This week my Twitter feed directed me to an article by Dana Nuccitelli in the Guardian titled: “New study undercuts favorite climate myth ‘more CO2 is good for plants”. The Guardian article was about a new study out of Stanford, published in the Proceedings of the National Academy of Sciences (PNAS). The intention of this blog post is to explain why this PNAS article does not, as Mr. Nuccitelli claims, “undercut” the “climate myth” that “more carbon dioxide is good for plants”.
As anyone who has studied any horticulture knows, it is an accepted fact in the academic literature that elevated concentrations of carbon dioxide are expected to result in increased yields and improved growth in the majority of green plants. There is a commonly accepted proviso to that axiom, which is that increased carbon dioxide concentrations will have little effect on plant growth in conditions where there is an established alternative limiting factor in that growth. So adding carbon dioxide to a plant that is lacking in essential nutrients will not increase growth rate as carbon dioxide does not represent the rate-limiting step in that growth. That doesn’t mean that the increases in carbon dioxide concentrations are not good for these plants (by decreasing water stress etc…), but simply that increasing concentrations of carbon dioxide will not increase crop yields when some other feature limits plant growth.
The effect of increased carbon dioxide concentrations on plant growth is not contentious. Even the folks at Skeptical Science admit as much (admittedly they only do so briefly before quickly changing the subject to start talking about how more plant growth will use up all the soil nutrients and will be likely to increase pest effects both of which are irrelevant to the science under discussion). So what Mr. Nuccitelli calls a “climate myth” in his Guardian article (that more carbon dioxide is good for plants) actually represents a demonstrated scientific fact confirmed in thousands of academic studies and used to increase crop yields in tens of thousands of greenhouses worldwide. Now at least three scientists have had the expression “extraordinary claims require extraordinary evidence” attributed to them, but regardless of who said it first, this axiom is very much applicable in the context of this report. Given the extraordinary claim made by Mr. Nuccitelli I was really interested in establishing what is so special about this PNAS study that allows it to “undercut” the globally accepted fact that increased carbon dioxide concentrations are generally good for green plants.
The PNAS article in question has a particularly unsexy title: “Nonlinear, interacting responses to climate limit grassland production under global change. It details the results of a “17 year study of California grassland exposed to full-factorial warming, added precipitation, elevated CO2, and nitrogen deposition”. Let’s start with the praise. Any 17-year scientific study deserves an award just for perseverance and this one is no exception. In the study the authors planted 132 plots of local, grassland plants and then varied how the plants were exposed to carbon dioxide, temperature, water and nitrogen. Then over 16 years they measured above-ground and below-ground net primary productivity (ANPP and BNPP, respectively) in the plots. The results of the study indicated that, in this one study, increased carbon dioxide concentrations resulted in no significant increase in net primary productivity (NPP); increased nitrogen resulted in significant increases in NPP while increases in temperature and precipitation had initially a positive, but ultimately a net negative effect on NPP. Now I was a bit surprised by the conclusion until I noticed that the study was undertaken in a Mediterranean climate biome. This cleared up a lot of confusion for me as I recognized that the results, while interesting, were likely not globally applicable as I will explain below.
The Stanford test site used in the study is located at the Jasper Ridge Global Change Experiment (JRGCE) site in the Jasper Ridge Biological Preserve in California. The site is “located on the east side of the Outer Coast Range of central California near the Stanford campus. Jasper Ridge has a Mediterranean climate with cool, wet winters and very dry summers (ref)”. Now this should immediately perk up your ears since the Mediterranean climate biome is a relatively rare one. Put another way, it is hard to overstate how small the area defined as having a Mediterranean climate really is from a global perspective or how minimally this area (in its natural state) contributes to global NPP. As the JRGCE web site points out: areas with Mediterranean climate are known for being wet and cool in winter and very dry and hot in summer. This means that growth is limited to a very short effective growing season and one that occurs primary in the winter time (as the authors point out, between November and May) when the days are shorter and the sunlight intensity is lower. Plants in Mediterranean climate zones are optimized to survive long droughts in summer and wet winters and include all the classic Mediterranean trees and vines like olives, grapes, figs, etc… (i.e. plants that store winter rains in their fruit and ripen in the hot summer sun) as well as grasses and small annual flowers. That is not to say that areas with Mediterranean climate can’t be very good for growing other crops, but this is thanks to complex irrigation systems intended to extend the growing season into the hot, dry springs and summers. The farms in these areas thus take advantage of the plentiful natural spring/summer sunlight and the availability of industrial fertilizers in addition to the availability of water for irrigation.
