truth data is highly variable. Fortunately, ISS LIS has
provided vital data of uniform quality out over the oceans.
Another problem with comparing GLM to ground sources is that the
comparison is not truly one-to-one. The GLM is an optical sensor,
whereas all of the ground-based networks consist of RF sensors. The RF
sensors look at fundamentally different physics and different parts of
the lightning flash, making one-to-one comparisons difficult. Since ISS
LIS is an optical sensor very similar to GLM (though with approximately
4x better spatial resolution), it is the only source of direct
comparison for GLM. Indeed, because ISS LIS is a heritage sensor of GLM,
with similar operation and data structure, it has provided particularly
easy/efficient inter-comparisons (i.e., both optical, both spaceborne,
and both detect lightning over land/ocean).
GLM-16 has been observed to have a substantially depleted flash DE over
the northwestern CONUS (e.g., Washington and surrounding states). A
detailed plot of GLM-16 flash DE for the period January 2018 to December
2019 is shown in Fig. 10. The GLM flashes were compared with
observations from ISS LIS (Fig. 10 left), as well as data derived from
two ground-based RF lightning detection networks - ENGLN (Fig. 10
middle) and GLD360 (Fig. 10 right). All three comparison datasets agree
on the basic structure of GLM-16’s northwestern CONUS DE depletion
(reduced to as low as 20-40%). The fact that both optical spaceborne
(ISS LIS) and RF ground-based (ENGLN and GLD360) measurements agree
provides increased confidence that the DE depletion is the result of GLM
instrument effects near the edges of its FOV.
ISS LIS continues to provide important one-to-one comparisons with GLM
data, and will remain a key dataset for current and future GLM
validation [e.g., Zhang and Cummins , 2020], as well as
potentially other future geostationary lightning observations, such as
from the forthcoming Meteosat Third Generation (MTG) Lightning Imager
(LI) [Kokou et al. , 2018]. ISS LIS further provides a single
observation system that will enable direct intercomparisons between GLM,
LI, and other space-based observations.