Study gives clearer picture of how land-use changes affect U.S. climate
from a Purdue University press release
Researchers say regional surface temperatures can be affected by land use, suggesting that local and regional strategies, such as creating green spaces and buffer zones in and around urban areas, could be a tool in addressing climate change.
A study by researchers from Purdue University and the universities of Colorado and Maryland concluded that greener land cover contributes to cooler temperatures, and almost any other change leads to warmer temperatures. The study, published on line and set to appear in the Royal Meteorological Society’s International Journal of Climatology later this year, is further evidence that land use should be better incorporated into computer models projecting future climate conditions, said Purdue doctoral student Souleymane Fall, the article’s lead author.
“What we highlight here is that a significant trend, particularly the warming trend in terms of temperatures, can also be partially explained by land-use change,” said Dev Niyogi, a Purdue earth and atmospheric sciences and agronomy professor, and the Indiana state climatologist. He is the study’s corresponding author.
Niyogi and Fall say the idea that land use helps drive climate change has been poorly understood compared to factors such as greenhouse gas emissions. But that is changing.
“People realize that land use cover also is an important force and not only at the local but also at the regional scale,” said Fall, whose doctoral research focuses on the impacts of land surface properties on near-surface temperature trends.
The researchers used higher resolution temperature data than previous studies, meaning the data was more detailed, Niyogi said. They also employed dynamic data on land-use changes from 1992-2001, which was derived from satellite imagery.
Niyogi said having an understanding of land use’s affects on climate change could have climatic and other benefits. For instance, creating green spaces and buffer zones in and around urban areas also could be aesthetically attractive, he said.
Among the study’s findings:
* In general, the greener the land cover, the cooler is surface temperature.
* Conversion to agriculture results in cooling, while conversion from agriculture generally results in warming.
* Deforestation generally results in warming, with the exception of a shift from forest to agriculture. No clear picture emerged from the impact of planting or seeding new forests.
* Urbanization and conversion to bare soils have the largest warming impacts.
In general, land use conversion often results in more warming than cooling.
The study took an approach called “observation minus reanalysis,” or OMR. Through this process, the researchers used temperature data from local ground observations, observation and computer modeling, Geographic Information Systems (GIS) and statistical methods. They were able to separate the effects of land use or cover from greenhouse warming and isolate the impact from each land use or cover type. The more detailed data provided a clearer picture of the effects of land surface properties on near-surface temperature trends.
“We showed this quantitatively for the first time,” said University of Maryland atmospheric and oceanic science Professor Eugenia Kalnay, who developed the OMR method with Florida State University Professor Ming Cai. She also is a co-author of the study.
While the effects of greenhouses gases like carbon dioxide are clear, Kalnay said, the study does suggest land use needs to be considered carefully as well.
“I think that greenhouse warming is incredibly important, but land use should not be neglected,” she said. “It contributes to warming, especially in urban and desertic areas.”
Another study co-author, Roger Pielke Sr., said the results indicate that “unless these landscape effects are properly considered, the role of greenhouse warming in increasing surface temperatures will be significantly overstated.” Pielke is a senior research scientist in atmospheric and oceanic sciences at the Cooperative Institute for Research in Environmental Sciences and the Department of Atmospheric and Oceanic Sciences at the University of Colorado in Boulder.
Purdue’s Gilbert Rochon and Alexander Gluhovsky also participated in the study. Rochon is associate vice president for collaborative research for Information Technology at Purdue (ITaP) and director of ITaP’s Purdue Terrestrial Observatory satellite and remote sensing data program. Gluhovsky is a Purdue professor in earth and atmospheric sciences and statistics.
The work was supported by the U.S. Department of Energy Atmospheric Radiation Measurement program, NASA, the National Science Foundation, and the National Oceanic and Atmospheric Administration.
###
ABSTRACT
Impact of land use cover on temperature trends over the continental United States: assessment using the North American Regional Reanalysis
We investigate the sensitivity of surface temperature trends to land use land cover change (LULC) over the conterminous United States (CONUS) using the observation minus reanalysis (OMR) approach. We estimated the OMR trends for the 1979-2003 period from the U.S. Historical Climate Network (USHCN), and the NCEP-NCAR North American Regional Reanalysis (NARR). We used a new mean square differences (MSDs)-based assessment for the comparisons between temperature anomalies from observations and interpolated reanalysis data. Trends of monthly mean temperature anomalies show a strong agreement, especially between adjusted USHCN and NARR (r = 0.9 on average) and demonstrate that NARR captures the climate variability at different time scales. OMR trend results suggest that, unlike findings from studies based on the global reanalysis (NCEP/NCAR reanalysis), NARR often has a larger warming trend than adjusted observations (on average, 0.28 and 0.27 °C/decade respectively).
OMR trends were found to be sensitive to land cover types. We analyzed decadal OMR trends as a function of land types using the Advanced Very High Resolution Radiometer (AVHRR) and new National Land Cover Database (NLCD) 1992-2001 Retrofit Land Cover Change. The magnitude of OMR trends obtained from the NLDC is larger than the one derived from the static AVHRR. Moreover, land use conversion often results in more warming than cooling.
Overall, our results confirm the robustness of the OMR method for detecting non-climatic changes at the station level, evaluating the impacts of adjustments performed on raw observations, and most importantly, providing a quantitative estimate of additional warming trends associated with LULC changes at local and regional scales. As most of the warming trends that we identify can be explained on the basis of LULC changes, we suggest that in addition to considering the greenhouse gases-driven radiative forcings, multi-decadal and longer climate models simulations must further include LULC changes.
