Tuesday, December 31, 2013

Massetti et al. - Part 1 of 3: Convergence in the Effect of Warming on US Agriuclture

Emanuele Massetti has posted a new paper (joined with Robert Mendelsohn and Shun Chonabayashi) that takes another look at the best climate predictor of farmland prices in the United States.  He'll present it at the ASSA meetings in Philadelphia - I have seen him present the paper at the 2013 NBER spring EEE meeting and at the 2013 AERE conference, and wanted to provide a few discussion points for people interested in the material.

A short background: several articles of contributors to this blog have found that temperature extremes are crucial at predicting agricultural output. To name a few: Maximilian Auffhammer and coauthors have shown that rice have opposite sensitivities to minimum and maximum temperature, and this relationship can differ over the growing season (paper). David Lobell and coauthors found that there is a highly nonlinear relationship between corn yields and temperature using data from field trials in Africa (paper), which is comparable to what Michael Roberts and I have found in the United States (paper).  The same relationship was observed by Marshal Burke and Kyle Emerick when looking at yield trends and climate trends over the last three decades (paper).

Massetti et al. argue that average temperature are a better predictor of farmland values than nonlinear transformations like degree days.  They exclusively rely on cross-sectional regressions (in contrast to the aforementioned panel regressions), re-examining earlier work Michael Hanemann, Tony Fisher and I have done where we found that degree days are better and more robust predictors of farmland values than average temperature (paper).

Before looking into the differences between the studies, it might be worthwhile to emphasize an important convergence in the sign and magnitude of predicted effect of a rise in temperature on US agriculture.  There has been an active debate whether a warmer climate would be beneficial or detrimental. My coauthors and I have usually been on the more pessimistic side, i.e., arguing that warming would be harmful. For example, a +2C and +4C increase, respectively, predicted a 10.5% and 31.6 percent decrease in farmland values in the cross-section of farmland values (short-term B1 and long-term B2 scenarios in Table 5)  and a 14.9 and 35.3 percent decrease in corn yields in the panel regression (Appendix Table A5).

Robert Mendelsohn and various coauthors have consistently found the opposite, and the effects have gotten progressively more positive over time.  For example, their initial innovative AER paper that pioneered the cross-sectional approach in 1994 argued that "[...] our projections suggest that global warming may be slightly beneficial to American agriculture." Their 1999 book added climate variation as an additional control and argued that "Including climate variation suggests that small amount of warming are beneficial," even in the cropland model.  A follow-up paper in 2003 further controls for irrigation and finds that "The beneficial effect of warmer temperatures increases slightly when water availability is included in the model."

There latest paper finds results that are consistent with our earlier findings, i.e., a +2C warming predicts decreases in farmland values of 20-27 percent (bottom of Table 1), while a +4C warming decreases farmland values by 39-49 percent. These numbers are even more negative than our earlier findings and rather unaffected whether average temperatures or degree days are used in the model.  While the authors go on to argue that average temperatures are better than degree days (more on this in future posts), it does change the predicted negative effect of warming: it is harmful.

Monday, December 23, 2013

The Red Queen strikes again

The weekend before last I attended an interesting CIMMYT meeting on remote sensing in Mexico City. Lots of cool stuff going on in remote sensing for agriculture, including use of drones in breeding programs, and near-term prospects for low-cost or free satellite data with high spatial and temporal resolution. But one of the most interesting parts of the meeting for me was catching up with Dave Hodson, a colleague who used to work in CIMMYT’s GIS group and now works full time on monitoring wheat rusts. He’s part of the Borlaug Global Rust Initiative (BGRI) which was started in the wake of the discovery of the UG99 strain of stem rust in 1999.

A quick review: rusts are a nightmare for wheat growers or breeders. They can decimate a wheat crop and can spread incredibly quickly and far. There are three main types of rust: stem rust, yellow (or stripe) rust, and leaf rust. One of the main precursors of the Green Revolution was improving rust resistance of wheat varieties, part of Norman Borlaug’s claim to fame. Breeders must continually make sure their varieties are not too susceptible to rust, and since rusts evolve over time it is often a race just to avoid going backward. That’s why it is often called Red Queen breeding, named after the scene in Throughthe Looking Glass where Alice learns she has to run just to stand still.

The same rust resistance genes were successful for a very long time, until the UG99 strain came along and proved to be a major problem for nearly all widely grown varieties. In stepped scores of wheat scientists, who quickly developed new resistant varieties that have since been widely adopted. With the help of Borlaug, and the Gates Foundation, the BGRI was set up to maintain an internationally coordinated system to monitor and respond to any future rust outbreaks.

Ok, now to the interesting part. A few weeks ago, surveyors in Ethiopia uncovered a sizable amount of wheat area (~10,000 ha) that had been wiped out by stem rust. These varieties were resistant to the known UG99 strains, so it seems that a new strain has emerged. It’s too early to know what this will imply, but an update was posted today on their website, including the picture of an affected field below.

As scary as rust is, the news isn’t all bad. The systems put in place by BGRI have already had several successes, though avoiding something bad happening rarely makes the news. For example, a few years ago, in late 2010, there was a big outbreak of yellow rust in Ethiopia. Roughly a third of the entire wheat crop was lost. This year, there were conditions favorable for yellow rust, and heavy incidence was spotted. But it was spotted early, and fungicides were used to contain the outbreak, and impacts were very small. (Why rusts seem to be happening more often is a topic of debate, and some would blame climate change, but that's a topic for another day).

As this new strain of UG99 emerges, you can see the capacity of BGRI and its partners spring to action. Samples of the spores have already been sent to labs around the world to assess what exactly they are dealing with. Fungicides are being targeted to the areas with active outbreaks. Modelers are looking at potential areas where the spores could spread in the near term, as shown in the figure below from their update. (Now is sowing time throughout much of the Middle East and West and South Asia, so spores reaching there could have big impacts). And breeders will likely soon be sending lines to Ethiopia for screening. 

To me this is a reminder of both how many things can go wrong when trying to produce food, but also how so many hard working, smart people help to bring resilience to modern agriculture. The next time you hear someone talking about “resilience” of agriculture as if it were solely the result of what particular mix of crops or soil biota are in a particular field, you should think about people like Dave Hodson and his colleagues. The resilience of modern agriculture, for better or worse, rests on the tireless but rarely celebrated work of people like them.

Thursday, December 19, 2013

The three wise men (of agriculture)

There’s a new book coming out soon that should be of interest to many readers of this blog. It’s written by Tony Fischer, Derek Byerlee, and Greg Edmeades, and called Crop yields and global food security: will yield increases continue to feed the world?” At 550 pages, it’s not a quick read, but I found it incredibly well done and worthwhile. I’m not sure yet when the public release will be, but I’m told it will be a free download in early 2014 at the Australian Centre for International Agricultural Research website.

The book starts by laying out the premise that, in order to achieve improvements in global food security without massive land use change, yields of major crops need to increase about 1.3% of current levels per year for the next 20 years. They explain very clearly how they arrive at this number given trends in demand, with a nice comparison with other estimates. The rest of the book is then roughly in two parts. First is a detailed tour of the worlds cropping system to assess the progress over the last 20 years, and second is a discussion of the prospects for and changes needed to achieve the target yield gains.

For some, the scope of the book may be too narrow, and the authors fully recognize that yield progress is not alone enough to achieve food security. But for me, the depth is a welcome change from a lot of more superficial studies of yield changes around the world. These are three men who understand the different aspects of agriculture better than just about anyone.

The book is not just a review of available information; the first part presents a lot of new analysis as well. Tony Fischer has dug into the available data on farm and experimental plot yields in each region, with his keen eye for what constitutes a credible study or yield potential estimate (think Warren Buffet reading a financial prospectus). This effort results in an estimate of yield potential and yield gap (the difference between potential and farm yields) by mega-environment and their linear rate of change for the past 20 years. The authors then express all trends as a percentage of trend yield in 2010, which makes it much easier to compare estimates from various studies that often report in kg/ha or bushels/acre or some other unit.
There are lots of insights in the book, but here is a sample of three that seemed noteworthy:

  1. Yield potential continues to exhibit significant progress for all major crops in nearly all of their mega-environments. This is counter to many claims of stagnating progress in yield potential.
  2. Yield gaps for all major crops are declining at the global scale, and these trends can account for roughly half of farm yield increases globally since 1990. But there’s a lot of variation. I thought it interesting, for example, that maize gaps are declining much faster in regions that have adopted GM varieties (US, Brazil, Argentina) than regions that haven’t (Europe, China). Of course, this is just a simple correlation, and the authors don’t attempt to explain any differences in yield gap trends.
  3. Yield gaps for soy and wheat are both quite small at the global level. Soy in particular has narrowed yield gaps very quickly, and in all major producers it is now at ~30%, which is the lower limit of what is deemed economically feasible with today’s technology. One implication of this is that yield potential increases in soy are especially important. Another is that yield growth in soy could be set to slow, even as demand continues to rise the most of any major crop, setting up a scenario for even more rapid soy area expansion.

Any of these three points could have made for an important paper on their own, and there are others in the book as well. But to keep this post at least slightly shorter than the actual book, I won’t go on about the details. One more general point, though.  The last few years of high food prices has brought a flurry of interest to the type of material covered in this book. For those of us who think issues of food production are important in the long-term, this is generally a welcome change. But one downside is that the attention attracts all sorts of characters who like to write and say things to get attention, but don’t really know much about agriculture or food security. Sometimes they oversimplify or exaggerate. Sometimes they claim as new something that was known long ago. This book is a good example of the complete opposite of that – three very knowledgeable and insightful people homing in on the critical questions and taking an unbiased look at the evidence.

(The downside is that it is definitely not a light and breezy read. I assigned parts of it to my undergrad class, and they commented on how technical and ”dense” it was. For those looking for a lighter read, I am nearly done with Howard Buffet’s “40 Chances”. I was really impressed with that one as well – lots of interesting anecdotes and lessons from his journeys around the world to understand food security. It’s encouraging that a major philanthropist has such a good grasp of the issues and possible solutions.) 

Tuesday, December 17, 2013

Yet another way of estimating the damaging effects of extreme heat on yields

Following up on Max's post on the damaging effects of extreme heat, here is yet another way of looking at it.  So far, my coauthor Michael Roberts and I have estimated three models that links yields to temperature:

  1. An eighth-order polynomial in temperature
  2. A step function (dummy intervals for temperature ranges)
  3. A piecewise linear function of temperature
Another semi-parametric way to estimate this to derive splines in temperature.  Specifically, I used the daily minimum and maximum temperature data we have on a 2.5x2.5mile grid, fit a sinusoidal curve between the minimum and maximum temperature, and then estimated the temperature at each 0.5hour interval.  The spline is evaluated for each temperature reading and summed over all 0.5hour intervals and days of the growing season (March-August).

So what is it good for? Well, it's smoother than the dummy intervals (which by definition assume constant marginal impact within each interval), yet more flexible than the 8th-order polynomial, and doesn't require different bounds for different crops like the piecewise linear function.

Here's the result for corn (the 8 spline knots are shown as red dashed lines), normalized relative to a temperature of 0 degree Celsius.

The regression have the same specification as our previous paper, i.e., the regress log yields on the flexible temperature measure, a quadratic in season-total precipitation, state-specific quadratic time trends as well as county fixed effects for 1950-2011.  

Here's the graph for soybeans:

A few noteworthy results: The slope of the decline is similar to what we found before:  A linear approximation seems appropriate (restricted cubic splines are forced to be linear above the highest knot, but not below). In principle, yields of any type of crop could be regressed on these splines.

Sunday, December 1, 2013

It's not the model. Really it isn't

There is a most lively discussion as to whether climate change will have significant negative impacts on US agriculture. There are a number of papers by my co-bloggers (I am not worthy!) showing that extreme heat days will have significant negative impacts on yields for all major crops except for rice. I will talk about rice another day. For the main crop growing regions in the US, climate models project a significant increase in these extreme heat days. This will likely, short of miraculous adaptation, lead to significant yield losses. To put it simply, this part of the literature has shown a sensitivity of yields to extreme temperatures and linked it with projected increases in these extreme temperature events

On the other hand, there are a number of papers, which argue that climate change will have no significant impacts on US agriculture. Seo, in a recent issue of Climatic Change, essentially argues that the literature projecting big impacts confuses weather ("panel models") and climate ("cross sectional models") and that using weather instead of climate as a source of identification leads to big impacts. As Wolfram Schlenker and I note in a comment this is simply not true for five reasons:

1) Even the very limited number of papers he cites, which use weather as the source of variation to identify a sensitivity, clearly state what this means when interpreting the resulting coefficients. There is no confusion here.

2) He fails to discuss the fact that the bias from adaptation when using weather as a source of variation could go in either direction.

3) It is simply not true that all panel models find big impacts and all Ricardian cross sectional models find small impacts. There are big and small impacts to be found in both camps.

4) There is recent work by Burke and Emerick, which uses the fixed effects identification strategy with climate on the right hand side! I wish I would have thought of that. They can compare their "long differences" (a.k.a. climate) sensitivity results to more traditional weather sensitivity results and find no significant difference between the two. This will either enrage both camps or make them very happy, since it suggests that the difference between sources of variation (weather versus climate) in this setting is not huge. 

5) The big differences in studies may finally not be due to differences in sensitivities, but differences in the climate model used. Burke et al. point out that uncertainty over future climate is a major driver of variation in impacts. We refer the reader to this excellent study, which discusses a much broader universe of studies and very carefully discusses the sources of uncertainty in impacts estimates.

We are of the humble opinion, that the most carefully done studies using both identification strategies yield similar estimates for the Eastern United States.