Soil organic matter dynamics and crop productivity as affected by organic resource quality and management practices on smallholder farms

Abstract

Crop productivity in low-input agricultural systems is largely a function of the soil’s capacity to hold and release nutrients in soil organic matter (SOM). Although sandy soils on most Zimbabwean smallholder farms inherently contain a small amount of SOM, large variability in soil productivity (fertility gradients) exists between adjacent fields or field sections within the same farm. Farmer management of such variability remains a challenge and sustainable option for soil productivity are required. This study was based on the hypothesis that SOM, a renewable resource, is the driving force behind sustainable crop productivity on depleted sandy soils. Relationships between maize yields and SOM contents, nitrogen (N) release patterns, and their links with organic matter management practices by farmers differing in resource endowment were evaluated under different Natural Regions (NR) in the smallholder farming areas of Chikwaka (NR II: >750 mm yr-1), Chinyika (NR III: 650-750 mm yr-1) and Zimuto (NR IV: 450-650 mm yr-1). The cumulative effect of applying known quantities of different quality organic resources on SOM formation and maize productivity was also evaluated on-station at Domboshawa (NR II) and Makoholi (NR IV) Experimental Stations following incorporation of sunnhemp (Crotalaria juncea) green manure, calliandra (Calliandra calothyrsus) prunings, cattle manure, maize (Zea mays) stover and pine (Pinus patula) sawdust into soil. Farmers’ perception of soil productivity was consistent with laboratory indices across the different rainfall zones. Criteria for ranking the most productive ‘rich’ and least productive ‘poor’ fields ranged from colour through elements of soil structure to crop response following external nutrient inputs. Laboratory analysis showed that rich fields contained significantly more soil organic carbon (SOC) ranging between 5 - 8 g kg-1, compared with between 3 - 6 g kg-1 for designated poor fields. Differences in SOC contents between rich and poor fields were wider in the old communal areas of Chikwaka and Zimuto with >70 years of smallholder farming than in Chinyika (<25 years), suggesting that the observed fertility gradients are a cumulative effect of years of differential management practices by the different farmer classes. Overall, rich fields received between 0.3 - 13 t C ha-1 compared to 0.1 - 6 t C ha-1 for poor fields with resource endowment apparently dictating the intensity of use. Organic inputs with a C:N ratio >25 (the bulk of available resources on-farm) contributed significantly to overall particulate organic matter (POM) size in sandy soils. The intensity of C management was reflected more in meso- POM (53-250 μm diameter) compared to the macro-POM (250-2000 μm diameter) fraction suggesting that the larger POM fraction has a high turnover and is not protected from degradation. However, early season (within five weeks of incorporation) N availability from these materials was low (<5% of added N) resulting in poor maize performance during the vegetative phase. This may justify the high organic matter loading strategy of up to 50 t ha-1 employed by those farmers who often achieve yields of >3 t ha-1 on coarse sands. Practical management options for smallholder farmers who usually access low quality resources may include pre-application treatments such as composting or organic/mineral N fertilizer combinations to enhance N availability. The overall size of the organo-mineral fraction (<53 μm diameter) in these soils was small (<250 g kg-1 soil) and stable, and was not influenced by quality and quantity of C inputs and time over which they had been applied. High quality organic materials (e.g. sunnhemp) apparently enhanced the N-supply capacity of the organo-mineral fraction without necessarily increasing its size. However, such materials (C:N <25) released between 15-25% of added N within five weeks of incorporation, suggesting that a significant proportion of N is lost before uptake. The challenge is to enhance the efficiency with which N release from high quality materials can be managed. Maize productivity, and most likely that of other cereals, on depleted sandy soils was related to within-season mineral N fluxes and labile POM fraction. Both factors were ii primarily a function of differential capacity to manage organic matter by different farmer classes. Sustenance of optimal maize yields on sandy soils may only be possible through regular supply of both high and low quality materials in combination with mineral fertilizers, particularly N. High quality C inputs are likely to enhance short-term nutrient supply capacity of a small organo-mineral fraction present, while slow decomposing materials would contribute towards the long-term maintenance of critical SOM pools.