So I consider surface station data both temporally and spatially sparse, subjected to manipulations that are less than forthright and in ways that are mathematically dubious.
I don't get that you would think the adjustments to surface temperatures are less than forthright. They are well documented in the relevant papers about the adjustments.
Here is a NOAA page on their temperature data.
This page at Berkeley Earth describes their data set and some of the adjustments.
As far as the sparseness of some regions of the globe goes we don't care so much what the actual temperature is globally as much as we care about how it's changing over time. If the global temperature is derived in a consistent manner then it's probably a reasonable representation of temperature change.
Satellites because of their orbital inclination don't cover the polar regions at all and have to deal with issues of observation angle for near polar readings. Also they have to make adjustments for clouds and high elevations messing up their readings.
I don't disbelieve the satellite readings, I just don't see any good reason to trust them over the surface measurements, particularly over the time period since the satellites went up (1979) when the surface systems have also been improved to higher standards.