According to the most recent report by the Intergovernmental Panel on Climate Change, land use change has a relatively minor impact on the recent rise in global average temperature.

Yet, as stressed in an earlier set of blogs on Iowa Dew Points and an apparent increase in stormy activity regionally, land use seems to be quite important at local and even regional scales.

Why the difference?

According to an article by Raddatz in a recent issue of Agricultural and Forest Meteorology, about 3.6% of Earth’s surface is covered in crops, and about 6.6% is in pasture. Figures 1 and 2 show what these percentages look like. Even if all the temperature trends associated with changes in land cover were all in the same direction, it would require large changes indeed to show up significantly in the average global temperature for a whole year.

Figure 1. Fraction of Earth’s surface covered with crops, rounded up to 4%.

Figure 2. Fraction of Earth’s surface covered with pasture, rounded up to 7%.

Further, crops grow actively only part of the year. So, for example, winter wheat or corn will lead to cooler maximum temperatures during their growing season. (Recall this is because more of the sun’s energy hitting the surface is going into evapotranspiration and less into heating). During the rest of the year, the stubble or plowed ground would have a different effect from surrounding green vegetation. In fact, the dormant fields could be warmer than grasslands if the grasses are green. Thus it is not surprising that converting natural land cover to crops or pasture does not always have the same effect on temperature change. In some areas, there is a cooling effect (e.g., if more moisture is being evaporated or transpired, or if more sunlight is reflected), while in other areas, there is a warming effect (e.g., more sunlight absorbed, less evaporation or transpiration). And finally, as pointed out in the Iowa Dew Points blogs, regional changes in land cover could have an indirect effect, like shifting the wind patterns. This can in some cases decrease the local influence on temperature. Figure 3 illustrates the “mixed” effect of crops.

Figure 3. Possible scenario showing effects of crops on the global average temperature, with the “red” representing a net warming effect over the whole year, and the “blue” representing a net cooling effect. This is only to illustrate that the net effect of crops will be mixed, rather than saying what the effect will be.

As noted in earlier blogs, cities also affect the temperatures, but they occupy a tiny fraction of Earth’s surface.

This does not mean that land use isn’t important. We live on land, which occupies only 30% of the globe. If we change our percentages of Earth’s surface to percentages of Earth’s land surface, they become bigger – 3.6% of Earth’s surface becomes 12% of Earth’s land surface, and 6.6% of Earth’s surface becomes 22% of Earth’s land surface! Furthermore, we humans aren’t evenly distributed: there are vast parts of Earth that are uninhabited. Human influence on land cover is where humans are. And, of course, those seasonal effects on temperature are important to us if they are happening where we live.

Figure 4. Fraction (12/100) of land surface on Earth covered with crops.

Figure 5. Fraction (22/100) of land surface on Earth covered with pasture.

Climate researchers know this. You have mainly heard about the predicted average annual temperature because it is the easiest “measure” that sums up all the details in one convenient number. This is particularly helpful in sending the message to governments, businesses, and individuals that Earth is getting warmer. Once you think about how this will affect how you live, one temperature is clearly not enough. You want to know how the temperature varies seasonally and where you live.

Also, climate models cover both a large area (the whole Earth) and a long time (decades), so they are expensive to run and require the biggest computers. And, they have to account for the various things that affect climate from the outside – the variability in the sun, the variation in greenhouse gases and particles, etc. By comparison, weather models are run for at most around 10 days or so. To make the runs doable, climate models work on points too far apart to really represent smaller-scale atmospheric motions (”weather”), smaller-scale regional effects (such as vegetation changes), and even terrain.

Some climate scientists have looked at regional climate changes by running weather-type (regional) models to describe what’s happening at the boundaries of a smaller area – such as the United States. (If the model only covers the United States, it still has to account for what the wind brings from the outside!) These models have taught us something. However, there is a worry that the climate models supplying the boundary information are themselves faulty because the effects of “weather” could affect the details of the local wind, temperature, etc.

So, in spite of the enormous costs, climate scientists are just now starting to run climate models on computers or clusters of computers working together that are powerful enough that global climate models can represent these regional changes better. The first such computer system was the Japanese Earth Simulator. It enabled model runs with grid points spaced at one-tenth the distance of most climate models. As more and more people start focusing on regional climate, and more computers with the capability of the Earth Simulator become available, the issues of our effect on Earth’s surface will be studied more intensely.

And our discussion didn’t even consider more long-term effects, such as the effect of land use on changes in greenhouse gas concentrations! Nor have we considered the effects of forests.

Reference: Raddatz, R.L., 2006. Evidence for the influence of agriculture on weather and climate through the transformation and management of vegetation: Illustrated by examples from the Canadian Prairies. Agricultural and Forest Meteorology, 142, 186-202.

At a recent meeting, someone commented to me that the “global-warming folks” must be wrong, since we haven’t had a strong hurricane season since 2005, and weren’t they saying that a warmer climate means more hurricanes?

Since we had work to do, I let the comment go, but decided later it would be a good subject for a blog. Particularly since the “official” hurricane season starts on 1 June in the United States.

