Background
The authors state that it has long been assumed that "deteriorating climate" -- defined as either cooling or warming -- "could shrink the carrying capacity of agrarian lands, depriving the human population of sufficient food," with "population collapses (i.e., negative population growth)" the unavoidable consequence; but they say that "this human-ecological relationship has rarely been verified scientifically," noting that at the high end of the temperature spectrum, "evidence of warming-caused disaster has never been found."

What was done
In a study designed to fill this research void, Zhang et al. performed time-series analyses to examine the association between temperature change and country-wide/region-wide population collapses in different climatic zones of the Northern Hemisphere (NH), focusing on all known population collapses over the period AD 800-1900. In addition, they say that regressions were run to estimate the relative sensitivity of population growth in the NH to climate change, where the independent variables employed were time and temperature anomalies.

What was learned
Of the 88 NH population collapses identified, fully 80% of them were caused by cooling, while 12% occurred during what the six scientists called "mild conditions," and only 8% of them were caused by warming. Thus, for the NH as a whole, it was no surprise that they discovered that "temperature was positive and highly significant in the regressions in which a 10% increase in temperature produced on average a 3.1% increase in population growth rate."

What it means
Historically, and for the Northern Hemisphere as a whole, warming and warmer times have most often been good times for humanity, as exemplified by the greater numbers of people the earth was able to support under such conditions, while cooling and colder times were typically just the opposite, with many significant population collapses, caused by what Zhang et al. describe as "Malthusian checks (i.e., famines, wars and epidemics)."