3 posts tagged “global warming”
Carl Zimmer's article Greenup of the Planet Is Not Black-and-White (Wired.com) describes research into the 'greenup algorithm', which determines when plants turn green each spring. Spring temperatures are a major factor - gradually rising average temperatures have led to earlier greenup in most parts of North America, with greenup occurring as much as three weeks earlier than it did in 1982. However, results are not uniform - in some parts of southern North America, greenup has been delayed.
One factor appears to be 'chill units'. Many plants need to go through a period of cold weather while dormant to greenup on schedule the next spring. Currently, areas north of 35 degrees latitude are cold enough in winter to satisfy the required chill units. Warming temperatures in southern regions appear to have reduced the number of chill units below critical values, causing plants to remain dormant when spring arrives until really warm weather finally triggers plants to start growing.
The article is an other example of the complex behavior of plants (see also Smarty Plants: Inside the World's Only Plant-Intelligence Lab). It also points out the dangers of over-simplifying natural systems and how they are influenced by changes such as global warming. Systems often display non-linear behavior, where small changes can result in large effects, or effects may suddenly reverse. As global warming continues, the apparent beneficial effects of longer growing seasons may reverse if winter temperatures become too warm. Unless we learn to deal with complexity, we may use over-simplified generalizations that can lead to poor predictions and decisions.
TVOntario is currently running a series called The Great Warming. The September 19th program had a comment from a Swiss Re (insurance) representative about how a small increase in average temperature can have a dramatic impact on how we perceive weather. In effect, he was trying to relate climate (long term trends) with weather (what we experience on a day-to-day basis).
Over a extended period of time, many observations of natural phenomenon can be represented by a 'normal distribution' graph that matches a specific value against the probability of the value occurring. The most probable values lie near to the average of all values. The 'spread' of the curve is related to the variability of the values (measured in 'standard deviations'), with high variability data resulting in a wider curve. One interesting aspect of normal distributions is that even extreme values can occur, although their probability becomes smaller with distance from the average.
According to Environment Canada data on Toronto for 1971-2000, July was the hottest month with an average
maximum daily temperature of 26.4C (79.5F). The variability is 1.2 standard deviations (calculated based on the daily average temperature). A normal distribution based on these values looks like the curve below (click on the image to enlarge it), with the temperature in Celsius at the bottom, and probability (where 1.00 is certainty) on the left.
Using a normal distribution allows us to make predictions. For example, the area below the curve to the right of a value tells us the likelihood of experiencing days with maximum temperatures exceeding that value. Using the curve, the probability of 30+C (86+F) degree days is quite small at about 0.13% (the area shaded red is barely noticeable). Note that this curve is an approximation of the real temperature data - from personal experience, the actual probability is higher. What we are interested in is the rate of change in the likelihood of having hot days with increasing average temperature.
If the average maximum temperature increases by 1 degree to 27.4C (81.3F) and the variability remains the same, we can shift the curve one degree to the right and again estimate the probability of 30+C degree days. The following graph shows the result, at around 1.5%. This represents more than a 10-fold increase.
Increasing the average temperature by another degree to 28.4C (83.1F) results in the following curve. The probability of 30+C degree days has increased to 9%, a 6-fold increase. Similar increases in probability would be found if we were looking at 33+C or 35+C degree days.
There is evidence to suggest that global warming not only increases average temperatures, but also the variability. Increasing variability by 25% increases the spread of the curve, and results in a probability of 30+C days of 14%. Note that the curve is wider both at the high and low ends - a higher probability of both hotter and cooler days.
Again, these normal distribution curves are an approximation of the real temperature data - considerable research and data modelling is required to estimate actual probabilities of unusually hot days. The intent was to show how even small increases in average temperatures can result in significant changes in our perception of the weather.
Technology Review has compiled a series of articles on global warming.
"Readily available energy technologies could be put in use today to forestall global warming. Technology Review examines some of these technologies and argues that they require not further refinement but a considered, long-term deployment strategy."
The articles describe both the causes and potential solutions for global warming. Noticeably absent is a discussion of how we can reduce the rising demand for energy through conservation and efficiency.
