What was done
In a study published back in 2001, Rustad et al. employed meta-analysis "to synthesize data on the response of soil respiration, net nitrogen mineralization, and aboveground plant productivity to experimental ecosystem warming at 32 research sites representing four broadly defined biomes, including high (latitude or altitude) tundra, low tundra, grassland, and forest."

What was learned
In the words of the 45 researchers involved in the analyzed studies, "there was a diversity of responses among sites, with experimental warming increasing plant productivity at 13 of the 20 sites for which data were available, decreasing plant productivity at two of the 20 sites, and having no significant effect on plant productivity at five of the 20 sites," the end result of which was that "the weighted mean increase in plant productivity due to warming was 19%." In addition, they report that "the response was greatest in the colder ecosystems characterized by lower mean annual precipitation."

What it means
The international team of scientists concluded that "the warming-induced increase in plant productivity may be a direct effect of either increased rates of photosynthesis at higher temperatures or (in those experiments with year-round warming) longer growing seasons, or an indirect effect of increased nutrient availability, resulting from increased rates of litter decomposition and nitrogen mineralization," the latter of which phenomena was increased by an average of 46% by the experimental warmings they analyzed. In addition, they note that "both direct and indirect effects of warming could be particularly important in higher latitude arctic ecosystems, which tend to be both temperature and nutrient-limited."

These several results, which the authors describe as "the most comprehensive synthesis of ecosystem response to experimental warming to date" -- which date was 2001 -- tend to suggest that warming is much to be preferred over cooling in terms of the temperature change's impact on ecosystem productivity.