A model for catchment soil erosion management in humid agricultural environments

Abstract

Intensive agricultural practices have critically contributed to the global increase in soil erosion and sediment fluxes. To reduce the impact of these practices, models able to represent the effect of changes in agricultural land‐use, farming and conservation practices are needed. Moreover, simulations spanning multi‐decadal periods can overcome the potentially confounding influence of climate variability on shorter‐term studies of impacts from agricultural change. Conceptual erosion models, such as the Morgan–Morgan–Finney (MMF) model, allow simulation of soil erosion rates and sediment fluxes over longer periods, while still retaining a general description of runoff and sediment generation processes. In addition, the modified‐MMF (MMMF) model offers improved representation of vegetation cover effects through measurable plant properties. However, as an annual model, the MMF model does not capture seasonal variability in climate, hydrology and land cover. Here, we propose a new model for humid environments based on the MMF model to address its limitations and improve its predictive ability, while retaining its simplicity and low computational and parameterization requirements. This includes monthly computation, representation of catchment hydrology based on delineation of saturated areas according to the topographic wetness index (TWI), and improved representation of ground and vegetation cover effects. The proposed model, MMF‐TWI, was applied in an agricultural catchment in the UK and performance compared to published data and MMMF simulation results. Land cover spatial and temporal variability, crop type as well as farming and conservation practices were found to have a significant influence on simulated sediment yields. Our findings demonstrate: (a) that MMF‐TWI model improves predictive ability compared to the MMMF model in humid environments, (b) the importance of capturing sub‐annual variability in climate, saturated areas and land cover, (c) the ability of MMF‐TWI model to represent impacts from farming and conservation practices, and (d) the potential for MMF‐TWI model to be applied as a soil erosion management tool. Copyright © 2017 John Wiley & Sons, Ltd.

Simulating a century of soil erosion for agricultural catchment management

“…We couple the reconstructions with a new catchment‐scale soil erosion model, MMF‐TWI, which is designed for use in humid agricultural environments (Peñuela et al ., ). MMF‐TWI is based on the Morgan–Morgan–Finney model (Morgan and Duzant, ), but incorporates new processes to capture sub‐annual variability in hydrology, vegetation cover and land management practices (Peñuela et al ., ). MMF‐TWI represents a compromise between process‐based models with higher parameter and computing demands and empirical models based on observations from certain regions (Prosser et al ., ; Merritt et al ., ).…”

“…We accessed records from stations that range 8–13 years in duration and 2.5–17 km from the lakes. We use these records to calculate the mean of the maximum monthly 30‐minute rainfall intensity ( I 30 , in mm h ‐1 ) for storm events discretized using the Rainfall Intensity Summarization Tool (RIST) (USDA, ) and to compute direct throughfall energy in MMF‐TWI on a mean monthly basis (Peñuela et al ., ). Storms were considered discrete events when separated by a six‐hour period with precipitation of less than 0.05 inches (1.27 mm) (Renard et al ., ).…”

“…MMF‐TWI is a conceptual catchment‐scale soil erosion model (Peñuela et al ., ) that builds on the Morgan–Morgan–Finney (MMF) set of models (Morgan et al ., ; Morgan, ; Morgan and Duzant, ). The modelling approach adopts a simplified representation of surface runoff and erosion processes while avoiding the greater parameter and computational demands of more physically‐based models.…”

“…We couple the reconstructions with a new catchment‐scale soil erosion model, MMF‐TWI, which is designed for use in humid agricultural environments (Peñuela et al ., ). MMF‐TWI is based on the Morgan–Morgan–Finney model (Morgan and Duzant, ), but incorporates new processes to capture sub‐annual variability in hydrology, vegetation cover and land management practices (Peñuela et al ., ).…”

“…MMF‐TWI represents a compromise between process‐based models with higher parameter and computing demands and empirical models based on observations from certain regions (Prosser et al ., ; Merritt et al ., ). As a conceptual model, MMF‐TWI provides a general description of runoff and erosion process while keeping computation and data requirements low (Peñuela et al ., ). This makes MMF‐TWI well suited for simulating past runoff, erosion and catchment sediment exports over centennial timescales using available historic datasets.…”