Sunday, February 01, 2015

A Long-Term View of Snowpack in New Mexico

By Sierra Rayne

One of the latest climate change talking points is snow. If it isn't climate change causing too much snow in the Northeast, it is causing too little snow in the West. Whatever the problem -- no matter how internally incoherent and contradictory -- climate change is the problem.

At the New Mexico Daily Lobo (the newspaper of the University of New Mexico), there is an interview with "a professor of Earth and planetary sciences specializing in large-scale climate change, about the state of climate change today and what the future looks like for New Mexico" in which the recent declines in New Mexico's snowpacks are discussed:

"I was a lead author on the IPCC (Intergovernmental Panel on Climate Change) report that came out in late 2013 -- that is assembled by the United Nations every six or seven years -- to assess climate change research...Among the things we're studying in my research group is the ongoing decrease in snow pack that is certainly predicted to occur as the climate warms up. We're seeing evidence of that in the data, and that has consequences for stream flow in a significant way. As snow pack at high elevations decreases, we have shorter snow seasons, the snow melts earlier, and evaporation rates go up. That's what we've been seeing for the past several decades."
If one looks at late season snowpack data for New Mexico, it is clear that some of the stations have significant declines in their snowpack over the last three decades on the benchmark beginning of March and April measurement dates.

But what needs to be discussed is our very poor understanding of long-term trends in snowpack, because this will tell us about whether the recent declines in some regions are part of a cycle and something not as unsual as very short-term records might suggest. The long-term view generally helps temper the short-term alarmism.

Of the 58 snowpack monitoring sites for New Mexico in the USDA National Water and Climate Center database, the average date of installation is 1982. Just 13 of the sites have snowpack records extending back before 1970, and only 7 of the sites provide snow data dating back to at least 1950. Thus, we have a very incomplete record of long-term snowpack data in the state. The dataset is heavily biased towards just the last few decades -- of which the snowpack tended to be higher than the historical average in the late 1970s and early 1980s. This doesn't do much to inform us about the causal relationships for trends over this short timespan, and is likely to lead to spurious conclusions.

No comments: