Land-sea shifts in future tropical rainfall
October 2017 – Will future tropical rainfall occur over land or ocean? A recent study in Journal of Climate explores the link between future changes in surface climate and where future rain will fall.
Changes in future rainfall in tropical regions simulated by climate models are very uncertain (Collins et al., 2013). A key unanswered question is whether rain that currently falls over land will shift over to the ocean. This is particularly important for tropical South America and Amazonia, which depend on heavy rainfall to maintain their rainforest and which are predicted to experience large drying in future by some climate models including the Met Office Hadley Centre model, but not by others.
A new study led by Dr Hugo Lambert at the University of Exeter in collaboration with Dr Rob Chadwick at the Met Office Hadley Centre shows how this difference in behaviour between climate models is linked to changes in future surface humidity and temperature, giving us the opportunity to understand model performance and compare it to observations (Lambert et al., 2017).
The tropical atmosphere
Figure 1: Heavy rain falls over areas of high surface relative humidity and temperature. It doesn’t tend to fall over areas of low humidity and temperature.
Tropical meteorology is quite different from that seen over the UK and Europe. Temperature and humidity vary strongly at the surface between rainforest, oceans and desert. Aloft, the tropical atmosphere is cooler, but quite uniform, with a temperature that reflects average conditions everywhere. Rainfall occurs when hot, humid air at the surface mixes upwards into the cooler atmosphere aloft, similar to convection currents seen in a pan of soup on the hob. Convection and associated rainfall is therefore most intense where the surface is warmest and most humid relative to the tropical mean, as seen in Figure 1.
Figure 2: Distribution of land and ocean in the present day, as a function of temperature and humidity.
In the present day, what is the temperature and relative humidity of areas of land and ocean? Or, to put it another way, which regions of Figure 1 would be land and which ocean? Desert regions may be hot or cooler, but always have low surface humidity so little rainfall is expected, as illustrated on the left-hand side of Figure 2. Tropical rainforests, however, are both hot and humid, and therefore associated with heavy rainfall; they therefore should be found in the upper right corner of Figure 2, as shown. These expectations match up with what is found in climate models.
In the future, surface temperature and humidity will change. Land is expected to warm more than the ocean, as it has during the historical record. However, relative humidity is expected to decrease over land but remain relatively constant over ocean. It turns out that the decrease in land humidity is important and would be expected to “move” some areas of wet land such as present-day Amazonia towards the left and away from the top-right of Figure 2, meaning less rainfall is expected there overall.
The new study finds that this is followed in most climate models and in the Hadley Centre model in particular (HadGEM2), but not all (e.g. FGOALS, IPSL). Interestingly, the models that follow expectations predict future decreases in rainfall over Amazonia. Models that do not follow expectations, predict no such shifts.
Future work with observations
Now that we have established a framework that links rainfall shifts to future surface temperature and humidity change in some climate models but not others, the next step is to test whether it works in the real world. Observations of tropical rainfall over both land and ocean are available from satellites as far back as 1979 (Adler et al, 2016 ). This is not a long enough period over which to examine climate change, but we can examine present day rainfall and changes caused by climate variability.
Future work will test whether the distribution of rainfall and land presented in Figures 1 and 2 occurs in the real world in the present day. We will then explore whether rainfall shifts match up with changes in surface conditions under shorter term climate variability caused by El Niño, which causes large variations in tropical climate, strongly affecting both China and Brazil. If we can understand how rainfall shifts work in the tropics in the present, we can begin to understand how it may change in the future, helping us to reduce uncertainty in climate model projections and provide more accurate information to decision-makers.
Future changes in tropical rainfall are expected to be dominated by shifts in rainfall location. Our work shows how these shifts are linked to changes in surface temperature and relative humidity, allowing us to develop an understanding of rainfall change. This is key for Brazil, which is predicted to experience substantial reductions in precipitation by most climate models, but also important for southern China and tropical Asia whose rainfall is strongly influenced by tropical processes. Our future work will test our ideas on present day mean climate and variability, with a particular focus on El Niño, which dominates year-to-year changes across the region. Our eventual aim is to reduce uncertainty in future projections of precipitation change in tropical regions.
This research was supported by the Climate Science for Service Partnership Brazil (CSSP Brazil) and CSSP China , which is supported by the Newton Fund, managed by the Department for Business, Energy and Industrial Strategy (BEIS).
Adler et al., 2016. An Update (Version 2.3) of the GPCP Monthly Analysis. Huffman, G.J., R.F. Adler, P. Arkin, A. Chang, R. Ferraro, A. Gruber, J. Janowiak, A. McNab, B. Rudolf, U. Schneider, 1997: The Global Precipitation Climatology Project (GPCP) Combined Precipitation Dataset. Bull. Amer. Meteor. Soc., 78(1), 5-20.
Collins M, Knutti R, Arblaster J, Dufresne J-L, Fichefet T, Friedlingstein P, Gao X, Gutowski WJ, et al. (2013). /Chapter 12 - Long-term climate change: Projections, commitments and irreversibility./ In: Climate Change 2013: The Physical Science Basis. IPCC Working Group I Contribution to AR5. Eds. IPCC, , Cambridge: Cambridge University Press.
Lambert, F. H., A. J. Ferraro and R. Chadwick, Land-ocean shifts in tropical precipitation linked to surface temperature and humidity change, /J. Clim./, Vol. 30, 4527-4545, 2017.