Responses of a nitrogen-saturated forest to a sharp decrease in nitrogen input

The reversibility of induced N saturation was investigated in a 46- yr-old pine (Pinus sylvestris L.) forest in northern Sweden Ammonium nitrate has b een applied annually since 1971 to plots (30 by 30 m) at average dosages of 36 (N1), 72 (N2), and 108 (N3) kg N ha(-1) yr(-1): with or without P and K addition (background N deposition is <4 kg ha(-1) yr(-1)). In 1990, after two decades of treatment, the largest N application (N3) was suspended, whi le N1 and N2 still received ammonium nitrate applications. Seven gears afte r the last application in N3, the N availability measured as N concentratio n in plants spine roots and needles and in leaves of the grass Deschampsia flexuosa (L.) Trin.] and activity of the enzyme nitrate reductase in leaves of D. flexuosa, and N-15 uptake by excised pine roots, was at the same lev els as in N1, although more than twice the amount of N has been applied in total to N3. The arginine concentrations in pine needles, concentrations of exchangeable mineral N in the organic layer and the uppermost 20 cm of the mineral soil were at the same levels as in the control plots. Thus, an exp erimentally induced N excess was, according to these measurements, to a hig h degree reversed 7 yr after the last N application. However, the compositi on of the understory vegetation still differed markedly from the untreated control 8 yr after the last N3 application.