The study investigates the nature and extent of erosion and deposition in the Maqalika catchment, Maseru. Components of the study include (i) the re-construction of erosion and land use history in the catchment from 1961 to 1985, (li) determination of dominant factors or combinations of factors responsible for the observed erosion distribution at each date, (iii) the identification and evaluation of erosion and deposition features, and their spatial and temporal variations, and (iv) identification of socia-economic problems associated with observed erosion and deposition. Methods employed for the re-construction of erosion and land use history include the use of sequential aerial photography of 1961, 1979 and 1985, orthophoto maps and review of literature from past studies. It has been possible to map erosion and land use for the three time periods pinpointing areas of major change. Results are presented in map form showing the spatial distribution of each erosion class and each land use category. It was however not possible to derive any meaningful relationship between erosion distribution and land use, on the aforementioned maps. The only observation made from the comparison of the maps is that erosion degree and distribution sometimes changes with land use, while land use sometimes changes in response to erosion. Major land use changes are the conversion of agricultural land to urban land use, and grazing land. Due to the mnlti-dimensional nature of soil erosion, hand 'actor analysis was employed to determine which factors or combinations of factors were dominant at each date. Despite the extensive research on the various factors affecting erosion such as those used for the USLE and SLEMSA there is a growing uncertainty as to which factors are more important to erosion. Soil erodibility has been found to be a component of the major controlling factor combinations in all three periods under study. At each date erodibility combined with a number of other factors determined the observed erosion distribution. As suggested by Mosley (1980), Cambell (1985) and Coleman and Scatena (1986) sediment from a catchment is derived principally from spatially limited portions of the catchment. Likewise eroded sediment becomes deposited in spatially limited areas with special characteristics which encourage deposition. Aerial photographic survey aided by ground survey and oblique photographs were employed to identify sediment sources and sinks within the catchment. Some sediment sources are fIxed such as gully floors and sides, while some change location from time to time such as construction sites. An evaluation of portions of the catchment for their ability to supply and deliver sediment has shown that the most eroded areas are not the most active sediment sources. Sediment yield is limited by either supply or transport. Sediment yield was estimated using reservoir survey data which indicate that there is a high rate of soil loss from the catchment. One flaw of this method as a measure of soil loss is that it treats the measured sediment yield as if it were contributed uniformly from the basin. This method however affords the researcher to estimate minimum erosion rates, taking into account that large amounts of sediment are stored at various places within the catchment. The possible socio-economic consequences of erosion and deposition have been identified. These include loss of cropland, destruction of roads and building sites which require methods of reclamation, sedimentation of small reservoirs and ponds, and the formation of gully bottom fills which are potential sediment sources. Conservation measures presently applied in the catchment are assessed and found to be irrelavant to the present erosion problem. Data from the reservoir survey revealed that the estimated rate of soil loss is more important to on-site erosion damage than to off-site damage in the form of the sedimentation of Maqalika reservoir. Appropriate conservation measures such as those suggested by Amimoto (1981) would be relevant to the study area, however the main constraint in their implementation would be lack of legislation and the absence of a sound land use policy. It is therefore concluded that the present land use situation which does not take into consideration the physical constraints of the catchment is partly responsible for accelarated erosion in the catchment.

Thesis (MScAgric (Soil Science))--University of Stellenbosch, 2010. / ENGLISH ABSTRACT: Lesotho is a country with an international reputation for the severe degree of soil erosion in its landscape. Despite several national soil conservation projects, soil erosion continues at an astounding rate. One of the reasons for this is possibly that the interactions between soil properties and erosion in Lesotho are not understood. Soil erosion is a site specific, cyclic phenomenon, controlled by geomorphological thresholds. To control soil erosion, the processes and soil properties which influence soil erosion in the specific place must be understood. In this study the soil properties of a highly eroded sub-catchment in Maphutseng, Lesotho was investigated. The gully extent in the sub-catchment, in 1957 and 2004 respectively, was mapped from aerial photos. These maps show where in the landscape gullies developed during this time. The