Posts Tagged ‘Model’
If you’ve been on the internet at all in the past week, you’ve probably seen these lovely images from NASA, visualizing the height of tree canopies around the world. They’ve been on science sites along with art ones. In a sense, that alone is useful: using beautiful visuals to make people think about the world on a larger scale. But where did these data come from, and what do they really mean?
Why is it important?
The two main sources of anthropogenic influence of atmospheric carbon are (1) the burning of fossil fuels, releasing carbon dioxide into the air and (2) deforestation, removing trees which store large amounts of carbon. 20-50% of carbon in the atmosphere is currently not accounted for in climate models – a huge amount. Knowing where exactly this carbon is coming from is important for both conservation and making socioeconomic decisions regarding energy use.
Some scientists hypothesize that these unaccounted-for changes in carbon flux are due to poor knowledge of forest stands – both where deforestation has occurred, and where forests are recovering after previous deforestation. By knowing the size of forest stands, as larger trees can store more carbon, and keeping track of changes in the sizes of these stands, the hope is to have more accurate models for carbon storage and atmospheric carbon.
In the face of climate change, these kind of data are also useful, if updated over time, in seeing how rising temperatures and increased carbon dioxide affect tree growth (and thus carbon sequestration). Since plants get their carbon from the air, it seems natural that increased carbon dioxide would increase tree growth. Thus canopy height maps can help us to test this hypothesis.
From where did these data come?
Most of the data comes from work by Michael Lefsky et al. in a 2002 paper pubslihed in Global Ecology and Biogeography and a 2005 paper published in Geophysical Research Letters. In the 2002 paper, Lefsky and his team measured canopy height using LIDAR (Light Detection and Ranging technology), which essentially sends down a laser beam from which distance can be measured based on return time. LIDAR uses a shorter wavelength than typical radar, making it more sensitive to smaller objects such as particles in the air. Thus the name, Light Detection and Ranging. It also has a narrower beam than radar and thus is more specific in its measurements. In the paper, Lefsky and co. took LIDAR measurements on temperate forest stands and compared them with field measurements of forest height, finding that they matched up very closely. They also were able to do field measurements on the complexity of forest stands, meaning how much undergrowth lives beneath the canopy, and were able to create an accurate equation for predicting biomass of a forest stand based on LIDAR alone.
The 2005 study was another sort of proof-of-purpose paper, once again showing that LIDAR can predict above-ground biomass. The authors did not have complexity variables this time, but were still very accurate in their predictions.
For the beautiful figures of canopy height released by NASA, Lefsky combined his own data with that from NASA’s MODIS (Moderate Resolution Imaging Spectroradiometer), which is housed on a satellite and images the earth’s surface every 1-2 days. Combining the data on terrain from MODIS with his own canopy height data, Lefsky crafted these images over several years.
What does it mean?
Previously, large-scale vegetative modeling was only able to be measured in 1 km swaths of land, done by MODIS alone. Due to deforestation, there is a great deal of variation in canopy height on a smaller than 1 km scale – and finally we have the tools to create maps of forest canopy and thus help us better track carbon storage at these sites. The map isn’t perfect – it is a model after all – but it is far more accurate than anything we’ve seen yet. Expect interpretation from Lefsky and others in the near future.
But this isn’t the end.
As I mentioned earlier, traditional thinking assumes that forest growth will assist in carbon storage through increased growth due to higher carbon dioxide in the atmosphere. But a recent paper (July 21, 2010) in PLoS ONE by Lucas Silva (“silva” means forest in Latin, lollers), Madhur Anand, and Mark Leithead suggests otherwise.
A major tradeoff for growth in trees (as I’ve discussed elsewhere) is that the opening of leaf stomata (pores) to absorb carbon dioxide also causes water loss through evaporation from these same stomata. The authors of the PLoS paper looked at tree growth through tree rings and compared it with isotope analysis to measure water loss. If the trees they studied had increased growth due to increased carbon dioxide, they would also expect more water conservation, as more carbon dioxide would be able to enter the leaves without as much water loss.
Studying 4 species of tree at 4 forest types in Canada, they found a 53% increase in water use efficiency over the last century. (They looked at both young and old trees to account for varying growth rates and energy use.) This seems like good news – the trees are absorbing carbon dioxide more readily. However, they also saw a decline in growth overall. This suggests that other stresses, such as water, nutrients, and temperature, are limiting their growth despite the ease of access of carbon dioxide.