As described in the PNAS paper, Mediterranean climate areas tend to be nitrogen limited for plant growth. To explain why, recognize that the climate regime does not encourage the growth of nitrogen fixing plants like legumes that struggle to grow/survive in the hot, dry summers. They instead favour grasses, small seasonal annual flowers and drought-resistant trees and vines. Thus the inclusion of nitrogen in this study presents an obvious way to find a limiting nutrient that would be expected to increase NPP. By including nitrogen in the study the authors have essentially included a positive control that demonstrates that, through their action, they can increase NPP.
The major concern with the Mediterranean climate, from a crop-growing perspective, is that a most of the rain comes in winter, often in short bursts. During the wet, growing season water is not a limiting feature in plant growth, rather the availability of strong sunlight and critical nutrients limit growth. This makes the way the study adds water a bit odd. As described in the paper, additional precipitation was not provided at the beginning or tail end of the rainy season (to extend the growing season) but rather the elevated precipitation was mostly added during, or immediately after, rainfall events. As the study put it: “The precipitation treatment was +50% of ambient rainfall, plus two 10-mm additions after the last rainfall event”. Thus, the increase in precipitation was in individual event intensity and not overall number of wet days. So in this study they took the expression “when it rains it pours” quite literally. Unfortunately this creates an obvious weakness: too much of a good thing. Too much rain will result in excess leaching of nutrients and, as the study showed, the wettest years had the worst NPP results. As well wetness would negatively correlate with sunlight intensity since the sun is not shining brightly when the rain is falling.
It is quite sad that such an important piece of research was placed in such an inopportune location. Doing a study of this kind in an area with a Mediterranean climate basically guaranteed the results observed. As described in the UC Davis web page on the region it is a combination of water stress, nutrient availability and winter sunlight intensity that determines growth in Mediterranean climates. Or put in a way that matters to readers of this blog, carbon dioxide availability would not be expected to represent a limiting feature in the growth patterns in this biome. Adding carbon dioxide would not be expected to have a significant effect on growth rate in this biome and not surprisingly, that is precisely what they saw. So the outcome of the study is confirmation of what any horticulturalist would have told you. That being said, confirmation of theories is one of the tasks we expect our universities to undertake and as such this study would be considered a success.
So let’s go back to the Guardian article in question which claims that the conclusions of this study debunk the “climate myth” that additional “carbon dioxide is good for plants”. Does the study do this? Well the answer to that is a categorical no. Rather the PNAS study demonstrates that in a Mediterranean climate water stress and winter sunlight intensity represent the biggest determinants for plant growth and that in grassland biomes nitrogen often represents the limiting nutrient in plant growth. The study also demonstrates that there can be too much of a good thing (i.e. rainfall in the wet season). The study says nothing about whether carbon dioxide is good for plants…don’t even get me started on what “good for plants” really means anyways.
So going further, does the PNAS study counteract the plentiful studies that show the greening of the northern hemisphere attributed to the increases in global carbon dioxide concentrations? No it does not. Does the PNAS study allow one to dismiss the comprehensive reviews that demonstrate increased yields associated with increased atmospheric carbon dioxide concentrations? No the study does not.
Now don’t get me wrong here, I am not claiming that adding carbon dioxide to the atmosphere at the rate we have over the last century is a good thing. What I am saying is that trying to extrapolate a single study from a geographically tiny biome to upend a generation of research on the effects of increased atmospheric carbon dioxide concentrations on plant growth might be a step too far, even for the most alarmist of authors. In this case Mr. Nuccitelli, in his Guardian article, has taken a gigantic and unsupported leap from the conclusions (and limitations) of this report. He has taken a small a study in a biome of limited global significance and extrapolated it to global proportions and come up with global conclusions that are not supportable using the data presented. This study does not undercut any “climate myths”.