gully maps were superimposed on maps of several soil erosion factors, to correlate the spatial distribution of the erosion factors with that of the gully distribution. A soil map was especially drawn for this. The spatial analysis shows that gully development between 1957 and 2004 was primarily confined to the area where duplex soils occur. The rest of the sub-catchment underwent negligible differences in gully extent during this time. The initiation of the gullies on the duplex soil area is ascribed to tunnel erosion. The high dispersibility of the duplex soil samples, sink holes which occur in this area and previous observations by researchers in this area gave evidence to this hypothesis. In the second part of the study the soil properties of seventeen soil profiles from across the study site were analysed. The difference in gully distribution between the duplex soils area and the rest of the catchment is ascribed to the high dispersibility of the duplex soils. No strong correlations could be found between the dispersion index and other determined soil properties. Segmented quantile regression was used to analyse the data further. Soil samples with moderate levels of total carbon (1.17%), iron oxide (0.9%) and effective cation exchange capacity (13.7 cmolc/kg), have below average dispersibility. When none of these stabilising agents are present in moderate amounts, soils with even low exchangeable sodium percentage values (0.68%) are dispersive. Furthermore, soils which have developed in colluvial material from basaltic origin were found to be less dispersive, presumably because of the amorphous clay minerals present in the volcanic material. The colour and increase in clay content between the A and B horizons of a soil can indicate the tunnel erosion potential of the soil. Dark coloured soils (values less than 4 and chromas less than 3) were found to have low dispersibility and free water can accumulate in the subsoils where the B horizon has much more clay than the A horizon. The accumulation of free water in the subsoil is necessary for tunnel formation. Thus soils with dark colours and/or a low clay accumulation index have low tunnel erosion potential. / AFRIKAANSE OPSOMMING: Lesotho is ‘n land met ‘n internasionale reputasie vir die ernstige graad van gronderosie waaronder die landskap gebuk gaan. Ten spyte van verskeie nasionale grondbewaringsprojekte duur die erosie teen ‘n verstommende tempo voort. Een van die redes hiervoor is heel moontlik dat die interaksies tussen grondeienskappe en erosie in Lesotho nie verstaan word nie. Gronderosie is ‘n plekspesifieke, sikliese verskynsel, wat deur geomorfologiese drempelwaardes beheer word. Om gronderosie te bekamp moet die prosesse en grondeienskappe wat gronderosie in die spesifieke plek beïnvloed, geïdentifiseer en verstaan word. In hierdie studie is die grondeienskappe van ‘n hoogs geërodeerde opvanggebied in Maphutseng, Lesotho ondersoek. Die dongaverspreiding in die opvanggebied, in 1957 en 2004 respektiewelik, is vanaf lugfoto’s gekarteer. Die kaarte wys waar in die landskap dongas gedurende hierdie tyd ontwikkel het. Die dongakaarte is op kaarte van verskeie gronderosie faktore gesuperponeer om die ruimtelike verspreiding van die erosie faktore met die donga verspreiding te korreleer. ‘n Grondkaart is spesiaal vir hierdie doel opgestel. Hierdie analise het gewys dat donga-ontwikkeling tussen 1957 en 2004 hoofsaaklik op die area waar dupleks gronde voorkom plaasgevind het. Die res van die opvanggebied het weinig verskille in donga verspreiding in hierdie tyd ondergaan. Die ontstaan van die dongas in die dupleksgronde word toegeskryf aan tonnelerosie. Die hoë dispergeerbaarheid van die dupleks grondmonsters, sinkgate wat in die area voorkom en vorige waarnemings deur navorsers in die area verleen bewyse aan hierdie hipotese. In die tweede deel van die studie is die grondeienskappe van sewentien grondprofiele van regoor die opvanggebied ontleed. Die verskil in donga verspreiding tussen die dupleksgrond area en die res van die opvanggebied is toeskryfbaar aan die hoë dispergeerbaarheid van die dupleks gronde. Geen sterk korrelasies is tussen die dispersiwiteits indeks en ander bepaalde grondeienskappe gevind nie. Gesegmenteerde kwantiel regressie is gebruik om die data verder te ontleed. Hierdie ontleding het gewys dat grondmonsters met matige vlakke van totale koolstof (1.17%), ysteroksied (0.9%) en effektiewe katioonuitruilkapasiteit (13.7 cmolc/kg), ondergemiddelde dispergeerbaarheid toon. Waar nie een van hierdie stabiliserings agente in matige hoeveelhede voorkom nie, is selfs gronde met baie lae uitruilbare natriumpersentasie waardes (0.68%) dispersief. Daar is ook gevind dat gronde wat vanuit kolluviale basaltiese afsettings ontwikkel het, minder dispersief is. Die kleur en verskil in klei persentasie tussen die A en B horison van ‘n grond kan as aanduiding dien van die grond se potensiaal vir tonnelerosie. Donker grondkleure (waarde laer as 4 en chroma laer as 3) wys op ‘n lae dispersiwiteit terwyl vrywater in die ondergrond van gronde waar die B horison veel meer klei as die E horison bevat kan akkumuleer. Die aansameling van vrywater in die ondergrond is noodsaaklik vir tonnelvorming. Dus het donker gronde en gronde met ‘n lae klei akkumulasie indeks ‘n lae potensiaal vir tonnelerosie.