Resource information
Surface runoff and associated erosion processes adversely affect soil and surface water quality. There is increasing evidence that a sound understanding of spatial-temporal dynamics of land use and management are crucial to understanding surface runoff processes and underpinning mitigation strategies. In this review, we synthesise the effects of (1) temporal patterns of land management of individual fields, and (2) spatio-temporal interaction of several fields within catchments by applying semivariance analysis, which allows the extent and range of the different patterns to be compared. Consistent effects of management on the temporal dynamics of surface runoff of individual fields can be identified, some of which have been incorporated into small-scale hydrological models. In contrast, the effects of patchiness, the spatial organisation of patches with different soil hydrological properties, and the effects of linear landscape structures are less well understood and are rarely incorporated in models. The main challenge for quantifying these effects arises from temporal changes within individual patches, where the largest contrasts usually occur in mid-summer and cause a seasonally varying effect of patchiness on the overall catchment response. Some studies indicate that increasing agricultural patchiness, due to decreasing field sizes, reduces the catchment-scale response to rainfall, especially in cases of Hortonian runoff. Linear structures associated with patchiness of fields (e.g. field borders, ditches, and ephemeral gullies) may either increase or decrease the hydraulic connectivity within a catchment. The largest gap in research relates to the effects and temporal variation of patch interaction, the influence of the spatial organisation of patches and the interaction with linear structures. In view of the substantial changes in the structure of agricultural landscapes occurring throughout the world, it is necessary to improve our knowledge of the influence of patchiness and connectivity, and to implement this knowledge in new modelling tools.
