Litterfall production modeling based on climatic variables and nutrient return from stands of Eucalyptus grandis Hill ex Maiden and Pinus taeda L.

Native grasslands in the Pampas of South America are increasingly being replaced by Eucalyptus and Pinus stands. The short rotation regimes used for the stands require high nutrient levels, with litterfall being a major source of nutrient return. To model the litterfall production using climatic variables and assess the nutrient return in 14-year-old Eucalyptus grandis and Pinus taeda stands, we measured litter production over 2 years, using conical litter traps, and monitored climatic variables. Mean temperature, accumulated precipitation, and mean maximum vapor pressure deficit at the seasonal level influenced litterfall production by E. grandis; seasonal accumulated precipitation and mean maximum temperature affected litterfall by P. taeda. The regression tree modeling based on these climatic variables had great accuracy and predictive power for E. grandis(N = 33; MAE（mean absolute error） = 0.65; RMSE（root mean square error） = 0.91; R2 = 0.71) and P. taeda（N = 108; MAE = 1.50; RMSE = 1.59; R2 = 0.72）. The nutrient return followed a similar pattern to litterfall deposition, as well as the order of importance of macronutrients（E. grandis: Ca > N > K > Mg > P; P. taeda: N > Ca > K > Mg > P） and micronutrients（E. grandis and P. taeda: Mn > Fe > Zn > Cu） in both species. This study constitutes a first approximation of factors that affect litterfall and nutrient return in these systems.