Vol.:(0123456789) 1 3 Climate Dynamics https://doi.org/10.1007/s00382-019-04668-6 Temperature variability and soil–atmosphere interaction in South America simulated by two regional climate models Claudio G. Menéndez 1,2,3  · Julián Giles 1,2,3  · Romina Ruscica 1,2,3  · Pablo Zaninelli 1,2,3,4  · Tanea Coronato 1,2,3  · Magdalena Falco 1,2,3,5  · Anna Sörensson 1,2,3  · Lluís Fita 1,2,3  · Andrea Carril 1,2,3  · Laurent Li 5 Received: 18 September 2018 / Accepted: 4 February 2019 © Springer-Verlag GmbH Germany, part of Springer Nature 2019 Abstract Interannual variability of surface air temperature over South America is investigated and, based on previous studies, thought to be partly the consequence of soil–atmosphere interaction. Annual and monthly averages of surface air temperature, evapo- transpiration, heat fuxes, surface radiation and cloud cover, simulated by two regional climate models, RCA4 and LMDZ, were analyzed. To fully reveal the role of soil as a driver of temperature variability, simulations were performed with and without soil moisture-atmosphere coupling (Control and Uncoupled). Zones of large variance in air temperature and strong soil moisture-atmosphere coupling are found in parts of La Plata Basin and in eastern Brazil. The two models show diferent behaviors in terms of coupling magnitude and its geographical distribution, being the coupling strength higher in RCA4 and weaker in LMDZ. RCA4 also shows greater amplitude of the annual cycle of the monthly surface air temperature compared to LMDZ. In both regions and for both models, the Uncoupled experiment tends to be colder and exhibits smaller amplitude of the interannual variability and larger evaporative fraction than the Control does. It is evidenced that variability of the land surface afects, and is afected by, variability of the surface energy balance and that interannual temperature variability is partly driven by land–atmosphere interaction. Keywords Interannual climate variability · Surface air temperature · Regional climate modeling · South America · Land– atmosphere interaction 1 Introduction South America is characterized by a large meridional exten- sion, broad range of tropical to extratropical climatic con- ditions and prominent orography. Sea surface temperature anomalies infuence its atmospheric circulation, rainfall and temperature (Nobre and Shukla 1996), with the El Niño Southern Oscillation playing an outstanding role (Cazes- Boezio et al. 2003; Andreoli and Kayano 2005; Poveda et al. 2006). The consequent year-to-year variations of the climate of South America have severe economic efects since a large part of its resources are based on agriculture (Field et al. 2014; Thomasz et al. 2019). Overall, the interannual vari- ability results from the superposition of diferent modes of large scale variability of the climate system, but the land surface state plays a role modulating surface air temperature and precipitation variability (Guillevic et al. 2002; Senevi- ratne et al. 2006; Guo and Dirmeyer 2013; Berg et al. 2015). While the interannual variations of the South American climate related to leading modes of climate variability has been widely analyzed (Garreaud et al. 2008), the role of land–atmosphere interaction has not yet been sufciently explored (Bedoya-Soto et al. 2018; Zaninelli et al. 2018). An appropriate representation of the land surface in climate models results in a better simulation of the South American climate. Barreiro and Díaz (2011) found that * Claudio G. Menéndez menendez@cima.fcen.uba.ar 1 Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Buenos Aires, Argentina 2 Centro de Investigaciones del Mar y la Atmósfera (CIMA/CONICET-UBA), Buenos Aires, Argentina 3 Instituto Franco-Argentino sobre Estudios de Clima y sus Impactos (UMI3351-IFAECI/CNRS-CONICET-UBA), Buenos Aires, Argentina 4 Facultad de Ciencias Astronómicas y Geofísicas, Universidad Nacional de La Plata, La Plata, Buenos Aires, Argentina 5 Laboratoire de Météorologie Dynamique, CNRS, Paris, France