If I had a dime every time I heard someone cite natural variability as an explanation for climate change symptoms, I’d probably be able to buy my dog that $300 dog bed she eyes every time we are in the pet store. Quips aside, natural variability is one of the most common skeptical positions out there; by natural variability, I’m referring to the dynamic ups and down of environmental conditions that are part of long-term and short-term regional and global climatic cycles. Natural variability encompasses phenomena like glacial to interglacial transitions and ongoing El Nino/La Nina climate oscillations. I’ve heard friends, relatives, strangers, and especially politicians state that the climate change effects seen around the globe are all within natural variability. Frankly, it’s a cop out: data abound that say that climate change is real, that temperature and other environmental conditions are outside of natural norms, and that humanity has a lot to lose by playing the ostrich with our heads in the sand.
Enter the research from a global team of climate scientists. Stephanie A. Henson and colleagues from around the world analyzed statistical models to paint a clearer picture of how climate change may impact key ocean health indicators in the coming decades. This recent work is essentially a one-two punch to proponents of the “natural variability” argument. Henson and her team used ocean-atmosphere circulation models created and checked by the global climate science community through the International Panel on Climate Change (IPCC). The Henson-lead team specifically used models created under the auspices of the CMIP5 exercise. Statistical models take existing data, make some assumptions about unknown conditions in the model environment, create sets of equations that “fit” or describe the existing data, and then use the produced equations to forecast future conditions. It’s important to reiterate that the forecast is based on historical observations and trends, so good historical data inputs are vital to create a decent model. The authors used two types of models for this research; one was a “business as usual” model which assumes no change in human greenhouse gas emissions, while the other factors in some human action to mitigate or curb greenhouse gas emissions and the effects of climate change.
Running models, not of the Nike variety
The models were run multiple times to produce results describing time trends of climate change symptoms in 1º latitude x 1º longitude boxes covering the entire globe. The model results (aka forecasts of future conditions based on past observations) were then analyzed to gauge the timing and pace of modeled climate change indicators in each 1º x 1º box.
The authors analyzed four climate change drivers in each assessed region; a climate change driver is an environmental condition, anticipated to change due to human-caused emissions, that determines the food web structure and health of a given area. The authors selected pH, oxygen concentration, temperature, and primary production level as the drivers they were interested in. The authors analyzed the 1º x 1º boxes for time of emergence and pace of emergence for each of these drivers, using seasonal ranges as points of comparison.
Time of emergence (ToE) is defined as the year when a given driver exceeds the seasonal average extreme consistently for the remainder of the observation period.Ffor example, if box A experiences a seasonal average maximum temperature of 90 º F for fifty years, then starts displaying temperatures throughout the 90’s during years 50-60, then displays maximum average temps of 100+ º in year 60 onwards, the time of emergence would be considered year 60. By defining time of emergence based on observed seasonal ranges, the authors account for the range of natural variability typical for each geographic area across the world ocean. Organisms are accustomed to dealing with seasonal extremes in the area that they live but anything beyond this can be stressful or even lethal. This is a particularly interesting way to look at the projected impacts of climate change, because it explicitly tackles the natural variability argument many skeptics cherish.
Henson and colleagues also define pace of emergence as the time period between when a driver first exceeds the seasonal average extreme of a given area and the time of emergence. So for the theoretical box A, the pace of change would be around ten years, as max temperatures first exceeded the historical norm in year 51 and then consistently exceeded the norm starting at year 60.
Number crunching nets some neat/nasty results
After combing through the model results for each region, the authors came up with some startling results. Wider swaths of the ocean display a slower pace of change and a later time of emergence with mitigation efforts, while the climate change drivers depart from seasonal variability more quickly under the business as usual scenario. The authors point out that a slower rate of change is vital to maintain marine life. Ocean flora and fauna are capable of adapting to ambient conditions, yet this takes time; fossil records and modern scientific evidence demonstrates that fast changes in environmental conditions typically kill organisms off, hence why we are not dodging dinosaurs on our daily commutes.
The results also indicate that multiple drivers are likely to emerge in each area and overlap; this will likely have unknown consequences on marine life. Under the “business as usual” model, which assumes we humans don’t take action to curb our greenhouse gas emissions, one or more of the examined drivers across 55% of the ocean emerge from the seasonal trend over the next 15 years. By 2050, 86% of the ocean will likely display multiple drivers that exceed natural variability. Intuitively, the emergence of multiple drivers doesn’t bode well. Think about how humans react to environmental stress; picture dealing with 100º F temperatures on a hot day with no shade…sounds pretty sucky right? Now imagine that the oxygen content of the air drops below normal, from 21% to maybe 17%. With this added stressor, a typical human would likely throw up, lose mental acuity, or pass out. Moral of the story: multiple environmental stressors pile up and are much harder to cope with compared to dealing with just one stressor at a time; this scenario is going to be reality for many ocean creatures under the business as usual model.
Why should I care about model results?
Overall, this research is a fantastic statistical tool demonstrating that natural variability is no longer a viable excuse to explain away the abnormal temperatures and other symptoms evident across the globe. Climate change is already underway and is restructuring marine environments. Change in marine environments matters to humans as we derive an inordinate amount of economic value and ecosystem services from a functioning ocean. Over 900 million people are projected to live in a low elevation coastal zone by 2060, 1 in 7 people currently get their protein from marine life, and the entire globe enjoys the climate-regulating benefits of ocean circulation and carbon dioxide uptake. This scientific study also reiterates how much mitigation efforts could help deter climate change catastrophe. How can you help? Push for climate change regulation, call your local and regional politicians, and work to avoid the dire 2100 predictions of Henson et al. under the business as usual model.
I am a third year PhD student at the University of Rhode Island Graduate School of Oceanography in the Lohmann Lab. My current research interests include environmental chemistry, water quality, as well as coastal and seabird ecology. When not in the lab, I enjoy diving, surfing, and hanging out with my dog Gypsy.