Deforestation and precipitation patterns in the arid Chaco forests of central Argentina

• Published in: Applied vegetation science Vol. 16; no. 2; pp. 260 - 271
• Main Authors: Hoyos, L.E., Cingolani, A.M., Zak, M.R., Vaieretti, M.V., Gorla, D.E., Cabido, M.R.
• Format: Journal Article
• Language: English
• Published: Lanna Blackwell Publishing Ltd 01.04.2013; Opulus Press; Wiley Subscription Services, Inc
• Subjects: Agriculture; Animal, plant and microbial ecology; Applied ecology; Biological and medical sciences; Cover types; Deforestation; Dry Chaco forest; Dry forests; Forest cover; Forestry; Fundamental and applied biological sciences. Psychology; Hydrologic data; Land cover; Landsat; Precipitation; Rainfall; Remote sensing; Shrublands; Soybeans; Variance analysis; Vegetation; Vegetation cover; South America; Argentina; America; Deforestation; Cover types; Forests; Arid environment; Space remote sensing; Plant ecology; Landsat; Rainfall; Landsat satellite; Dry Chaco forest; Rain; Precipitation; Plant sciences
• ISSN: 1402-2001; 1654-109X
• DOI: 10.1111/j.1654-109X.2012.01218.x

Aims: (1) to compare two series of precipitation data from different periods (1930–1950 and 1950–2000) in three sectors of the southern dry Chaco in the arid and semi-arid sub-regions; (2) construct maps showing the distribution of land-cover units for 1979, 1999, 2004 and 2010 for the same three sectors; and (3) assess the changes in land-cover units occurred between 1979 and 2010 in the three sectors. Location: Southern extreme of the dry Chaco in NE and NW Córdoba Province, central Argentina. Methods: We compared annual and growth period (November–March) precipitation among the three sectors and between two series of data corresponding to different periods (1930–1950 and 1950–2000) using repeated measures ANOVA, with the station as the subject variable, period as the within-factor and sector as the between factor. Using three Landsat MSS (1979) and nine Landsat TM (1999, 2004 and 2010) images we mapped the distribution of eight land-cover units for the whole study area. For each sector (NE, NW and W), we performed a change detection analysis between 1979 and 2010. Results: The classification of Landsat MSS and TM images resulted in reliable land-cover maps (overall accuracy 80%). Our results showed that vegetation cover in the area is highly disturbed and that the present status of vegetation cover differs among the three sectors. In the more humid sector, the land-cover changes have been dominated by replacement of closed forests by crops, while in the driest portion of the study area forest loss was not related to agriculture. Additionally, we found that significant increases in precipitation have occurred in all three sectors, but the increase was highest in the humid sector. Conclusions: The differences observed among the three sectors suggest that precipitation may have effectively played a dominant role in the process of forest conversion to agriculture.