Environ Monit Assess (2019) 191:635 https://doi.org/10.1007/s10661-019-7803-7 Spatial and temporal variation of the vegetation of the semiarid Puna in a pastoral system in the Pozuelos Biosphere Reserve Verónica Rojo & Y. Arzamendia & C. Pérez & J. Baldo & B. L. Vilá Received: 4 February 2019 /Accepted: 1 September 2019 # Springer Nature Switzerland AG 2019 Abstract This study aimed to analyze the spatial and variation of the vegetation cover, biomass, dominant temporal variation of the vegetation in the northern species, and vegetation indices. Ten vegetation units Argentine Puna, utilizing both field sampling and (with differences in composition, cover, and high and remote-sensing tools. The study was performed within low stratum biomass) were identified at our study site. the Pozuelos Biosphere Reserve (Jujuy province, Ar- The diversity of these vegetation units correlated with gentina), which aims to generate socio-economic devel- geomorphology and soil type. In the dry season, the opment compatible with biodiversity conservation. Our vegetation unit with greatest vegetation cover and bio- study was designed to analyze the dynamics of the Puna mass was the Festuca chrysophylla grassland, whereas vegetation at local scale and assess and monitor the in the wet season, the units with greatest cover and seasonal (dry and wet seasons), interannual, and spatial biomass were vegas (peatlands) and short grasslands. The Festuca chrysophylla grasslands and short grass- lands were located in areas with clay soils, except V. R ojo : Y. Arzamendia Instituto de Ecorregiones Andinas (INECOA-CONICET-UNJu), peatlands, associated with valleys and coarse-texture Alberdi 47, 4600 San Salvador de Jujuy, Argentina soils. The vegetation indices used (NDVI, SAVI, and MSAVI2) were able to differentiate functional types of * : : : V. Rojo ( ) Y. Arzamendia J. Baldo B. L. Vilá vegetation and showed a good statistical fit with cover VICAM: Vicuñas, Camélidos y Ambiente, Buenos Aires, Argentina values. Our results suggest that the integrated utilization e-mail: [email protected] of remote-sensing tools and field surveys improves the assessment of the Puna vegetation and would allow a Y. Arzamendia : J. Baldo Facultad de Ciencias Agrarias, Universidad Nacional de Jujuy, periodic monitoring at production unit scale taking into Alberdi 47, 4600 San Salvador de Jujuy, Argentina account its spatial and temporal variation. C. Pérez . Laboratorio de Investigación de Sistemas Ecológicos y Keywords Puna rangelands South American Ambientales, Universidad Nacional de La Plata, 1900 La Plata, camelids . Spatial distribution . Vegetation indices . Argentina Drylands J. Baldo : B. L. Vilá CONICET: Consejo Nacional de Investigaciones Científicas y Técnicas (National Research Council), Buenos Aires, Argentina Introduction B. L. Vilá Departamento de Ciencias Sociales, Universidad Nacional de The Puna or Altiplano is a high mountain plateau locat- Luján, Avenida Constitución y RN 5, 6700 Buenos Aires, ed at 3500 m above sea level in the Central Andes Argentina region, with cold climate, strong solar radiation, low 635 Page 2 of 18 Environ Monit Assess (2019) 191:635 atmospheric humidity, and wide thermal amplitude the former leave the soil unprotected (Cendrero et al. (Aceituno 1996). In Argentina, the Puna extends over 1993). However, little research was done about the the northwestern part of the country, and the main impact of this plan on the natural vegetation. Ref. "Gray activity of the rural areas is the herding of llamas (Lama 2005" is cited in the body but its bibliographic informa- glama) and sheep (Ovis aries)(Göbel2001; Yacobaccio tion is missing. Kindly provide its bibliographic infor- 2007;Lamas2012; Wawrzyk and Vilá 2013). Together mation in the list.Gray M. 2005. Geodiversity and with vicuñas (Vicugna vicugna), a wild South American geoconservation: what, why, and how?. Geodiversity camelid, these livestock species are the main regional & geoconservation,22(3):4-12. herbivores (Arzamendia et al. 2006; Vilá 2012; Previous studies describing the vegetation of the Arzamendia et al. 2014; Rojo 2017). The herding sys- Argentine Puna have been focused on its composition tems in the region are primarily based on the natural and cover (Cabrera 1957; Cabrera 1968; Ruthsatz and vegetation, whose main plant communities are shrub Movia 1975; Cabrera and Willink 1983; Castañeda and steppes. Peatlands, a local type of mountain wetlands González 1991; Bonaventura et al. 1995), and the known as “vegas” or “bofedales”, are also present in current knowledge regarding both its ecological areas with higher soil moisture (Ruthsatz and Movia (Arzamendia et al. 2006; Arzamendia et al. 2008, 1975;Cabrera1976; Caziani and Derlindati 1999; Talamo et al. 2010) and