Nutrient balances in African land use systems across different spatial scales: A review of approaches, challenges and progress

Nutrient balances are useful tools as indicators of potential land degradation and for optimizing nutrient use, and are thus highly relevant in the African context. A comprehensive literature review on nutrient balances in Africa was carried out to illustrate the main approaches, challenges, and progress, with emphasis on issues of scale. The review showed nutrient balances being widely used across the continent. The collected dataset from 57 peer-reviewed studies indicated, however, that most of the balances were calculated at plot and farm scale, and generated in East Africa. Data confirmed the expected trend of negative balances in the continent for nitrogen and potassium, where >75% of selected studies had mean values below zero. For phosphorus only 56% of studies showed negative mean balances. Several cases with positive nutrient balances indicated that soil nutrient mining cannot be generalized across the continent. Land use systems of wealthier farmers mostly presented higher nitrogen and phosphorus balances than systems of poorer farmers (p < 0.001). Plots located close to homesteads also usually presented higher balances than plots located relatively farther away (p < 0.05). Partial nutrient balances were significantly higher (p < 0.001) than full balances calculated for the same systems, but the later carried more uncertainties. The change in magnitude of nutrient balances from plot to continental level did not show any noticeable trend, which challenges prevailing assumptions that an increasing trend exists. However, methodological differences made a proper inter-scale comparison of results difficult. Actually, the review illustrated the high diversity of methods used to calculate nutrient balances and highlighted the main pit-falls, especially when nutrient flows and balances were scaled-up. Major generic problems were the arbitrary inclusion/exclusion of flows from the calculations, short evaluation periods, and difficulties on setting of spatial-temporal boundaries, inclusion of lateral flows, and linking the balances to soil nutrient stocks. The need for properly describing the methods used and reporting the estimates (i.e. appropriate units and measure of variability and error) were also highlighted. Main challenges during scaling-up were related to the type of aggregation and internalization of nutrient flows, as well as issues of non-linearity, and spatial variability, resolution and extent, which have not been properly addressed yet. In fact, gathered information showed that despite some few initiatives, scaling-up methods are still incipient. Lastly, promising technologies and recommendations to deal with these challenges were presented to assist in future research on nutrient balances at different spatial scales in Africa and worldwide.