Our work described in Kane et al. (2008) was a first attempt to examine the effect of precipitation patterns and seasonality on net dissolved inorganic nitrogen (DIN) retention in watersheds across the Long Term Ecological Research (LTER) network and related sites. The principal result of this study was that net DIN retention, quantified as atmospheric DIN import minus riverine DIN export, decreased with increasing variation in precipitation across eight diverse sites in North America and Puerto Rico. The comment by Lajtha (2009) raises important points mainly associated with interpreting a large body of watershed biogeochemical data from diverse sites when spatial autocorrelation and statistical independence are issues of concern.Lajtha notes that the spatial pattern of N deposition coincided with the spatial pattern of seasonality of precipitation, leading to a spurious correlation between seasonal precipitation variability and wet N deposition. This is an important point to consider, as wet N deposition rates varied by over an order of magnitude across the regions we investigated (table III in Kane et al., 2008). Thus, even if DIN exports across sites were similar, 'pristine' areas with markedly low rates of N deposition (e.g. BNZ, 0Ð3 kg N ha 1 year 1 ) could appear to have very low retention capacity compared to areas of high N deposition. As such, a spatial correlation between N deposition and precipitation seasonality could mislead interpretation (figure 6 in Kane et al., 2008). Although spatial autocorrelation may have influenced the exact * Correspondence to: E. S. Kane, Michigan Technological University, School of Forest Resources and Environmental Science, Houghton, MI 49931, USA. E-mail: eskane@mtu.edu nature of the relationship between DIN retention and precipitation seasonality among study sites, there are two reasons why we believe the overall message of our paper remains valid. First, when we consider only the relationship between variation in seasonal precipitation and seasonal DIN retention for sites exhibiting similarly high N deposition rates, the relationship remains significant (Figure 1). Second, besides net DIN retention we also reported the relationship between variation in (seasonal) precipitation and net DIN retention efficiency (net DIN retention/DIN import). This analysis also showed a negative trend between DIN retention and precipitation variability across sites. Taken together, the data and findings discussed herein reinforce precipitation variability as a control over DIN retention, while also substantiating the need for caution in interpreting input-output budgets at multiple scales.To asses the validity of this relationship in a more quantitative way, one needs to determine the degree of spatial correlation among explanatory variables in our data set. Quantification of spatial autocorrelation could be done with Moran's I (Moran, 1950) and Geary's C (Geary, 1954) statistical indices. As pointed out by Lajtha, we did not assess the spatial inferences of each study site on the stat...