The NERC and CEH trademarks and logos ('the Trademarks') are registered trademarks of NERC in the UK and other countries, and may not be used without the prior written consent of the Trademark owner. The success of agroforestry in semi-arid areas depends on efficient use of available water and 32 effective strategies to limit tree/crop competition and maximise productivity. On hillsides, 33 planting improved tree fallows on the degraded upper section of bench terraces is a 34 recommended practice to improve soil fertility while cropping continues on the lower terrace 35 to maintain food production. This study examined the influence of tree fallows on soil water 36 content ( w ) and evaporation (E s ). Alnus acuminata Kunth (alnus), Calliandra calothyrsus 37 Meissner (calliandra), Sesbania sesban L. (sesbania), a mixture of all three species, or sole 38 crops (beans (Phaseolus vulgaris L.) or maize (Zea mays L.)) were grown on the upper 39 terrace. The same sole crops were grown on the lower terrace. Four management regimes 40 (unpruned, root, shoot and root+shoot pruned) were applied to the tree rows adjacent to the 41 cropping area. Neutron probe and microlysimeter approaches were used to determine  w and 42 E s when the trees were c. 3.5 years old. Sesbania and alnus increased  w by 9-18% in the 43 cropping area on the lower terrace but calliandra reduced  w by 3-15%. After heavy rain, E s 44 comprised 29-38% of precipitation in the tree-based treatments and 53% under sole crops. 45Absolute values declined as rainfall decreased, but E s as a proportion of rainfall increased to 46 39-45% in the tree-based treatments and 62% for sole crops. Root+shoot pruning of alnus 47 and the tree mixture increased  w in the cropping area but had no significant effect in the 48 other tree-based treatments. The results suggest that sesbania and alnus can be planted on 49 smallholdings without compromising water supply to adjacent crops, whereas calliandra 50 decreased water availability despite reducing E s . These results provide a mechanistic 51 understanding of reported effects on crop yield in the same site. 52 53

This version available http://nora.nerc.ac.uk/6342/ NERC has developed NORA to enable users to access research outputs wholly or partially funded by NERC. Copyright and other rights for material on this site are retained by the authors and/or other rights owners. Users should read the terms and conditions of use of this material at http://nora.nerc.ac.uk/policies.html#access This document is the author's final manuscript version of the journal article, incorporating any revisions agreed during the peer review process. Some differences between this and the publisher's version remain. You are advised to consult the publisher's version if you wish to cite from this article. www.springerlink.comContact CEH NORA team at noraceh@ceh.ac.ukThe NERC and CEH trade marks and logos ('the Trademarks') are registered trademarks of NERC in the UK and other countries, and may not be used without the prior written consent of the Trademark owner. 2002, 2006, 2007Bayala et al. 2008). In semi-arid Kenya, Jackson et al. were not pruned to maximise woody biomass production and reflect the objectives of subsistence farmers. Sub-treatment plots (6 m wide x 5 m long)were randomly allocated in each main treatment. Sole crops were grown continuously on the lower terrace (12 m wide).Alnus and calliandra were planted in September 2000 using potted seedlingsand sesbania was planted in March 2001 using bare-rooted seedlings. The phased planting ensured that all species could be harvested simultaneously as sesbania, a shrubby species, matures sooner than calliandra and alnus, which are both trees. A single row of each species was planted in the tree mixture.Based on previous studies (Siriri and Raussen 2003), the least competitive species, sesbania, was situated adjacent to the crops, calliandra was planted in the central row, and alnus, believed to be the most competitive, was grown furthest from the crops. Main and sub-plots were separated by 4 and 2 m wide walkways to provide access and minimise interference (Fig. 1).A relatively mild pruning regime was chosen as a compromise between effective control of competition and maximum production of woody biomass and green manure for soil improvement. Pruning was implemented simultaneously for all tree species when calliandra and alnus were 12 months old and sesbania was six months old to avoid compromising the growth of young trees. Shoot pruning involved removing all branches from the lower third of the crown of trees adjacent to the cropping areas on the lower terrace and the sole crop plots on the upper terrace, and was repeated before each cropping season; prunings were returned to the plots from which they came.Root pruning was carried out to a depth of 30 cm when the trees were young and 50 cm when they were over three years old. Mean values for specific treatments provided by REML may vary depending on how treatments are structured in the analysis, providing an explanation for the differing mean crop yields shown in Tables 2 and 3. Table 2 compares crop yield adjacent to unpruned trees wit...

Terrace benches on the hillslopes of Western Uganda exhibit a crop-yield gradient: upper portions of terraces produce less than the lower sections. We investigated the soil factors responsible for this yield variation on 30 terraces along 5 toposequences in what was predominantly a Ferralsol. Two levels of spatial analysis were conducted: (1) variation within individual terraces; and (2) differences across hill-slope positions. A greenhouse experiment further examined this fertility gradient using soils extracted from the upper, middle and lower parts of the terrace. This included a nitrogen (N) treatment of 70 mg N kg À1 soil. In the fields sorghum (Sorghum bicolor) grain yield differed significantly (p 0Á05) across terraces, increasing from 0Á4 Mg ha À1 on the upper 40 per cent of terrace to 2Á5 Mg ha À1 on the lower 40 per cent. Soil bulk density decreased down the terraces from 1Á41 g cm À3 to 1Á18 g cm À3, causing an increase of hydraulic conductivity from 1Á6 cm h À1 to 7Á3 cm h À1, from the upper to the lower part, respectively. Organic carbon (C) and total N increased from upper to lower terrace sections. In the greenhouse, sorghum growing on soils from which the soil physical limitations have been removed did not show significant yield differences across a terrace as observed in the field. Response to N was most pronounced on the upper terrace sections, increasing sorghum dry matter from 3 g pot À1 to 15 g pot À1.On the lower terrace, N amendment increased dry matter from 4 g pot À1 to 11 g pot À1 . Technologies to improve the adverse soil physical conditions and the N limitations on upper terrace parts while preventing tillage-and soil-erosion-induced terrace scouring are required.

No abstract

No abstract