The next step: can we learn about tree growth from Lefsky’s maps? Are they accurate enough? It would be great to measure biogeochemical measures, such as water use efficiency, and compare this to large-scale forest size data. A girl can dream…
Okay, class: what have we learned?
Lasers are cool! The LIDAR technology, originally created for studying atmospheric chemistry, reapplied to study canopy heights has allowed us to visualize our forests in a new way. (And make some beautiful pictures.) There was a lot of work put into it – and to accurately measure how our forests are changing, increasing work will have to be done to keep the maps updated to create an index of canopy height on our planet.
However, we’ve also learned that we cannot necessarily rely on traditional hypotheses in times of climate change. While trees have the capacity to remove carbon from the atmosphere and store it, other factors can confound these effects, as we read in the PLoS ONE paper. While more work certainly needs to be done on this front (using large-scale climate measures for growth instead of dendrochronology, for example), their results are certainly sobering.
So, as usual, we need to do more work! We need to learn more! We have to challenge our hypotheses, and challenge new results that support or disprove them. It’s always easier when you have a mystery to solve: where is all that carbon anyway?
Cohen, W., Harmon, M., Wallin, D., & Fiorella, M. (1996). Two Decades of Carbon Flux from Forests of the Pacific Northwest BioScience, 46 (11) DOI: 10.2307/1312969
Lefsky, M., Cohen, W., Harding, D., Parker, G., Acker, S., & Gower, S. (2002). Lidar remote sensing of above-ground biomass in three biomes Global Ecology and Biogeography, 11 (5), 393-399 DOI: 10.1046/j.1466-822x.2002.00303.x
Lefsky, M., Harding, D., Keller, M., Cohen, W., Carabajal, C., Del Bom Espirito-Santo, F., Hunter, M., & de Oliveira, R. (2005). Estimates of forest canopy height and aboveground biomass using ICESat Geophysical Research Letters, 32 (22) DOI: 10.1029/2005GL023971
Running, S. (1999). A Global Terrestrial Monitoring Network Integrating Tower Fluxes, Flask Sampling, Ecosystem Modeling and EOS Satellite Data Remote Sensing of Environment, 70 (1), 108-127 DOI: 10.1016/S0034-4257(99)00061-9
Silva, L., Anand, M., & Leithead, M. (2010). Recent Widespread Tree Growth Decline Despite Increasing Atmospheric CO2 PLoS ONE, 5 (7) DOI: 10.1371/journal.pone.0011543
I have a tendency to root for the underdog. I rooted for the Phillies throughout the 90s, when my heroes Lenny Dykstra and Darren Dalton could rarely lead them to a win. It’s a mixture of a desire for upheaval, that the unexpected can happen, as well as pure sympathy for the ones who always lose.
Do you know who always loses in science? Dirt. No one cares about it. I mean, it’s a mixture of poop and rotting plants and animals. It harbors fungus and worms and bacteria. And while it is generally accepted that it has an important role to play, it tends to be overlooked because it’s just not all that exciting on the surface. Why study ground up brown stuff when you can study WHALES or CANCER?
That’s why I love it when dirt wins, as it does in this early-access article from PNAS entitled “The impact of soil microorganisms on the global budget of δ18O in atmospheric CO2” (doi:10.1073/pnas.0905210106). The closest I’ve heard soil come to being included in the climate debate is the possibility of pumping liquid carbon dioxide deep beneath the earth’s surface to sequester it: not the most dignified of positions. But this article provides evidence that it is more involved than that, and helps to mollify some discrepancies between prior models and observed measurements in carbon dioxide.
One of the most common ways to trace the origins of oxygen in the atmosphere is through isotope analysis. Carbon dioxide cannot dissolve in water on its own, but needs to be made into the ions HCO3+ and H+ so that it can be transported in fluids. An enzyme, carbonic anhydrase (CA), switches carbon dioxide between its dissoluble and soluble forms, and is found in both plants and animals. It is an incredibly important enzyme for both respiration and photosynthesis. If CO2 were ionizing on its own, without an enzyme, it would take far longer, and the systems would be far less efficient. However, it leaves a mark: a heavy isotope of oxygen. The new CO2 molecule, built by CA, adds 2 more neutrons to oxygen, creating a δ18O isotope which we can trace, thus tracing the activity of CA.
