Plant regeneration at varying levels of ecosystem recovery: a case of Namulonge wetland system, Uganda

A dissertation submitted to the Directorate of Research and Graduate Training in partial fulfillment of the requirements for the award of the degree of Master of Science in Botany of Makerere University

Namulonge wetland system has been exposed to various land use changes. This study, supported by the GlobE wetlands of East Africa, was conducted at National Crop Resource Research Institute (NCRRI), Namulonge, in a modified wetland with three distinct fallow patches of known age, i.e., short-term fallow sites, medium term and long-term fallow sites. The aim was to investigate the level of wetland habitat recovery in terms of species composition, vegetation structure and biomass changes after staggered periods of fallowing. Variation in wetland plant species composition, vegetation structure and biomass were studied using eight transects located on a vegetation change gradient. A replica was also done in a natural wetland system as a reference site. Along each transect, samples were taken ineight quadrats from which data collection on vegetation biomass and other attributes was done every after eight weeks for a period of 24 weeks. Fisher‟s alpha (α), and Shannon–Wiener (H′) indices were used to calculate the alpha diversity at each fallow scale. One-way Analysis of Variance (ANOVA) was used to ascertain the differences in species diversity between the three fallow periods. Analysis of Similarity (ANOSIM) was adopted to calculate the variation in community species composition, and a pairwise test was carried out to determine the fallow sites that differ significantly from the other. The Kruskal-Wallis test was used to assess whether there is a significance difference in the plant biomass across fallow periods and across a moisture gradient. Percentage relative frequency of each species was adopted as a measure of relative distribution of the species within both fallow and natural wetland sites, and Pearson correlation coefficient adopted to explore the relationship between the most evenly distributed species between fallow and natural wetland site. To determine whether there is a significant difference in vegetation attributes between two sites, a paired t-test was adopted. The primary study area (fallow wetland patches) comprised of (i) cropland planted with rice, maize, sweet potatoes, and vegetables, (ii) fallow sites dominated by grasses and herbs, and (iii) seasonally flooded grassland fallow dominated by Phoenix reclinata and light bushes of Triumfeta macrophylla, Bridelia micrantha and Sesbania sp. The natural wetlands within the same environment setting comprised swampy bushed grassland, and papyrus swamp. Altogether, 215 plant species were recorded from both fallow and natural wetlands, of which 155 species were recorded from the fallow sites and 164 species from the control site. There was a very weak positive relationship in species composition between and within sites fallowed for short- and medium-term periods (r = 0.0752, p=0.132,). Plant composition and abundance between sites fallowed for shorter periods was significantly differed from those fallowed for longer periods (r = 0.123, p=0.045,). Sites fallowed for short term and medium term periods share, to a greater extent, a number of species with a very high similarity index in the plant species composition. Sites fallowed for shorter periods were more diverse than medium and long-term fallowed sites, Shannon-Wiener (Ҥ) =3.344 and Fisher‟s alpha (α)=25.85. There was no significant difference in the plant biomass across the short-term, medium term and long-term fallow sites (ꭓ2= 3.0927, p=0.213, df=2). Leersia hexandra was evenly distributed within both fallow and natural wetland sites with percentage relative frequency of 81% in fallowed area and 79.4% in natural sites. There was a very low positive correlation in species distribution between fallowed and natural wetland sites (r=0.059, n=33, p = 0.3793). Leersia hexandra, Desmodium salicifolium, Cyperus papyrus, Ipomoea cairica, Thunbergia alata, Cylosorous interruptus, Triumfetta macrophylla and Mimosa pigra were evenly distributed within the natural wetland with relatively high relative cover, indicating high degree of dominance than other species. On the other hand, Leersia hexandra, Crassocephalum vitellinum, Panicum trichocladum, Panicum maximum, Desmodium salicifolium, Ludwigia abyssinica, Paspalum scrobiculatum, Cyperus latifolius, and Ageratum conyzopides were found evenly distributed within the degraded sites. These species are early wetland colonizers which when left undisturbed pave a way to a more stable wetland macrophyte community provided the wetland moisture is not drained out but rather improved. In terms of species diversity, there was a significant different in plant species diversity between fallow and natural (r= 0.321, p< 0.001, df = 63). Human activities especially agriculture have a great bearing on the primary productivity, functionality, and integrity of wetlands but if the disturbance is removed the wetland left to recover without any intervention, it attains the stable status within the shortest time possible. If wetlands are used on a rotation basis without totally dysfunction them, they can ably perform their functions and attain the desired state. In order to balance agricultural production/food security and use of wetlands, emphasis should be put on the time the wetland will be used without adversely affecting its ability to regains its natural functionality.

GlobE Wetlands Project of East Africa