Ecological Archives A022-045-A2

Cristian Correa and Andrew P. Hendry. 2012. Invasive salmonids and lake order interact in the decline of puye grande Galaxias platei in western Patagonia lakes. Ecological Applications 22:828–842.

Appendix B. Deforestation for pasture in Aysén, Chilean Patagonia.

The land-cover in Aysén underwent an extensive transformation during the first half of the 20th century, due to the sudden arrival of settlers and livestock companies sponsored by the government. Broad extensions of burnt trunks scattered on pastures are still an eloquent proof of the dramatic changes to the landscape (Fig. B1). In order to study the long-tem effects of deforestation on galaxiid fishes, we calculated the % area deforested within the basins of the lakes sampled, and used it as an explanatory variable of galaxiid density.

FigB1
 

   FIG. B1. Pasture in deforested area by Lago Castor. Note the old, burnt trunks still scattered, and patches of secondary forest in the background. Photo by C. Correa, January 2009.


 

We based our land-cover classification on a digital map from CONAF-CONAMA (1999). We used Gajardo’s (1994) ‘vegetation associations’ as a easonable hypothesis of the composition and distribution of forests in Aysén before colonization (Yarrow and Torres 2009). [The classification of the Aysén region sensu Gajardo was provided by CONAF, along with the attribute table of their main data set (CONAF-CONAMA 1999).] Within the basins of the 25 lakes sampled in this study, eight of Gajardo’s vegetation associations were represented; these were re-classified to distinguish originally forested areas from areas naturally devoid of forests (a few additional naturally-devoid-of-forest areas were identified using CONAF-CONAMA 1999). Potentially forested areas comprised evergreen and deciduous forests, while areas naturally devoid of forest high Andean peaks, rivers, lakes, Patagonian steppe, high-mountain shrubs, and outcrops.

We then used data from the a relatively recent assessment of land-cover in the region (CONAF-CONAMA 1999) to quantify deforestation. First, we reclassified the land use data into six categories: old-growth native forest, secondary native forest (regrowth), pasture, shrub, exotic forest, and naturally devoid of forest (Table B1). Then, we calculated the % of lakes’ basins covered by each of these categories, excluding from the analysis the areas naturally devoid of forest (see previous paragraph). We treated pastures, shrubs (but Gajardo’s high-mountain shrubs), and exotic forest (CONAF-CONAMA 1999) as deforested areas, while regrowth forest as native forest, assuming that substantial forest ecosystem functions have, for decades, been restored.

The results indicate little to complete deforestation within the basins of the lakes studied, with a median deforestation of 49 % (Table B2; Fig. B2). The proportions of deforested area per basin were transformed before being used as explanatory variable (see main article).

FigB2
 

   FIG. B2. Deforestation in the study area. Besides lakes (blue), all colored areas were formerly forested; gray indicates areas naturally devoid of forest such as high peaks and Patagonian steppe, or no data (in Argentina). Black contours show boundaries of the lake basins studied.


 

TABLE B1: Reclassification of land-cover data.

