Resorption proficiency and efficiency of leaf nutrients in woody plants in eastern China

(i) To explore variations in nutrient resorption of woody plants and their relationship with nutrient limitation and (ii) to identify the factors that control these variations in forests of eastern China. We measured nitrogen (N) and phosphorus (P) concentrations in both green and senesced leaves of 172 woody species at 10 forest sites across eastern China. We compared the nutrient resorption proficiency (NuRP) and efficiency (NuRE) of N and P in plant leaves for different functional groups; we further investigated the latitudinal and altitudinal variations in NuRP and NuRE and the impacts of climate, soil and plant types on leaf nutrient resorptions. On average, the leaf N resorption proficiency (NRP) and P resorption proficiency (PRP) of woody plants in eastern China were 11.1mg g(1) and 0.65 mg g(1), respectively; and the corresponding N resorption efficiency (NRE) and P resorption efficiency (PRE) were 49.1% and 51.0%, respectively. Angiosperms have higher NRP (are less proficient) values and lower NRE and PRE values than gymnosperms, but there are no significant differences in NRP, PRP and PRE values between species with different leaf habits (evergreen vs. deciduous angiosperms). Trees have higher NRE and PRE than shrubs. Significant geographical patterns of plant nutrient resorption exist in forests of eastern China. In general, NRP and PRE decrease and PRP and NRE increase with increasing latitude/altitude for all woody species and for the different plant groups. Plant functional groups show more controls than environmental factors (climate and soil) on the N resorption traits (NRP and NRE), while site-related variables present more controls than plant types on PRP and PRE. NRP increases and PRP and NRE decrease significantly with increasing temperature and precipitation for the overall plants and for most groups, except that significant PREclimate relationship holds for only evergreen angiosperms. Leaf nutrient resorption did not show consistent responses in relation to soil total N and P stoichiometry, probably because the resorption process is regulated by the relative costs of drawing nutrients from soil versus from senescing leaves. These results support our hypothesis that plants growing in P-limited habitats (low latitudes/altitudes or areas with high precipitation/temperature) should have lower PRP and higher PRE, compared with their counterparts in relatively N-limited places (high latitudes/altitudes or areas with low precipitation/temperature). Our findings can improve the understanding of variations in N and P resorption and their responses to global change, and thus facilitate to incorporate these nutrient resorption processes into future biogeochemical models.