systematic aspects (Ospina Ruthsatz 2012). A key aspect when herders define live- González et al. 2013, 2016) is scarce and fragmentary. stock movements throughout the grazing sites is the Few studies have been conducted at a scale that can existing heterogeneity of vegetation communities along provide data which contribute to the development of the landscape (Göbel 2001). Several issues such as the rangeland management strategies and few detailed land degradation throughout the drier sector of the Puna maps of the vegetation community distribution in the region have emerged in the last years (Navone et al. area are available (Ruthsatz and Movia 1975; Cendrero 2006). One of the main reasons pointed out as a cause of et al. 1993). Attributes such as cover and biomass of this land degradation is overgrazing, but evidence to the natural vegetation, soil characteristics, and vegeta- confirm this correlation is scarce (Genin et al. 1995; tion indices (such as NDVI and SAVI) have not been Buttolph, and Coppock 2013). One of the barriers to thoroughly analyzed in the semiarid Puna of Argenti- understand the effect of grazing in the Puna region is the na. Most reference values for the Puna vegetation are lack of basic information of the vegetation dynamics at from Bolivia, where species composition and local local scale (Genin et al. 1995). environmental conditions are different from those in Like most drylands, the Puna shows wide intra- and Argentina (Navarro 1993; Genin et al. 1995; Alzérreca inter-annual climate variability with sequences of dry et al. 2001; Genin and Alzérreca 2006; Muñoz García and humid cycles (Aceituno 1996). Additionally, during et al. 2013). the past 100 years, this region has suffered significant The rough geomorphology and harsh climatic condi- climatic changes, characterized by a persistent warming tions of the Puna impose severe limitations on field trend and an increase in the elevation of the 0 °C iso- studies, especially those that require intense monitoring therm (Morales et al. 2012). and sampling over large areas. In this context, vegeta- The natural heterogeneity of geological, geomorphic, tion indices, which are remote-sensing tools, have been and soil characteristics is known as geodiversity and has extensively used to assess the vegetation in drylands been proposed as an important feature that evidences the worldwide (Jackson and Huete 1991; Bannari et al. complexity of natural systems (Gray 2005; Stavi et al. 1995; Sonnenschein et al. 2011; Xue and Su 2017). 2018). Regarding this issue, in the Pozuelos Basin, However, in the Argentine Puna, remote-sensing tools Jujuy, Argentina, Cendrero et al. (1993) found an altitu- have been used only at a large scale, to study the dinal range in which lower altitude areas present more regional physiognomic distribution and vegetation pro- developed soil and higher vegetation cover. In this ba- ductivity in the Puna region of Salta and Jujuy Provinces sin, a plan for improvement of sheep was implemented (Baldassini et al. 2012), but not to study the vegetation during the 1960s, with the introduction of new breeds. at the local scale. To conduct local-scale research, re- One of the results of this plan was an increase in the mote sensing together with field data may allow number of sheep in relation to that of llamas, with the complementing the characterization of different plant consequent increase in the risk of land degradation, as communities defined by their structural attributes Environ Monit Assess (2019) 191:635 Page 3 of 18 635 (Gould 2000). In Peru, the Normalized Difference Veg- (UNESCO-MAB Biosphere Reserve Directory). In this etation Index (NDVI) has been implemented to analyze Reserve, the main rural activity is the herding of sheep the vegetation trends in “bofedales” at a more precise and llama on the natural vegetation. scale (Mazzarino and Finn 2016), but the wetter envi- The study site was located at an altitudinal gradient ronment of the Peruvian Puna as compared with the through the following morphodynamic environments of Argentine Puna hinders the possibility of result compar- the reserve: western sierras (main massif and fluvial isons between these two regions. In Bolivia, Genin and valley), piedmont and paleo-lake (as defined by litholo- Alzerreca (1995) also implemented the NDVI to analyze gy, superficial deposits, and plant cover), which consti- the vegetation, using NOAA-AVHRR images. This con- tutes the buffer and transition area of the reserve, where stituted one of the first attempts at integrating remote pastoral activities are allowed (Cendrero et al. 1993). sensing and fieldwork while focusing on rangeland Local climatic conditions are cold, dry, and with a wide management in the region. The use of remote-sensing