Geoscience Reference
In-Depth Information
1
1
1
111
1
1
1
1
1
111
2
5
1
1
1
2
1
1
1
11
11
1
1
1
2
2
3
5
3
1
1
1
1
2
2
1
1
1
6
7
8
2
2
2
1
3
811
2
11
1
11
3
6
1
2
4
69
1
1
3
3
3
1
1
1
2
3
11
2
11
11
1
1
RMSE x = 0.8 km
Accuracy = 2.36 km
RMSE y = 1.13 km
N = 245
RMSE x = 0.86 km
Accuracy = 1.85 km
RMSE y = 0.94 km
N = 545
F14
F15
11
1
1
11
1
1
1
1
1
11
1
1
1
1
1
1
1
2
2
1
1
1
1
1
3
5
2
1
1
13
8
3
1
1
1
1
2
2
2
3
10
5
1
1
2
7
8
2
5
9
1
1
3
8
5
1
3
5
2
3
2
1
1
2
11
1
1
RMSE x = 0.74 km
Accuracy = 1.55 km
RMSE y = 0.76 km
N = 601
RMSE x = 0.85 km
Accuracy = 1.83 km
RMSE y = 0.93 km
N = 201
Figure 20.7
Geolocation accuracy of the centroid positions of OLS lights from DMSP satellites F-10, F-12, F-14,
and F-15. The numbers printed on the grid cells indicate the percentage of observations in which
the OLS light centroids were found in that grid cell position relative to the actual location of the light.
The results (Figure 20.9) show that point sources of light were detected in solitary OLS pixels 38%
of the time, in two OLS pixels in 28% of the detections, and in three OLS pixels for 13% of the
observations. This phenomenon was caused by the substantial overlap in the footprints of adjacent
OLS pixels (Figure 20.3 and Figure 20.4).
20.4 CONCLUSIONS
The DMSP-OLS provides a global capability to detect lights present at the Earth's surface. This
chapter provides the first quantitative assessment of the area and positional accuracy of DMSP-
OLS-observed nighttime lights.
Light sources from isolated oil and gas platforms with areas as small as 0.1 km
were detected
2
in this study. Since these platforms are heavily lit, the 0.1-km
area approximates the detection
limits of the OLS for other heavily lit sources. For detection, the aggregated radiances within an
OLS pixel must produce a DN value that exceeds the background noise present in the PMT data.
A larger area of lighting would be required for OLS detection of more dimly lit features than the
2