The peer reviewed paper which this press release discusses is
Fall, S., D. Niyogi, A. Gluhovsky, R. A. Pielke Sr., E. Kalnay, and G. Rochon, 2009: Impacts of land use land cover on temperature trends over the continental United States: Assessment using the North American Regional Reanalysis. Int. J. Climatol., DOI: 10.1002/joc.1996.
To Willis Eschenbach
Hi Willis
I agree with you, all the more as all the ‘inconsistencies’ you mention are listed in our conclusion. Excerpt:
“In addition, our analysis shows that there is not always a straightforward relationship between the different types of conversions: for example, (1) both conversion of urban to barren and the opposite resulted in slightly negative OMRs; (2) there was a weak warming of areas that shifted from bare soils to grassland/shrubland and for the opposite as well and (3) both conversion from forest to grassland/shrubland and the opposite were associated with a weak warming. In a number of cases, our estimates were hampered by the lack of significance due to a small number of samples. All these considerations lead us to conclude that the effects of LULC changes on temperatures trends are significant but more localized studies need to be conducted using high-resolution datasets”.
These results are what we obtained from the analysis that examined the trends in areas that changed from one type to another using the NLCD 1992–2001 Retrofit Land Cover Change. Of course, the OMR values used in this specific analysis span the same period (as explained in section 2 of the paper).
While there is still much work to do, our results demonstrate a clearly discernible effect due to landscape change.
Some contributions in this blog pointed out the shortness of the study period to assess changes due to land use. Agree, but for now, despite the shortness of the period of acquisition, the NLCD dataset (30 meters grid spacing) is the only one that breaks down the data into non-changed vs. changed areas (a total of 87 classes). A closer look at this dataset shows that during this short period, some conversion have been important: e.g. (i) urbanization, especially from agriculture and forest to urban; and (ii) conversion to croplands, especially from barren and grass/shrub.
As mentioned in some contributions, there are some odd conversions, such as urban to barren or urban to grass/shrub, but our study did not invent these conversions. That’s the NLCD datasets that identifies such land use/cover conversions.
That said, it is important to have in mind that OMR values reflect not only land use/cover changes, but also a number of other factors such as various climatic and non-climatic biases that affect near-surface temperature trends, including the quality of station siting which, so far, has not been included in temperature trend assessments. Moreover, factors such as near-surface moisture and wind speed can influence temperature trends and make them unrepresentative of the regional trends. As a result, OMR should not simply be read as “warming” or “cooling”.
Last but not least, the issue that is presented in the press release focused only on temperature trends with respect to LULC types. The press release only reports on part of our research [in this paper and a number of other peer reviewed papers] on the issue of the uncertainties associated with homogeneity adjustments and multi-decadal surface temperature trends.
Regards
Dr. Souleymane, many thanks for your comments.
I believe that the answer to the inherent contradictions I pointed out lies in the nature of your study.
In your study, you used the difference between the USHCN data, and “reanalysis” data. I have great problems with “reanalysis” data. For example, see my analysis of the NCEP reanalysis data here. It shows that the NCEP reanalysis data does a very poor job at replicating the amplification behaviour of the atmosphere.
If you are going to use reanalysis data to do the type of analysis you have done, it seems to be incumbent on you to first verify the reanalysis data. You do not appear to have done so. In your study you say
However, this is a necessary but not sufficient investigation. The NARR is based on observations, so we would expect it to be close. But that is not enough.
What you need to look at is, when the USHCN disagrees with the NARR, what is the reason for the disagreement? These disagreements could be from a variety of causes, including LU/LC changes, random error, bad computer programming or assumptions, or other environmental factors.
You have assumed that the difference is due to changes in LU/LC. However, this is something that needs to be proven rather than asserted.
And in fact, your study establishes that there is a very good chance that the differences are merely random. I say this because of the contradictions in the results that I highlighted above. Since they go both ways, it certainly leaves random fluctuations high on the list of possible reasons.
There are other possibilities as well. For example, in the map shown at the top of this thread (Figure 5 in your study), there are huge areas where the NARR is significantly different from the USHCN data. One of them is roughly contiguous with the Rocky Mountains … so the first thing I would check is to see whether there is a correlation of the OMR (observation minus reanalysis) data with elevation data. You say there is a “qualitative correlation” that explains this (shown in Fig. 6 of your study), but absent a quantitative correlation, I am not convinced. I get nervous when someone reports a qualitative correlation and does not proceed to do and report a quantitative correlation.
Because frankly, I have a very hard time believing that there have been large LU/LC changes over the entirety of the Rocky Mountains as shown in the map at the top of this thread. For another example, look at the area shown in blue that stretches from Southern California to Nevada … I’ve driven that stretch many times, and there is no common LU/LC change that covers that area. Over much of it, there is little development of any kind. So clearly, while LU/LC may be a factor, there are also some other thing at play that affect large areas.
In short, while an OMR analysis may be able to be related to LU/LC at some future date, there are obviously huge confounding factors which have not been considered, much less allowed for in your analysis. You cannot simply say that OMR and LU/LC are correlated without removing those factors.
And this is clearly demonstrated by your results. If the OMR method were valid and the confounding factors had been removed, we would not see the curiosities I note above, viz:
• when forest is converted to barren, the OMR rises … and when barren is converted to forest, the OMR rises.
• when grass/shrub is converted to barren, the OMR rises … and when barren is converted to grass/shrub, the OMR rises.
• when urban is converted to barren, the OMR drops … and when barren is converted to urban, the OMR drops.
These contradictions demonstrate that using the OMR method to analyse LU/LC is not ready for prime time. It may be in the future, but at present, it gives contradictory results.
Again, Dr. Fall, many thanks for your comments. I am impressed that you are willing to publicly defend your results, as this is extremely uncommon with climate scientists.
Well, I was just speculating about elevation being a factor in my post just above, but today I chanced onto this …
SOURCE