Modelling of mean annual air temperature (1979-2010) for the study of mountain permafrost in the SW of San Juan

DARIO TROMBOTTO LIAUDAT; CARLA C. TAPIA BALDIS; FEDERICO OTERO; VILLARROEL CRISTIAN DANIEL

Potsdam

Congreso; XI. International Conference On Permafrost; 2016

It is the aim of the present study to estimate and statistically model the mean annual air temperature (MAAT) for the southwestern part of the Andes in the province of San Juan, Argentina (31° S to 33° S and 71° W to 69° W), an area with hardly any register of instrumental data. At a large scale the MAAT may provide a good approximation of permafrost distribution. The Argentine Central Andes have a mean altitude of 3500 m asl with several peaks of 6000 m asl and higher. The reanalysis data indicate a MAAT of 0.88 °C between 1979 and 2010 at a height of 3685 m asl.The method of Gruber (2012), based on the correlation between MAAT and permafrost, in order to calculate the extension of permafrost (PE) and its derivation for the calculation of a global index of permafrost zones (PZI) was used. The extension of large areas of permafrost is obtained by extrapolating the MAAT isotherms that indicate its limits.The set up MAAT model is based on the values of the data set from the National Center for Environmental Prediction (NCEP) Climate Forecast System Reanalysis (CFSR) (Saha S et al., 2010) available on the website: cs.ncep.noaa.gov/cfsr/. The used period of reference in order to figure the MAAT in the present study comprises 01.01.1979 to 01.01.2011. The chosen NCEP-CFSR grid resolution is 0.5 °x 0.5° and MAAT data were calculated at 3685 m asl. The heights were taken from the digital elevation model of the Argentine Republic (MDE-AR) with a spatial resolution of 45 m (http://www.ign.gob.ar/).The MAAT grid was readjusted to the topographical MDE-AR grid, obtaining a grid layer with the spatial MAAT distribution in function of topography , that is to say of 30 arc seconds of spatial resolution (see figure 1).The lack of meteorological stations in the high mountains of the studied area complicates the validation of the interpolated MAAT values. Only two registers of meteorological stations could be compared to the MAAT model: Portezuelo de la Guardia and Campamento Pachón (see Arenson et al., 2010).Portezuelo de la Guardia in the surroundings of the Bramadero river basin at 31° 53 S and 70° 11 W recorded a MAAT of -2.15 °C between 2009-2014 at a height of 4019 m asl, whereas the value estimated by the model for this height is -0.93 °C. The nivo-meteorological station of the camp of the Pachón project (Glencore-Xstrata) at 31° 34 S and 70° 2 W indicated a MAAT of 1.3 °C in the year 2010 at 3600 m asl. The value estimated by the model for the same site is 2.28 °C.In the study area (see fig. 1) cryogenic belts with possible permafrost between 3700 and 4500 m asl and likely permafrost above 4500 m asl were previously established. The average height of the 0 °C isotherm (ca. 3900 m asl) indicates geomorphologically quite precisely the mean lower permafrost limit of sedimentary cryogenic debris slopes, which may contain ice in their matrix, while the limit of cryoforms with more specific creeping (rock glaciers, protalus lobes and protalus ramparts) locally lower the permafrost limit down to 3600-3700 m asl, where it is possible for permafrost to persist even under slightly positive MAAT temperatures (0.5 to 1 °C).Thus considering the 0 °C isotherm as lower permafrost limit in situ, the surface covered by permafrost in the studied area is of 5546 km2 approximately.BibliographyArenson LU, Pastore S, Trombotto D, Bolling S, Quiroz MA, Ochoa X. 2010. Characteristics of two Rock Glaciers in the Dry Argentinean Andes based on Initial Surface Investigations. GEO 2010, Calgary, Alberta.Gruber S. 2005. Mountain permafrost: Transient spatial modelling, model verification and the use of remote sensing. PhD Thesis, Universität Zürich, Zürich. Available from: http://dx.doi.org/10.5167/uzh-18100 (Accessed: 09 April 2009)Saha S, et al. 2010. NCEP Climate Forecast System Reanalysis (CFSR) 6-hourly Products, January 1979 to December 2010. Research Data Archive at the National Center for Atmospheric Research, Computational and Information Systems Laboratory. http://dx.doi.org/10.5065/D69K487J. (Accessed 02 november 2015)