Characterizing hillslope-riparian-stream interactions: A scaling perspective, Maimai, New Zealand

Historically, much small catchment hydrology has focused on hillslopes and related flow processes. Concurrently, much riparian research has focused on nutrient transformation. Little research has attempted to integrate hillslope and riparian investigative approaches with catchment runoff in order to estimate their relative roles in stormflow and baseflow generation, the partitioning of old and new water, impact on stream water residence time, and subsequent control of stream chemistry. Furthermore, little is understood about landscape controls on streamflow generation and composition. To address this, we discretized the Maimai catchments into their main landscape components (hillslopes and riparian zones). Isolation and classification of the response characteristics of each landscape unit was possible through instrumentation of each landscape unit from a trenched and gauged hillslope to riparian zones at successively larger catchment scales. This landscape discretization approaches was tested with a continuity-based hydrometric model, as well as isotopic/solute mass-balance hydrograph separation techniques, to assess the spatial and temporal sources of catchment runoff. We quantified the sequencing of landscape unit contributions to catchment runoff. We used this landscape discretization approach to elucidate controls on DOC export dynamics. Results suggest that the relative timing of riparian and hillslope source contributions, connections and disconnections of dominant runoff contributing areas, as well as the internal dynamics of these zones, are a first approximation of catchment controls on catchment runoff DOC concentrations and mass export. We present a simple method for quantifying the local contributions of hillslope area and riparian area along a stream network based on gridded digital elevation data. The method enables us to compute characteristics of a catchment such as the distribution of inputs to the stream network and the riparian to hillslope area ratios along the stream network. For the entire catchment the ratio between riparian and hillslope area was 0.14. We calculated this ‘buffer-capacity’ along the stream network, and found the values were below 0.14 for 75% of the stream length and below 0.06 for 50% of the stream length. The combination of plot scale investigative approaches, landscape discretization, and a new method for landscape analysis provided insight into the first-order controls on runoff and solute dynamics and served as a framework for investigating the role of catchment scale and landscape organization in runoff generation.