Agriculture, Ecosystems and Environment 139 (2010) 736–741 Contents lists available at ScienceDirect Agriculture, Ecosystems and Environment journal homepage: www.elsevier.com/locate/agee Relationship between soil  15 N, C/N and N losses across land uses in New Zealand B.A. Stevenson a,∗ , R.L. Parfitt b , L.A. Schipper c , W.T. Baisden d , P. Mudge c a Landcare Research, Private Bag 3127, Hamilton, New Zealand b Landcare Research, Private Bag 11052, Palmerston North, New Zealand c Department of Earth and Ocean Sciences, University of Waikato, Hamilton, New Zealand d GNS Sciences, PO Box 31312, Lower Hutt, New Zealand article info Article history: Received 4 July 2010 Received in revised form 27 October 2010 Accepted 28 October 2010 Available online 26 November 2010 Keywords: Nitrogen isotopes Pastoral systems Dairy soils N loss Land-use change abstract Several of the major processes that result in N loss from soil (nitrification, ammonia volatilization, and denitrification) discriminate against 15 N and fractionate the stable N isotopes, thus  15 N of ecosystem components has been suggested as an indicator of ecosystem N leakiness. This concept has been applied more successfully to native systems (primarily forest) than to managed systems where N inputs are greater and N cycling processes have potentially been modified. We analysed 210 New Zealand soils (0–100 mm depth) from different land-use systems (increasing in intensity of land use management from indigenous, to plantation forestry, pasture under drystock, pasture under dairy, and cropping) for  15 N and measures of N availability (total N, C/N, and N mineralization) to determine whether increas- ing intensity of land use management would lead to increased soil  15 N values. Mean soil  15 N differed between land uses with intensively managed cropping having the highest mean soil  15 N (6.2‰) fol- lowed by dairy (5.4‰), drystock (3.8‰), forestry (2.8‰) and indigenous forests (2.1‰). Over all land uses there was a negative correlation between  15 N and the soil C/N ratio ( = -0.73) and regression analysis indicated a relatively strong linear relationship between  15 N and C/N (r 2 = 0.56, P < 0.001) when crop- ping sites (where significant loss of soil C had occurred), and sites with C/N ratio >18 were excluded. Typical N balances for each land use showed that total N loss (and in particular fractionating N losses from ammonia volatilization and nitrate leaching) also increased with increasing land-use intensity. Our results indicate that soil  15 N may be a useful tool in assessing potential N losses in different soils. © 2010 Elsevier B.V. All rights reserved. 1. Introduction Human activity has altered the nitrogen cycle through land- use change, agricultural intensification, and use of fossil fuels (Vitousek et al., 1997; Matson and Vitousek, 2006). The New Zealand landscape is dominated by managed land uses (pastoral land alone accounts for 45% of total land area), and the intensity of management of these lands has increased considerably during the last few decades. For example, N fertilizer addition has increased from approximately 50 Gg N in 1989 to 335 Gg N in 2005 (Parfitt et al., 2006). Increasing N additions may deplete the soils ability to store additional N (Schipper et al., 2004), and loss of reactive N from soil to waters and the atmosphere is becoming increas- ingly problematic, particularly in areas where intensification of agriculture is occurring (Galloway et al., 2008; Parliamentary Commissioner for the Environment, 2004). Indicators of the extent of N enrichment of soils, and of potential N loss would be beneficial ∗ Corresponding author. Tel.: +64 859 3797; fax: +64 859 3701. E-mail address: stevensonb@landcareresearch.co.nz (B.A. Stevenson). to establish where remedial practices to reduce N loss to surface and ground waters should be focused. Robinson (2001) suggested that  15 N (the ratio of 15 N to 14 N in relation to atmospheric N 2 ) of ecosystem components repre- sents an integrated signal of ecosystem N processes because the isotopic composition of these components is driven by the iso- topic composition of inputs and outputs that can be modified by N fractionation within the system. Nitrification, denitrification and ammonia volatilization are processes that fractionate the N iso- topes in soil (Delwiche and Steyn, 1970), and loss of N through these processes will generally result in an increase of 15 N in the remaining ecosystem components (see Table 1). Thus the  15 N of ecosystem components can represent a measure of N leakiness or the degree of N saturation of a system (Pardo et al., 2006; Högberg, 1990, 1991). In a study of global patterns in N isotopes of soil and veg- etation, Amundson et al. (2003) suggested that  15 N soil [and  15 N (plant–soil) ] accurately portrays the sensitivity of ecosystems to increased rates of anthropogenically derived N, and implied that ecosystems with high  15 N soil should be considered sensitive to additional N inputs. Considerable research has been carried out in 0167-8809/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.agee.2010.10.020