Since δ18O is a heavy isotope compared to oxygen-16, the normal form, it preferentially remains in leaf tissues during transpiration and evaporation. Eventually these leaves die and fall to the soil, where they are broken down. The amount of δ18O in the soil has traditionally been used as a measurement of plant photosynthesis. The possibility of CA activity in soil has been disregarded bcause of the high levels of δ18O in just the top few centimeters of soil, indicating that it is due just from decomposing leaves.
The 18 authors of this paper decided that this assumption wasn’t good enough. What if microorganisms are creating δ18O in the soil due to their own CA activity? What would this mean for the overall oxygen budget? First of all, it would mean that plant photosynthesis would have a lesser role. It could also change the estimations of photosynthesis vs. respiration in our atmosphere, since the microbes could be either photosynthetic algae or cyanobacteria, or respiratory little buggers.
The authors took the measurements of δ18O at different soil depths from 7 different major earth ecosystems from the field, and also created the artificial conditions in chambers with to determine if δ18O levels differed between the two. They also used this “chamber-flux” data to estimate different rates of δ18O creation under different CA catalyzation levels. These data showed that naturally measured δ18O levels were greater than the control levels without CA: up to 300x in the more productive ecosystems! This provided clear evidence that δ18O is being created by soil microorganisms through CA enzymatic activity on their own.
This information was consistent with previously observed and modelled δ18O curves, shown in the figure above. The top half shows the observed δ18O levels in dark blue dots, with the modelled line in black, with the frames increasing in CA activity from left to right. In the right frame, with CA activity at 300x the left frame, the modeled and observed δ18O creation rates overlap. (The curve is based on latitude — northern latitudes, with much vegetation and high photosynthesis on the left, decreasing in photosynthetic production as we move southwards.) This provides more evidence that CA is present in soils, as in an ideal world, observations and models will match up.
The bottom half of the figure is based on the concept of isoflux. This is a measurement of CO2 in the atmosphere, with positive values indicating photosynthesis, while negative values indicate greater respiration, which removes CO2. The “soil invasion” line, in orange, goes from showing no-change in the no-catalyzation scenario, to absorbing nearly as much CO2 as respiration.
So, really, what is this paper saying? First of all, don’t ignore the dirt! Soil microbes may be small, but they are vast in number and can really have an impact on our element cycling. More than anything, this paper helps to adjust previous models. It suggests that the soil may be more of a carbon dioxide sink that we previously thought, because we now have evidence that respiration is taking place due to this increase in δ18O from CA use.
To me, what this paper really shows is how little we know. We’re trying to model oxygen and carbon in the atmosphere and earth, and there’s so little way of knowing. If it weren’t for this enzyme, carbonic anhydrase, that happens to incorporate a heavy oxygen isotope, where would we be? Modelling is important, don’t get me wrong. But it is also incredibly frustrating because we really don’t know enough to create very accurate models. This paper is a little slice, sure; but we could be missing huge impacts just because they are untraceable.
Wingate, L., Ogee, J., Cuntz, M., Genty, B., Reiter, I., Seibt, U., Yakir, D., Maseyk, K., Pendall, E., Barbour, M., Mortazavi, B., Burlett, R., Peylin, P., Miller, J., Mencuccini, M., Shim, J., Hunt, J., & Grace, J. (2009). The impact of soil microorganisms on the global budget of 18O in atmospheric CO2 Proceedings of the National Academy of Sciences DOI: 10.1073/pnas.0905210106
You may have heard that we’ve been having a bit of a problem called “global warming” or “climate change.” The debate is what to do about it — can individuals, day-by-day, affect the amount of greenhouse gases in the atmosphere based on decisions involving diet, waste, and choices of consumption? What types of alternative energy are the most efficient and effective? How does industry need to change in order to yield or reduce carbon emissions? Is this a problem that we can actually solve?
The journal Climatic Change published an article online on November 21 by Timothy Garrett entitled “Are there basic physical constraints on future anthropogenic emissions of carbon dioxide?” (open access: doi:10.1007/s10584-009-9717-9). You could easily blow this off as just another doomsday scientist — but the way he structures his argument, stepping back from the issue and thinking about human civilization in relation to its environment in a new way, makes it stand out.
