Item Details

Title: Conservation Farming and Changing Climate: More Beneficial than Conventional Methods for Degraded Ugandan Soils

Date Published: 2017
Author/s: Drake N. Mubiru, Jalia Namakula, James Lwasa, Godfrey A. Otim, Joselyn Kashagama, Milly Nakafeero, William Nanyeenya and Mark S. Coyne
Data publication:
Funding Agency :
Copyright/patents/trade marks:
Journal Publisher: Sustainability
Affiliation: National Agricultural Research Organization (NARO), P.O. Box 7065, Kampala, Uganda; (J.N.); (J.L.); (G.A.O.); (J.K.); (M.N.); (W.N.)
2 Department of Plant and Soil Sciences, University of Kentucky, 1100 S. Limestone St., Lexington,
KY 40546-0091, USA;
Keywords: land degradation; land management; conservation farming


The extent of land affected by degradation in Uganda ranges from 20% in relatively flat
and vegetation-covered areas to 90% in the eastern and southwestern highlands. Land degradation
has adversely affected smallholder agro-ecosystems including direct damage and loss of critical
ecosystem services such as agricultural land/soil and biodiversity. This study evaluated the extent
of bare grounds in Nakasongola, one of the districts in the Cattle Corridor of Uganda and the yield
responses of maize (Zea mays) and common bean (Phaseolus vulgaris L.) to different tillage methods in
the district. Bare ground was determined by a supervised multi-band satellite image classification
using the Maximum Likelihood Classifier (MLC). Field trials on maize and bean grain yield responses
to tillage practices used a randomized complete block design with three replications, evaluating
conventional farmer practice (CFP); permanent planting basins (PPB); and rip lines, with or without
fertilizer in maize and bean rotations. Bare ground coverage in the Nakasongola District was 187 km2
(11%) of the 1741 km2 of arable land due to extreme cases of soil compaction. All practices, whether
conventional or the newly introduced conservation farming practices in combination with fertilizer
increased bean and maize grain yields, albeit with minimal statistical significance in some cases.