Wednesday, January 6, 2016

2016 paper #3 Climate change and land use

Just gave a very quick read of "Ordonez, A., S. Martinuzzi, V. C. Radeloff and J. W. Williams, 2014. Combined speeds of climate and land-use change of the conterminous US until 2050. Nature Climate Change 4, 811-816."

Ordonez and colleagues present a short article representing lots of work mapping climate change and land use across the US. They examined where climate change and land use change would affect ecosystem most rapidly as independent and combined effects. 

For climate change, the strongest effects will be seen in the Great Plains where there will be no topographic refugia for a warming climate. There will also be high rates of land use change, much of urbanization. So the Plains are screwed. The West appears to have relatively low rates of change for climate and land use. The ne and se US will be mixed.  




5 comments:

  1. Is there a PDF of this anywhere? I am paywalled.

    ReplyDelete
  2. https://www.researchgate.net/profile/Volker_Radeloff/publication/273739806_Combined_speeds_of_climate_and_land-use_change_of_the_conterminous_US_until_2050/links/550b1b4e0cf265693cef67e5.pdf

    If the link doesn't work, I went to Google Scholar and searched the title.

    ReplyDelete
  3. TNX. Color me unimpressed. I am always suspicious of papers which are essentially "models all the way down". When one of those models is misrepresented in the paper, (ironically, by co-authors!) then perhaps it should have failed peer review.

    My suspicion was raised by Figure 1, which bizarrely appears to show conversion of land in northwest Maine to cropland, pasture, and urban uses. You have to dig down into the methods section of the paper to see what is going on:

    "We focused on five land-cover classes: forests, rangelands, pasture, croplands and urban. Present and future land-use coverage for each of the five land-cover classes were determined based on an econometric model(11,12) that predicts spatially explicit land-use changes across the conterminous US for the period 2001–2051. These projections of future land-use change are driven only by alternative economic incentives, which at the timescale of this study (∼50 yr) are more important drivers of land-use change than climate."

    This papers in question are (11) Radeloff, V. et al. Economic-based projections of future land use in the conterminous United States under alternative policy scenarios Ecol. Appl. 22, 1036–1049 (2012), and a paper which shares many co-authors and builds upon it, (12) Lawler, J. J. et al. Projected land-use change impacts on ecosystem services in
    the United States. Proc. Natl Acad. Sci. USA 111, 7492–7497 (2014). Two of the co-authors of these two papers, Radeloff and Martinuzzi, were also co-authors of the Nature Climate Change paper we are looking at here.

    The critical line in the methods area quoted above is "These projections of future land-use change are driven only by alternative economic incentives." This is true of (11) but false of (12) (RED FLAG! PUBLICATION DENIED PENDING REVISION!) which in part built upon (11) by offering, arguendo, an alternative baseline for land-use change (2007-2012, rather than 1992-1997) to test how that would alter the impact of their policy models. Why these periods?

    "The first baseline scenario (1990s trend) assumes continuation of exogenous factors driving land use during a 5-y period from 1992 to 1997. The second baseline scenario (high crop demand) increases the price of agricultural commodities relative to the 1990s trend with concomitant pressures to expand agricultural lands, which more closely resembles the 5-y period from 2007 to 2012. The two scenarios allow us to gauge the sensitivity of our results to different assumptions about the underlying drivers of land-use change." (12)

    There may be a deeper error here. Figure 1 in (12) shows the percentage change in land use as a function of the existing type of land cover, not total land cover - so the increases in crop use in northern New England under the "High Cropland Demand" scenario show a large percentage increase - but only over the current 0-5% of land which is cropland at the beginning of the model run. Conversion rates in Figure 1 of Nature Climate Change paper suggest that the percentage was misapplied - to the total land area. Given my math-challenged status I could be wrong here 8-P.

    ReplyDelete
  4. All my Google apps reset (and are behaving badly right now)... anyways, I appreciate the more in-depth reading of this. When I see these types of papers I expect more. At least I would expect best case scenarios vs. worst case vs. median response. That is, using the climate models with the most temp/precip change with the most land use change (which seems reasonable is you consider the latter drives the former) and a minimal change scenario and a median change. I would have also liked to have seen simple line graphs of urban, forest, agriculture cover over time (and best if given error).

    ReplyDelete
  5. Well, the article does offer some comparisons between the RCP 2.6, 6.0, and 8.5 models (see, for example, figure 3). But again, its models on models, and it should be noted that those in 11 and 12 were designed to test POLICY impact on land-use changes, and were imperfectly re-purposed here.

    ReplyDelete