Garrett creates a new economic model, essentially reducing civilization to production and energy consumption. The standard model used in the International Panel on Climate Chance (IPCC) Special Report on Emissions Scenarios (SRES) includes the variables p (population) and g (standard of living), which are difficult to predict, causing difficulty in creating reliable models to calculate the climatic state even 50 or 100 years from now. Garrett argues that these variables are unnecessary, as they are simply responses to energy consumption and efficiency; that we should instead think about civilization as a huge furnace, which needs more energy as productivity increases, but is also inextricably linked to past production. “The present and future are influenced even by the most distant past, and the past cannot be erased.” (Waxing philosophical, are we, Garrett?)
Essentially, he boils down the human-planet system to physics. Carbon dioxide, the output of energy consumption, exits civilization at a constant rate, but accumulates over time. This tradeoff is represented as the variable η, which is the “rate of return” of energy to a system. It essentially represents a feedback loop in which the greater the energy consumption and production, the greater is the potential for more consumption and production. (Remember that carbon emissions are tied to this production and energy consumption.) What is most important to note is that if η > 0, the system is growing, meaning energy consumption is increasing; when η < 0, it is shrinking; and when η=0, system growth is at a standstill.
It seems obvious that energy consumption would be tied to production in general. But if energy consumption also is linked to economic growth, then we would have another way to think about how humans, energy and the environment interact. Garrett based his calculations on the assumption that there is some constant value, λ, which links energy consumption to economic value through the equation:
a = λC, where a = global primary energy consumption and C = global economic value
If his argument is true, λ has to be constant with time. To show the existence of this constant, Garrett turned to data for world energy production (and thus consumption) a from the Annual Energy Review (2006) and global economic production P from United Nations data and looked at the whole 36-year timeframe for which he had data. As you can see from his figure (below), the ratio of a/C for λ stayed constant at around 0.306 exajoules per trillion for the entire period. (Also note the dramatic increase in η since the industrial revolution.) (FYI: P is production rate in 1990 dollars/year.)
Garrett does admit that, since he has such a short length of data to work with, this constant could only apply to this 36-year period and not more. However, I find his evidence sufficient to consider the model further. As we accumulate more data on energy and economic production, it will be interesting to see if this constant λ is, in fact, constant.
This is a simple concept: that economic growth and increased energy use are linked. However, what it implies about how to reduce energy use is harder to grasp. This paper suggests that to bring η below zero and thus lose our forward acceleration of energy use, we have to actually shrink our economy. For some reason, saying “shrink our consumption” seems doable, but when it is tied to the economic success of countries, developing or stable, it seems like far more of an impossible task. In this way, Garrett’s paper points out a flaw in current discussions about climate change: we want to reduce emissions, but at the same time keep living our lives the way we do, keeping production high and the economy growing.
The next question is: what if we change our energy to non-carbon-based sources, such as wind or solar power? For η to equal or be less than zero, we would need to make a switch to non-carbon sources at the same rate as η itself, the rate of return. The 2005 value for η is 2.1% growth per year (see figure above; According to Garrett, “2.1% of current annual energy production corresponds to an annual addition of approximately 300 GW of new non-carbon emitting power capacity — approximately one new nuclear power plant per day.”
Garrett’s paper seems to present us with an impossible task: up the building of non-carbon-based energy sources while simultaneously downgrading our economy. (This gets even more complicated with the news that we may be heading towards a uranium shortage, so nuclear power may not be realistic.) It’s a hefty charge, and one that makes the future seem quite bleak. However, this work should be taken with a grain of salt. As much as simplification can be helpful in understanding a system, we cannot just give up. Other factors can help mitigate our carbon emissions — if we don’t believe this then we’re wasting our time — and work should still continue to figure out those methods.
More than anything, I think that this is a really interesting way to think about humans and civilization on this planet. When we’re talking and thinking about climate change, it’s easy to play the blame game and assign roles to different parts of the world or society, whether they be developing nations, industrialization and globalization, or rich people in mansions with enormous carbon footprints. This paper makes the reader step back and realize that it’s not one thing — it’s the entire planet. One country’s change isn’t going to do it. While biking to work makes me feel good, we need everyone to bike to work in order to reduce our consumption dramatically enough to reduce emissions, and hopefully get our η in the negative.
Garrett, T. (2009). Are there basic physical constraints on future anthropogenic emissions of carbon dioxide? Climatic Change DOI: 10.1007/s10584-009-9717-9