Influences of the landscape pattern on riverine nitrogen exports derived from legacy sources in subtropical agricultural catchments

Changes in the landscape pattern can disturb legacy nitrogen (N) release by influencing hydrological and biogeochemical processes; thus, understanding the effects of landscape patterns on riverine N exports from legacy sources is critical for preparing water quality improvement strategies. In this study, an empirical statistical model that incorporates the net anthropogenic nitrogen input (NANI), runoff coefficient, and residential land area percentage was used to quantify the contribution of legacy sources to annual riverine ammonium nitrogen (NH4+–N), nitrate nitrogen (NO3−–N), and total nitrogen (TN) exports in eight adjacent agricultural catchments in subtropical southern China. The results indicated that annual riverine NH4+–N, NO3−–N, and TN exports from legacy sources ranged from 0.36–1.03, 3.09–4.89, and 3.94–6.79 kg ha−1 year−1, respectively, during the 2012–2017 period. Redundancy analysis (RDA) was used to analyze the interactions between legacy N release and landscape metrics at both the landscape and class levels. The RDA results suggested that higher dispersion, lower shape complexity, and greater heterogeneity in landscape patches can enhance the release of legacy N at the landscape level. In agricultural and residential areas, higher release of legacy N was associated with patches that are unfragmented and have a low shape complexity, whereas in woodland areas, the opposite was true. These analyses provide scientific support for preparing legacy N control strategies from the perspective of landscape ecology.