Observational requirements for climate monitoring

After any decision to deploy, a monitoring system will need to be in place to evaluate changes to important climate variables. The most obvious of these are surface temperature and precipitation, but stratospheric ozone, stratospheric water vapor, stratospheric temperature, surface air quality, sea-ice extent, diffuse radiation and surface UV, and ocean circulation are also critical monitoring targets. Gaining a deep understanding of SAI's impact on these climate responses is important for validating models and improving their predictions of future climate evolution.

Most of these variables are well measured at present, so this activity determines how long of a record is necessary to see SAI-related changes, and identifies potential risks to the observing systems that would provide those records. The necessary record length will be determined through analysis of "time of emergence" (ToE) — how long it takes for a change caused by SAI to become visible against natural background variability. ToE incorporates information about seasonal and interannual variability as well as the characteristics of the observing systems. The modeling will determine the ToE for different quantities and different observing systems, in order to recommend the minimum set of observations for tracking these impacts.

The first step is defining ToE for the SAI context. Multiple definitions and methods for ToE exist, predominantly concerned with detecting a forced signal against background variability. A few papers have examined ToE for mitigation strategies, but ToE for SAI has not yet been a focus. The first deliverable is therefore a paper comparing these methodologies and their implications for an SAI ToE, concluding with a recommendation for a preferred method.

Once the method is established, subsequent papers will address individual variables, covering both the ToE and the observational requirements for learning and detection. The exact division into individual studies required is not yet clear, but it is likely that at least each of the following would need to be considered individually, in each case assessing when signals would emerge across SAI scenarios, and when we would expect to be able validate model predictions and reduce uncertainty using the observed response: (a) stratospheric water vapor and stratospheric temperature, (b) stratospheric ozone and surface radiation, (c) ocean circulation, (d) Arctic temperature and sea-ice extent, (e) regional extremes in surface temperature and surface air quality (f) global and regional precipitation pattern changes. These correspond to the climate response uncertainties.