Climate change in Central America and Mexico: regional climate model validation and climate change projections

TLDR: In this article, a regional climate model for Central America and Mexico has been proposed to simulate spatial and temporal variability of temperature and precipitation and also to capture regional climate features such as the bimodal annual cycle of precipitation and the Caribbean low-level jet.

Abstract: Central America has high biodiversity, it harbors high-value ecosystems and it’s important to provide regional climate change information to assist in adaptation and mitigation work in the region. Here we study climate change projections for Central America and Mexico using a regional climate model. The model evaluation shows its success in simulating spatial and temporal variability of temperature and precipitation and also in capturing regional climate features such as the bimodal annual cycle of precipitation and the Caribbean low-level jet. A variety of climate regimes within the model domain are also better identified in the regional model simulation due to improved resolution of topographic features. Although, the model suffers from large precipitation biases, it shows improvements over the coarse-resolution driving model in simulating precipitation amounts. The model shows a dry bias in the wet season and a wet bias in the dry season suggesting that it’s unable to capture the full range of precipitation variability. Projected warming under the A2 scenario is higher in the wet season than that in the dry season with the Yucatan Peninsula experiencing highest warming. A large reduction in precipitation in the wet season is projected for the region, whereas parts of Central America that receive a considerable amount of moisture in the form of orographic precipitation show significant decreases in precipitation in the dry season. Projected climatic changes can have detrimental impacts on biodiversity as they are spatially similar, but far greater in magnitude, than those observed during the El Nino events in recent decades that adversely affected species in the region.

Figures

Fig. 2 Annual mean surface air temperature comparison between the a RCM baseline (RCM.BL) and b CRU data for the period 1961–1990 and between the c RCM.BL and d NARR data for the period 1980–1990. The RCM SAT bias is shown relative to e CRU and f NARR data

Table 1 Lapse rates for the RCM baseline simulation (RCM.BL) and the comparison data sets

Fig. 3 The RCM SAT bias (RCM.BL-CRU and RCM.BL-NARR) in C plotted as a function of elevation. The horizontal bars indicate one standard deviation around the mean SAT bias at every elevation bin

Table 3 Model bias in the mean (l) and variability (r; standard deviation) of seasonal PRCP relative to the GPCC data for six regions defined in Fig. 7

Fig. 16 Projected change in PRCP (colors, % change with respect to RCM.BL), MSLP (contours in mb) and 10 m winds (m/s). Percentage change within ±20% is not significant at the 95% significance level

Fig. 4 Spatially and temporally averaged annual cycle of precipitation for the Central American landmass as described by GPCC, CMAP, RCM.BL, and GCM.BL data. CMAP data is averaged over the period 1979–1990 whereas all other data sets use a 30 year (1961–1990) average

Frequently asked questions

Q1. What contributions have the authors mentioned in the paper "Climate change in central america and mexico: regional climate model validation and climate change projections" ?

Here the authors study climate change projections for Central America and Mexico using a regional climate model. A large reduction in precipitation in the wet season is projected for the region, whereas parts of Central America that receive a considerable amount of moisture in the form of orographic precipitation show significant decreases in precipitation in the dry season. The model shows a dry bias in the wet season and a wet bias in the dry season suggesting that it ’ s unable to capture the full range of precipitation variability. 

Q2. What are the future works mentioned in the paper "Climate change in central america and mexico: regional climate model validation and climate change projections" ?

In the highlands of Nicaragua and Costa Rica, the future SAT distributions lie completely outside the present-day distributions, which shows that warmest temperatures in the baseline run are lower than the coldest temperatures in the scenario run. An increase in the atmospheric water vapor in the future and the associated increase in the latent heat release results in a decreased moist-adiabatic lapse rate and a higher increase in temperatures in the free troposphere ( Santer et al. 2005, 2008 ). Besides, the drying pattern in most of Central America under the A2 scenario is very similar to the dry anomalies observed during the El Niño events ( Neelin et al. 2003 ), which leads us to speculate that present-day El Niño-like conditions might be the norm in the future. In future studies, daily data should be used in order to fully understand the utility of a regional model over coarse resolution GCMs and to study changes in climate extremes under the future scenario. 

Q3. What is the main forcing mechanism of climate variability in Central America?

El Niño Southern Oscillation (ENSO), which is the dominant mode of SST and atmospheric variability in the Pacific, is the main forcing mechanism of climate variability in Central America (Giannini et al. 

Q4. What is the pressing issue in conservation biology?

One of the most pressing issues in conservation biology is the observed global decline in amphibian populations (Stuart et al. 2004), which are particularly striking in Central and South America. 

Q5. What is the RCM's prediction of an increase in the variability of land SATs?

In addition to warming, the RCM also predicts an increase in the variability of land SATs in the region, which could be due to increased SST variability in the equatorial Pacific. 

Q6. What regions are projected to experience a decrease in precipitation in the future scenario?

The IAS region, central American landmass, and the eastern Pacific warm pool region are projected to experience a reduction in precipitation under the future scenario. 

Q7. What is the effect of observed SST on the land SATs?

The authors note that since the baseline experiment uses observed SST as a surface boundary condition, its effect onthe land SATs is realistically simulated by the model for the present-day simulation. 

Q8. What makes the GPCC data more reliable?

The GPCC database is an integration of several precipitation data sets including the CRU and GHCN data, which makes the GPCC data more reliable. 

Q9. What is the spatial pattern of the regional model SAT bias relative to the CRU data?

The spatial pattern of the regional model SAT bias relative to the CRU data (RCM.BL-CRU) suggests that the bias may depend on grid-point elevation. 

Q10. How many grid-points are considered in the model bias analysis?

only grid points with elevation differences within 100 m (i.e., 772 out of 1,005 grid-points at CRU resolution) are considered in quantifying model biases. 

Q11. What is the significance of the model bias relative to CRU?

Given the sparsity of high elevation observations that are available for use in the CRU data set, the significance of the model bias relative to CRU is difficult to assess. 

Q12. What was used to divide the model domain into climatologically similar regions?

These fifteen variability modes were used as an input to the K-means clustering algorithm (MacQueen 1967) to divide the model domain into climatologically similar regions. 

Q13. What is the projected decrease in precipitation in the future scenario?

The Caribbean Sea and the Central American landmass are projected to receive less precipitation in the future scenario than the modern-day values in the dry season. 

Q14. Why is the GPCC database a better model for wet and dry seasons?

Due to strong seasonality of precipitation in the region, the mean climate setting and the model evaluation are discussed for wet and dry seasons separately. 

Q15. What is the result of the convergence of low-level winds in the eastern Pacific?

Highest values of precipitation (6–10 mm/ day) in the eastern Pacific between 5 and 10 N are a result of the convergence of low-level winds.