While satellite data provide significant advantages over conventional surface-based observations, there are also important limitations in how the data can be used to infer information about the state of the atmosphere. Specifically, information about the atmosphere is obtained indirectly by measuring the properties of electromagnetic radiation arriving at a space-borne sensor. That is, weather variables, such as precipitation or surface temperature, need to be inferred from quantities that provide a proxy for the weather variables of interest.
For instance, the amount of precipitation can be inferred from the amount of energy (radiation) that is emitted when water vapor condensates in the atmosphere and errors are therefore inherent to any satellite-based estimate of precipitation1. Another important source of error is the contribution of the underlying land surface (e.g., vegetation) to the electromagnetic signal measured by a satellite-based instrument.
Some of the important advantages of satellite data include:
The high vantage point provides a broad field-of-view and permits the observation of large-scale weather phenomena in one single view.
Satellite data provide measurements of atmospheric observations in areas of the globe that are less equipped with surface-based stations.
The continuous observation permits the monitoring of and warning for short-lived phenomena such as tropical cyclones or snow storms.
Advanced communication systems are an integral part of satellite systems and provide quasi real-time access to weather data.
In situ measurements at weather stations, on the other hand, can be directly interpreted, but are more prone to be influenced by local factors and may therefore not be representative for larger areas. Also, the density and distribution of surface-based observations is inhomogeneous across the globe and particularly sparse in less developed (and less populated) areas and over ocean surfaces.
It is important to note that direct measurements from surface stations are indispensable to:
(i) provide accurate and precise measurements of meteorological variables that cannot easily be derived from remote sensing
(ii) monitor small-scale phenomena
(iii) provide independent validation and calibration data for satellite estimates
Satellite observations, on the other hand, are critical to the generation of warnings and forecasts of hazardous conditions such as storms, tropical cyclones or high winds and are more difficult to be derived from station data. (Satellite-derived temperature data is used to measure the duration of cold cloud tops over a region for the determination of accumulated rainfall (3mm of precipitation for each hour that cloud top temperatures are measured to be less than 235 K, see Arkin, P. and Meisner, B. “The Relationship between Large-Scale Convective Rainfall and Cold Cloud over the Western Hemisphere during 1982-84,” American Meteorological Society Monthly Weather Review Vol. 115, Issue 1. Pp/ 51-74 (1987).)
More than 90 percent of the volume of data assimilated in global numerical weather prediction models today comes from space-based systems. Thus, the strengths and weaknesses from surface and space-based meteorological observations are complementary and satellite-based systems are not a substitute for observations collected on the ground.
– Direct weather observation (accurate/precise)
– Narrow measuring footprint
– Long-term climate records
– Local sampling
– Indirect weather observation (proxy)
– Synoptic broad field of view
– Short records (< 10-15 years)
– Continuous global coverage
Source – World Bank