Search

Despite the clear evidence of competition for water between trees and crops, there have been very few studies comparing simultaneous water use by differing tree species in drylands. Comparative water use by dryland trees was measured in Senegal using heat balance gauges at the end of the wet season and in the dry season. Significant differences between tree species were found for maximum rates of water use per unit leaf area. Indigenous species may be better adapted to the dry environment than exotic species but the indigenous species Acacia seyal Del. used more water per unit leaf area than all other species. The exotic species Azadirachta indica Adr. Juss. consistently used less water per unit leaf area than most other species. There were significant differences in amounts of water used per unit leaf area by differing provenances of the same tree species. Water use in the dry season varied by a factor of three between two provenances of Acacia aneura F. Muell ex Benth. indicating potential to select provenances for drylands based on their water use characteristics. Absolute rates of water use as well as differences in sapflow between species were greatest when soils were moist suggesting that comparative sapflow studies will be most informative when carried out during the wet season. Water use rankings of the differing tree species were broadly maintained irrespective of season.

Survival, biomass production, nutrient use and fertility of soil were examined in experimental stands of 10-year-old N-fixing indigenous Acacia nilotica (L.) Del. (eight provenances), A. senegal (L.) Willd. (one provenance), Acacia tortilis (Forsskal) Hayne (six provenances), Acacia raddiana syn A. tortilis (Forssk.) Hayne ssp. raddiana (Savi) Brenan (three provenances) and exotic N-fixing A. holosericea A. Cunn. ex G. Don (one provenance), A. cowleana Tate (one provenance), Acacia aneura F. Muell. ex Benth. (twelve provenances), Prosopis juliflora (Sw.) DC. (one naturalised provenance), Prosopis cineraria (L.) Druce (five provenances) and P. chilensis (Molina) Stuntz (two provenances) growing on a coastal site in semi-arid Senegal. Similar data were obtained for the non-N-fixing fast-growing exotic species Azadirachta indica Adr. Juss. and Eucalyptus camaldulensis Dehnh. of the same age that were growing on adjacent plots. P. juliflora produced more biomass than the other species and showed the greatest potential for use as a multipurpose tree. A. aneura showed the greatest potential for fodder production and despite its small stature, it ranked second for biomass among the N-fixing species. Of the indigenous species, A. nilotica was the most efficient user of non-renewable elements in biomass production. A. raddiana seemed profligate in its consumption of both N and P in wood production. E. camaldulensis was efficient in its use of N in wood but its wood contained large concentrations of P and K. A. aneura, P. juliflora and A. nilotica had the smallest concentrations of P in wood. A. raddiana and A. indica contained large concentrations of K in their wood in contrast to A. aneura and A. nilotica which were the most efficient users of K in wood production. There were few significant differences in fertility in soils beneath the differing tree species. Nevertheless, plots containing P. juliflora and P. cineraria tended to be the most fertile. Soil beneath A. indica was more fertile than soil beneath some N-fixing species. The extent of removal of plant nutrients in biomass suggests that without inputs of nutrient elements, harvest of wood and foliage is unlikely to be sustainable at the measured rate of production for more than two tree rotations of 10 years duration.