Quantifying the hydrologic impacts of afforestation in Uruguay: A paired watershed study
Substantial portions of grassland are being converted to managed forestland in Uruguay. Long -term paired watershed studies in other continents indicate that water yields from managed forestlands will be reduced compared to water yields from grasslands. Very few long-term paired watershed studies to quantify changes in water yields due to afforestation have been conducted in South America. This study was initiated in 1999 to determine the hydrologic impacts of changing land use from grassland to pine plantation in Uruguay. Two adjacent watersheds (~100 ha) located in the Tacuarembó River basin were selected for the paired watershed study. Outflow rates were continuously measured on each watershed. Rainfall and meteorological conditions were also measured continuously on the site. During the initial pretreatment period (July 01, 2000 through June 2003) both watersheds were in pasture. In July 2003, one watershed (LC-PINE) was planted with loblolly pine, while the other (LC-PAST) remained in pasture. Data collection has continued through 2008. Reductions in annual water yields from the pine plantation have ranged from no reduction in the third year to a 28% reduction in the fourth year since tree planting. The year with the greatest yield reduction was characterized by a very dry period followed by a very wet period. The water yield reduction over the last three years of the study has been 15%. Distributions of outflow rates with time have also changed since tree planting. Peak flow rates from LC-PINE were reduced on average by 50% two years after planting and by 75% four years after planting. Times to peak flow rate at LC-PINE increased on average by 11 minutes two years after planting and by 26 minutes four years after planting. The Richards-Baker flashiness index was calculated using a six minute time step to determine the flashiness of both watersheds. The flashiness index for the LC-PINE watershed relative to the LC-PAST watershed was on average 50% lower during the fourth year after planting than during the pretreatment period. Flow duration curves for six minute intervals also show reduction in high flow rates from the LC-PINE watershed relative to the LC-PAST watershed. Baseflow rates from the forested watershed have not changed since planting. This study will continue through the pine growth cycle which will include pruning, thinning and harvest management practices typically used in the region.