Warming is global, but GHG sources and sinks are local (with minor atmospheric-chemical exceptions). Humanity is global, but only local bits (e.g., c40cities.org) have the means and will to act to reduce global GHG concentrations. So how to help the willing to act most effectively, given limited means and great need?

Effective action requires not only will, but knowledge (of current and probable future states), predictive skill (actions X maps to effects Y), and awareness sufficient to link them. I believe we can profitably (in the sense of marginal benefits exceeding both marginal and opportunity costs) promote positive effective GHG action by extending current tools to "close the loop" linking LULC and GHG for urban/regional planners, facilitating better comparison of alternative actions.

As you know, land use (LU) choices (and our future plans) mostly drive anthropogenic GHG production: where we dwell (e.g., dense cities vs sprawling exurbs), how we dwell (e.g., low surface/volume ratios vs high), where and how we work, how and where we get our nutrients (e.g., energy, food, water) and set our wastes. Land cover (LC) affects not only GHG production, but also GHG consumption (sinks), as well as geophysics (e.g., UHIs, albedo) and geochemistry (e.g., runoff composition and dynamics) more broadly. Fortunately, due to AQ regulations and processes developed over the past 20 years, MPOs in the US (e.g., Indianapolis), and similar urban- and regional-scale entities around the (developed) world, are already both monitoring current LULC, and planning for the future. The process, and toolchain, goes like this (esp in places that are not "in attainment"):

1. MPOs plan

   • locations of residences, businesses, and other {transportation endpoints, emission loci}.

   • transportation network connecting the endpoints

   • regional economics, including (e.g.) housing stock and vehicle fleet

   using regional development models (e.g., UrbanSim)

2. Outputs from step 1 are input to a transportation planning model (TPM, or travel-demand model, TDM), which outputs (et al) trip/vehicle dynamics, link saturations, and mobile-source emissions. E.g., how it's done in Phoenix.

3. Mobile-source emissions from step 2 are combined with area- and point-source emissions from step 1, and assimilated, for future periods of interest (in the US, typically up to 20 years out).

4. An air-quality model (AQM, e.g., CAMx, CMAQ) inputs emissions (from step 3) and past meteorology (and sometimes forecast climatology), and outputs (et al) predictions for

   • concentrations of substances of particular legal concern (in the US, CACs, aka CAPs)
   • (often rather baroque) statistics regarding their dynamics

for various future periods and "modeled episodes" of interest.

6. If the modeled concentrations are within regulatory bounds (in the US, NAAQS), the plans may be approved and (more importantly) funded (in the US, by federal legislation such as ISTEA, TEA-21, SAFETEA-LU, and now MAP-21).

One must note, the above activities and toolchain are not neither spatially or temporally marginal: they

• are mandated by law, and backed by "real money," in many large polities

• have now been in operation for decades

• support a significant regulatory ecosystem, including planning/modeling software and consultants

So what's wrong with this picture? Two main things:

1. It doesn't do GHGs. Fortunately, this seems "in process":

   • Kevin Gurney's group (et al) have done regional-scale CO2 budgets, and are drilling down to finer scales

   • AQMs have long "done CH4" (atmospheric-chemistry-wise) and folks in the CMAQ community are incorporating CO2 and N2O.

So GHGs should be relatively easily integrated into both the emission inventories and model mechanisms.

2. It doesn't facilitate smaller-scale scenario planning. As Gurney indicated in T 4 Dec talk @ AGU 2012 ("Quantification of fossil fuel CO2 at the building/street level for large US cities"), planners want "mitigation guidance tools" with "direct policy interface and Q&A capability." They want to be able to answer questions of the form

   • "We have N $ to spend. How much CO2e can we reduce with that?"

   • "We're considering two main development alternatives. Which produces better GHG outcomes?"

So how to fix these problems? Two main ways:

1. Continue the development of regional AQMs into "Earth region models" (ERMs) by (notably)

   1. advancing the regional-scale modeling of GHG sources. Work on CO2 needs deepened (i.e., Hestia) and broadened (to CH4 and N2O, the work on which is IMHO embryonic compared to CO2).

   2. regional-scale modeling of GHG sinks. Work on this (of which I'm aware) is much more rudimentary (and typically directed to "natural areas"--since apparently humans are not part of nature), but is beginning: e.g.,

      • USGS' LandCarbon project: see Sohl et al 2012 in Applied Geography

      • CMAQ recently incorporated NH3 bidirectionality

2. Closing the plan-outcome visualization loop. This is much more just straight-up software engineering (which is my background--work experience, bachelors degree):

   • current models are quite hard to run individually

   • integrating the individual models into a single data pipeline is even more difficult

   • current model products are relatively unconsumable: considerable "massage" is required to answer the questions that scenario planning asks.

   The difficulty of running the individual models, integrating the toolchain, and visualizing the forecasts, is what feeds the consultants. Conversely, it's the barrier to entry to smaller-scale and less-developed polities, and to all but the most wealthy non-state actors. Minimally, we should be able to find ways to

   • {better, more transparently} couple existing tools
   • better communicate results from current pipelines

So my proposal seems pretty simple--tool the LULC/GHG connection. Transform current, useful, mandated tools into Earth region models that not only answer scientific questions about GHGs, but facilitate planning about GHGs. Don't reinvent wheels, and avoid requiring new policies/funding: instead, exploit what folks are already using.

Because this is not a temporary or local need: as long as we're gonna occupy this planet (or others) with technology of similar power and extent, we will need to better anticipate the consequences of our actions. Big feet gotta step careful.