We don't have experience with losing the infrastructure in a country with 300 million people, most of whom don't live in a way that provides for their own food and other needs. We can go back to an era when people did live like that...10 per cent would be 30 million people, and that is probably the range where we could survive as a basically rural economy.
A good mental model for the next century.
The geophysical algorithms specifically designed for use with the
Nimbus-7 SMMR for obtaining sea surface temperatures, near-surface
winds, atmospheric water vapour, cloud liquid water, rain rates, snow
cover, sea-ice concentration, multiyear ice fraction, and ice
temperature are summarised in Gloerson et al (1984).
In the Arctic, the presence of a perennial ice cover further
complicates the problem of determining sea-ice concentration. The
presence of sea ice ranging in age from newly formed thin ice to thick
multiyear ice results in ambiguous microwave signals, which pose
additional difficulties in accurately calculating sea-ice
concentration. The problem of determining ice concentration in the
presence of a mixture of radiometrically different ice types is
essentially the problem of determine the ice-type distribution. Some
attempts in differentiating first-year and multiyear ice cover in the
Arctic using the single-channel ESMR-5 data have met with only limited
success (Carsey 1981; Campbell et al 1983). The accuracy of computing
sea-ice concentration when two radiometrically different ice types are
present and when significant spatial and temporal variations in ice
temperature occur has been improved with the use of multispectral
radiances from the SMMR.
[Goes on to describe an algorithm to process Nimbus-7 SMMR data and
determine concentration of sea-ice and multiyear fraction -- the
amount of ice that has survived at least 1 summer.]
-- Barry Saltzman (ed), Advances in Geophysics, Vol 27