CONAF-CONAMA 1999

Correa and Hendry, this article

Field: DESC_USO

Field: RECENT_COV

ROTACION CULTIVO-PRADERA

Pasture

ESTEPA PATAGONICA

Naturally devoid of forest

MATORRAL ABIERTO

Shrub

MATORRAL ARBORESCENTE ABIERTO

Shrub

MATORRAL ARBORESCENTE SEMIDENSO

Shrub

MATORRAL PRADERA ABIERTO

Shrub

MATORRAL PRADERA SEMIDENSO

Shrub

MATORRAL SEMIDENSO

Shrub

PRADERAS PERENNES

Pasture

BOSQUE NAT.ACHAPARRADO ABIERTO

Old-growth native forest

BOSQUE NAT.ACHAPARRADO DENSO

Old-growth native forest

BOSQUE NAT.ACHAPARRADO SEMIDENSO

Old-growth native forest

BOSQUE NAT.ADULTO-RENOVAL ABIERTO

Old-growth native forest

BOSQUE NAT.ADULTO-RENOVAL SEMIDENSO

Old-growth native forest

BOSQUE NATIVO ABIERTO

Old-growth native forest

BOSQUE NATIVO ADULTO DENSO

Old-growth native forest

BOSQUE NATIVO ADULTO SEMIDENSO

Old-growth native forest

BOSQUE NATIVO ADULTO-RENOVAL DENSO

Old-growth native forest

PLANTACION

Exotic forest

RENOVAL ABIERTO

Regrowth native forest

RENOVAL DENSO

Regrowth native forest

RENOVAL SEMIDENSO

Regrowth native forest

ÑADIS HERBACEOS Y ARBUSTIVOS

Naturally devoid of forest

OTROS TERRENOS HUMEDOS

Naturally devoid of forest

VEGAS

Naturally devoid of forest

AFLORAMIENTOS ROCOSOS

Naturally devoid of forest

CORRIDAS DE LAVA Y ESCORIALES

Naturally devoid of forest

TERRENOS SOBRE LIMITE VEGETACION

Naturally devoid of forest

NIEVES

Naturally devoid of forest

LAGOS-LAGUNAS-EMBALSES-TRANQUES

Naturally devoid of forest

RIOS

Naturally devoid of forest

 

TABLE B2: Recent land-cover within the basins of the 25 lakes studied (CONAF-CONAMA 1999). Only areas forested before colonization were included in this analysis.

Land-cover

Land-cover modified from CONAF-CONAMA (1999)

Mean, median, and range % cover

Deforested

Pasture

21, 2 (0 – 63)

Shrub

27, 9 (0 – 93)

Exotic forest

0, 0 (0 – 5 )

Subtotal deforested

48, 47 (4 – 93)

Native forest

Old-growth

41, 45 (0 – 91)

Regrowth

12, 8 (0 – 39)

Subtotal native forest

52, 53 (7 – 96)

 

LITERATURE CITED

Arnold, J. G., R. Srinivasin, R. S. Muttiah, and J. R. Williams. 1998. Large area hydrologic modeling and assessment part I: Model development. Journal of the American Water Resources Association 34:73–89.

Beyer, H. L. 2004. Hawth's Analysis Tools for ArcGIS. Available at http://www.spatialecology.com/htools. (Accessed October 2009).

CONAF (Corporacion Nacional Forestal de Chile), and CONAMA (Comisión Nacional del Medio Ambiente). 1999. Catastro y evaluación de los recursos vegetacionales nativos de Chile. Informe nacional con variables ambientales. CONAF-CONAMA, Santiago, Chile.

Gajardo, R. 1994. La vegetación natural de Chile. Clasificación y distribución geográfica. Editorial Universitaria, Santiago, Chile.

Gassman, P., M. Reyes, C. Green, and J. Arnold. 2007. The soil and water assessment tool: historical development, applications, and future research directions. Transactions of the ASABE 50:1211–1250.

Jarvis, A., H. I. Reuter, A. Nelson, and E. Guevara. 2008. Hole-filled  seamless SRTM data V4, International  Centre for Tropical  Agriculture (CIAT). http://srtm.csi.cgiar.org

Yarrow, M., and M. Torres. 2009. The ecological and cultural landscape of the Aysén river basin. Pages 341–356 in R. Neves, J. Baretta, and M. Mateus, editors. Perspectives on integrated coastal zone management in South America. IST Press.


†  We obtained lake basins and stream networks from the hydrology model Soil and Water Assessment Tool (SWAT), implemented in the ArcGIS extension ArcSWAT 1.0.7 (Arnold et al. 1998; Gassman et al. 2007). Input data were the digital elevation model (DEM; Jarvis et al. 2008) and a parameter that defined the minimum drainage area necessary to originate a stream, which we set to 200 ha. We enhanced the fit by editing small cell-clusters of the DEM, at four locations. Lake polygons (CONAF-CONAMA 1999) were then superimposed onto the new stream network to guide the manual placement of lakes’ outlets points, which were subsequently selected in ArcSWAT to obtain lake basins’s polygons.

‡  Using the reclassification in Table B1, we created a thematic raster, with cell size equal to that of the DEM. Then, we based the calculation of land-cover proportion within basins on this raster and the basins’ polygons, using Hawth’s tools (Beyer 2004).


[Back to A022-045]