What was done
The authors "assessed how well the current day state-of-the-art reanalyses and CGCMs [coupled global climate models] are reproducing the annual mean, seasonal cycle, variability and trend of the observed SAT [surface air temperature] over the Arctic Ocean for the late 20th century (where sea ice changes are largest)."

What was learned
Lin et al. report that "large uncertainties are still found in simulating the climate of the 20th century," noting that on an annual basis, "almost two thirds of the IPCC AR4 [Fourth Assessment Report] models have biases that [are] greater than the standard deviation of the observed SAT variability," additionally noting that the models "can not capture the observed dominant SAT mode variability in winter and seasonality of SAT trends." They also state that "the majority of the models show an out-of-phase relationship between the sea ice area and SAT biases," and that "there is no obvious improvement since the IPCC Third Assessment Report."

What it means
Not only does it appear that state-of-the-art climate models have a long way to go before they can adequately simulate even the past climate of the Arctic Ocean (much less predict its future), we have the word of the six scientists who evaluated them in this study that their creators have made "no obvious improvement" in the models' simulation ability since the time of the Third Assessment Report several years earlier. Does that make what has transpired in the interim something akin to beating a dead horse? Or has it been more like attempting to steer a ship that is dead in the water? However one characterizes the activity, it has apparently gotten us nowhere.