hts 
ed 
Elvidge, Christopher 
2 RADIANCE CALIBRATED NIGHTTIME LIGHTS: 1999-2000 
The basic methods for generating global maps of nighttime lights using cloud-free sections of OLS time series has been 
described by Elvidge et al., 1997 and 1999. Under normal operations the OLS low-light gain is set high and the 
resulting data of urban centers is saturated. It is possible for the gain to be turned down to avoid saturation, but a set of 
three gain settings are required to cover the full dynamic range of settlement brightnesses from bright urban cores to the 
diffused lighting present in rural environments. This style of data acquisition is referred to as “low-gain”. NGDC 
produced a global product calibrated to at-sensor radiances using 28 nights of data acquired in 1996-97 (Elvidge et al., 
1999). Figure 1 shows a sample of the radiance calibrated global nighttime products from 1996-97. Although the 
product has not had an atmospheric correction applied, we did find that the cumulative brightness of individual states of 
the USA to be highly correlated to carbon emissions from fossil fuel consumption (Figure 2). It is anticipated that the 
relation to energy related carbon emissions will improve once the nighttime data are corrected to surface radiances 
(atmospheric correction), the statewide use of non-fossil fuel energy sources (hydroelectric, geothermal, and nuclear 
power), and diffuse versus point source emissions are factored in to the analysis. 
NGDC has recently developed an atmospheric correction to extract surface radiance values from nighttime visible band 
. OLS data. Using a radiative transfer model, this algorithm corrects for path length differences encountered across 
individual OLS scan lines, surface terrain variations, and differences in the atmospheric transmission properties for 
different seasons and latitudinal bands. 
Previous global nighttime lights products generated by NGDC were not entirely satisfactory due to low numbers of 
cloud-free observations in humid-tropical regions. At NGDC's request, the U.S. Air Force has acquired seven nights of 
gain controlled OLS data each lunar cycle for a fifteen month time period (January, 1999 through March, 2000). This 
data will be processed using the recently developed atmospheric to generate the world's first global map of nighttime 
lights calibrated to surface brightness levels. The lights sources will be divided into four primary source groups: human 
settlements, gas flares, biomass burning , and fishing boats. Fires (biomass burning) will be identified based on their 
short temporal duration. Gas flares will be identified based on their extreme high brightness levels, location, and limited 
spatial extent. Fishing boats will be identified using a land-sea mask. 
3 INDEPENDENT ESTIMATION OF NATIONAL FOSSIL FUEL TRACE GAS EMISSIONS 
Figure 2 shows the relationship between cumulative brightness and annual carbon gas emissions from fossil fuel 
consumption for individual states of the USA for 1996. With the radiance calibrated global map of nighttime lights 
being produced for 1999-2000 by NGDC (described above), it would be possible to generate a similar plot for each 
nation of the world. The precise relationship between cumulative brightness and fossil fuel trace gas emissions 
undoubtably varies among countries due to differences in electric power source apportionment, per capita income, 
lighting technology in use and cultural preferences for lighting. However, countries can likely be grouped into clusters 
that have similar income levels, power sources, and lighting styles. Of these, we anticipate income level would be the 
predominant factor affecting the clustering. For countries where emissions levels are in question, it then becomes 
possible to use the data within clusters to make independent estimates of fossil fuel trace gas emissions based on the 
cumulative brightness of individual nations. By analyzing the outliers it may be possible to identify countries that are 
either under-reporting or over-estimating their emissions. By tracking the cumulative brightness of countries over time 
it may be possible to confirm or call into question reported increases or decreases in energy related trace gas emissions. 
4 GENERATING A HIGH SPATIAL RESOLUTION GLOBAL DEPICTION OF FOSSIL FUEL TRACE GAS 
EMISSIONS 
4.1 Assembly of National Level Data on Annual Fossil Fuel Consumption 
One of the premier sources of data on national fossil fuel consumption levels is the International Energy Agency (IEA). 
This organization, based in Paris, is an autonomous agency linked with the Organisation for Economic Co-operation and 
Development (OECD). The IEA collects and publishes data on the consumption of fossil fuels broken down by final 
use sector. This style of breakdown, indicating the estimated quantity of each fossil fuel type used in electric power 
generation, industrial, commercial, transportation, and residential use provides the easiest starting point for assembling 
an emissions database suitable for use with the nighttime lights. The IEA data could be augmented with data from other 
sources, such as the U.S. Department of Energy's Carbon Dioxide Information and Analysis Center (CDIAC - 
http://cdiac.esd.ornl.gov/cdiac/home.html) and Energy Information Administration (http://www.eia.doe.gov/). 
  
International Archives of Photogrammetry and Remote Sensing. Vol. XXXIII, Part B7. Amsterdam 2000. 399