In 2005, a couple of papers (see references with asterisks, below) came out that implied that there could be more strong tropical cyclones in a warmer climate. (”Tropical cyclone” is the more general term for such storms; “hurricanes” are tropical cyclones that affect North and Central America and the Eastern Pacific north of the Equator.) These papers were well-timed, because 2005 was a devastating North Atlantic hurricane season, with four – Emily, Katrina, Rita, and Wilma, reaching Category 5 on the Saffir-Simpson scale (sustained winds of 155 miles per hour (135 knots or 249 kilometers per hour – henceforth km/hr). Katrina was the most devastating hurricane in memory, with a death toll (well over 1000) exceeded only by the “1900 storm” that destroyed Galveston, Texas and killed between 6000 and 12,000 people. Hurricane Wilma had the lowest central pressure (882 millibars) of any recorded Atlantic hurricane, with sustained winds of 175 miles per hour or 292 km/hr. (The strongest tropical cyclone was Typhoon Tip, whose central pressure dipped to 870 millibars with sustained winds of 190 mph (305 km/hour) on 12 October 1979).

Figure 1. Number of named tropical storms (blue) and named hurricanes (red) by year. From the U.S. National Climatic Data Center.

Finally, 2005 was the year they ran out of names and had to start using Greek letters to name hurricanes, with Zeta, the 26th and last storm, occurring between 30 December 2005 and 6 January 2006. (For the North Atlantic list, names starting with Q, U, X, Y, and Z are left out; the remaining hurricanes were named for the first five letters of the Greek alphabet).

The arguments used for strong hurricanes in a warming climate related to the warming of the sea-surface temperature. Basically, a hurricane is like a heat engine, getting its energy primarily from water vapor evaporating from the warm sea surface, and cooling off at cloud top, around 15 kilometers above the surface.

Figure 2. Globally-averaged sea-surface temperature anomaly (sea surface temperature minus mean for 1961-1990). Data from Climate Research Unit, Hadley Centre, UK. (http://www.cru.uea.ac.uk/cru/data/temperature/)

Although there is variation from region to region, the global average of carefully-compiled sea surface temperatures (Figure 2) does indicate a warming. The warming is due to more greenhouse gases in the atmosphere. These gases trap heat in Earth’s lower atmosphere, land surface, and ocean.

However, there are changes superimposed on this long-term trend. In the North Atlantic, these changes can take several decades. The relatively few strong hurricanes during the 1970s and 1980s follow more strong hurricanes in the 1950s and 1960s so this “natural variability” is important as well. One familiar example of natural variability is El Nino, which spreads warm surface waters eastward across the Equatorial Pacific Ocean and affects wind and weather patterns over much of the earth. As noted in previous blogs, aerosols and solar variability can also affect temperature changes on earth, but the effect of the sun is probably fairly minor over the last several decades.

Other things being equal, warmer sea surface temperatures would mean stronger hurricanes. However, other things are not equal. Certain wind patterns favor hurricane development, while other wind patterns do not. For example:

• Converging winds (more air flowing horizontally into an area than leaving) favor hurricanes. Hurricanes are storms with air near the surface spiraling in to the center, until it reaches the eye wall, where it spirals upward and slightly outward. Such motions are favored in regions where the air is slowly moving upward. This happens where winds converge into an area.

• Little wind change (called wind shear) with height favors hurricanes. If the wind changes enough with height, it can disrupt the air circulation in a developing tropical storm, keeping it from developing into a hurricane.

• Wind patterns are much harder to predict in climate models. For example, researchers have found that fewer hurricanes occur during El Nino years. This is because El Nino warms the eastern equatorial Pacific, and this leads to wind shear over the Atlantic basin. But it is not clear how the warming climate will affect the occurrence of El Ninos. If there are more in the future – this effect could offset that of the generally warming sea surface temperatures. Indeed, a new paper by Knutson and colleagues has just pointed out such a possibility. However, it is interesting to note their caution and list of caveats (mostly that the input to their modeling studies is based on global climate models that are still not adequate at regional scales).

What about 2008? On 22 May, the U.S. Climate Prediction Center issued a “2008 Hurricane Outlook” that called for a “90% probability of a near-normal or above-normal hurricane season” in the United States, with the above-normal season more likely (65% chance). Among the factors considered was La Nina (the “cold” phase of El Nino).

As for the rest of the world, the northern hemisphere has already experienced one of the most deadly tropical cyclones in recent history, Cyclone Nargis, which devastated parts of Myanmar and killed tens of thousands of people.

For the longer-term future, the warmer oceans should lead to stronger tropical cyclones – when the wind conditions favor their formation and growth. The real question is how often the favorable wind conditions will happen.

References

*Emanuel, K. 2005: Increasing destructiveness of tropical cyclones over the last 30 years. Nature, 436, 686-688.

Knutson, T.R., et al., 2008: Simulated reduction in Atlantic hurricane frequency under twenty-first century warming conditions. Nature Geoscience, doi:10.1038/ngeo202.

*Webster, P.J., G.J. Holland, J.A. Curry, and H.-R. Chang, 2005: Changes in Tropical Cyclone Number, Duration, and Intensity in a warming environment. Science, 309, 1844-1846.

Hurricane Statistics from
NCDC: Climate of 2005: Atlantic Hurricane Season Summary.
http://www.ncdc.noaa.gov/oa/climate/research/2005/hurricanes05.htlm

Acknowledgments: I wish to acknowledge Caspar Ammann of NCAR for checking this blog and pointing out the Knutson reference.