Global Dimming, Stilling, Evaporative Demand and Transpiration.

It is commonly thought that with increasing global temperature, evaporative demand will increase, drying out the soil and leading to drought and soil water deficit. Of course this is a superficial view, taking no note of changes in precipitation, for example. Surprisingly, however, observations over the last 40 years of evaporative demand, as assessed by measurement of pan evaporation rate, for example, show that evaporative demand over land has been decreasing in many parts of the world (Peterson et al., Nature 1995).

Roderick and Farquhar (Science 2002) showed that decreases in net radiation (“dimming”) are an important feature in the phenomenon.  Measurements are ideally needed of all the parameters that determine evaporative demand: net radiation, windspeed, humidity deficit, and (weakly, only, when humidity is expressed as humidity deficit) temperature. The use of temperature alone to predict changes in evaporative demand, as is common in the United States in the context of assessing drought intensity, has been demonstrated several times (e.g. Sheffield et al., Nature 2012) to cause errors of considerable magnitude and even of sign when applied to cases where global change is occurring. The diminution of pan evaporation rate in Australia was shown (Roderick et al., GRL 2007) to be caused by decreases in net radiation (“dimming”) and in windspeed (“stilling”). Subsequently those findings have been repeated in several areas around the world.

There has been speculation on the reasons for dimming and stilling, yet their origin remains unclear. The impacts of increasing aerosol loadings (e.g. in China, India) on solar radiation are now widely understood but dimming has been reported in many apparently pristine regions (e.g. NW Australia) implying a role for clouds. The underlying measurements are challenging, yet need to be intensified and made easily available more widely. Given the importance of evaporative demand to transpiration in agriculture and other human endeavours, it would be helpful if such measurements could be accessed in much the same way that carbon dioxide (and some other greenhouse gases) concentrations are coordinated and made readily available on the web. We discuss some of the issues involved.

Pan evaporation records from well-managed sites are valuable because of their widespread and extensive temporal nature. Any moves to discontinue such measurements because of the human labour involved should be examined in detail. We need to ensure, for example, that there is sufficient overlap with expensive, sophisticated measurements of net radiation, etc., that a synthetic pan evaporation rate can be calculated for all sites involved. We discuss such requirements. Given that measurements of net radiation are not widespread, it would seem that pan evaporation records should be with us for some time.