| dc.description.abstract | This study was initiated to address degradation in the semi-arid rangeland of southern Kenya by assessing the level of pastoral and agro-pastoral communities' awareness, what they are doing about it, difficulties they face in re-vegetation and identify strategies to alleviate those difficulties. Participatory community surveys and feedback workshops, laboratory tests and field trials at station and on-farm level were conducted. Nine divisions across three districts, namely Makueni, Kajiado and Taita- Taveta were surveyed involving 353 households with diverse background and resource endowment. Communities confirmed that rangelands are degraded thus constraining livestock production. Approximately 79% (range 16-97% among divisions) of the communities knew about this degradation caused by a number of factors, principally continuous overgrazing and drought. About 35% of the respondents estimated that bare patches in excess of 50% in some of the grazing areas exist and soil erosion is evident. Previously dominant and important forage plants (mostly herbaceous and a few browse species) had either declined drastically or disappeared altogether. Over 70% of the households knew methods to stop further degradation and / or restore the rangeland integrity. These included deferred grazing, protecting bare patches with thorny branches, terracing and reseeding. About 62% of the households had tried at least one method. Of the above methods, reseeding was identified as the likely opportunity that is researchable on the basis that it was the most known among the communities yet it ranked third in adoption. However, reseeding had previously failed or had not been attempted at all because of inappropriate land tenure, labour costs, lack of or poor seeds, drought, poor establishment, loss by wind and runoff, low persistence and lack of know-how. Land tenure is ever-changing in the pastoral areas because while some communities are still in communal group ranches many have agitated and succeeded to sub-divide them into small family parcels. In the latter situation as in the agro-pastoral areas the family heads control the land and are mostly older and rarely accommodate new technologies. Younger family members are constrained as they cannot freely decide to implement their wishes in any part of the land. It is a social issue that calls for sustained public awareness and possibly exposure to success case studies. Labour costs constraint arises from failure to see tangible returns to investment because where, for instance, arable farming is concerned the same people would readily open up large acreage within a short time. So, there was need to demonstrate that investing in pasture is worthwhile through on-farm trials and field tours. The rest of the difficulties in reseeding were related to seed quality and establishment for which a series of experiments were designed and carried out. Seed unavailability meant testing the published techniques of harvesting local materials and storing them for future use without compromising viability. Further, lack of seeds meant that any material obtained from whatever source would be tried in most cases without success because plant species have different growth requirements, especially soils type and moisture. Where the right species is matched with right soil type and attendant moisture level, establishment could be poor depending on the planting technique used. Seeds offour grass species, namely Cenchrus ciliaris, Chloris roxburghiana, Enteropogon macrostacyus and Eragrostis superba, considered the most important by the - ....• communities were harvested using two recommended methods: (1) cutting the stalk of a mature panicle with the flag leaf, keeping them under shade for 3-5 days before threshing (2) stripping the seeds when dry. Seeds were put separately in four types of containers, namely aluminum tins, nylon sacks, brown paper bags, and white cotton cloth bags. They were stored in a traditional grain store. Sub-samples of the seeds were periodically collected and caryopses tested for viability through germination in Petri dishes for 14 days. Harvesting with stalks was superior to stripping, particularly for E. superba. Different species had different levels of viability right from harvesting time and over the storage period, regardless of method of harvesting. Daily seed germination increased by 37.5% on average with storage for all species. The overall best container material was the metal tins with 5.7% higher daily seed germination than the worst container which was the cloth bags. Seeds were sown in plastic pots filled with three soil types (rhodic FERRALSOL, orthic FERRALSOL and histic GLEYSOL) and subjected to four regimes of watering (250 ml daily, 500 ml every other day, 750 ml every third day and 1000 ml every fourth day) for 70 days. The highest seedling population was that of E. macrostachyus (31.27 ± 3.493) while the least was that of C. roxburghiana (7.63 ± 1.009), which was similar to that of E. superba. Seedling population did not differ between the FERALLSOLs but was four-fold that on the GLEYSOL. These differences persisted into maturity where tillering and biomass was lower on the GLEYSOL and not different between the FERRALSOLs. However, plants growing in the GLEYSOL suffered less moisture stress as they maintained a relatively higher level of greenness for a longer period. Watering every third day produced the highest seedling population (23.94 ± 3.612) and daily -- watering the least (8.02 ± 1.243). These results reflected soils' differences in water holding capacity and chemical activity, especially leaching and salts build up. On-farm trials involved sowing a mixture of the four grasses using two techniques: (l) broadcasting of seeds onto the soil surface among existing vegetation or bare ground, and (2) preparing the seedbed by pitting along contours and placing the seeds on the crescents of the pits. Sowing on pits had 89.7% more seedlings than broadcasting. However, individual species establishment varied from site to site due to variation in soil fertility and the concomitant variation in rainfall. Poor reseeding results could further be explained by introduction of un-adapted seed material. It is possible that seeds brought in, mostly from KARl - Kiboko, were of a different ecotype. Genotyping of C. roxburghiana using the random amplified polymorphic DNA (RAPD) technique found that there is a significant genetic variability within and between four natural populations, with the Kiboko population being the least diverse. Greater genetic diversity within a population where seed is obtained from implies a greater chance of obtaining a good establishment when reseeding. Possibly, other species have similar variation making it imperative that decisions on where to obtain seeds are more judicious. The overall conclusions are that southern Kenya rangelands are degraded and the communities are not only aware of this fact but also they have attempted to address it in a number of ways albeit at a limited level. Reseeding difficulties are of social, economic and biophysical kind. So, strategies to alleviate the problem of rangelands degradation must be of necessity be multi-disciplinary because each component is almost equally crucial. | en |
