Browsing by Author "Odera, Patroba"

Now showing 1 - 7 of 7
  • Thumbnail Image
    Item
    Open Access
    A remote sensing assessment of irrigation land use land cover change in the Sokoto Rima River Basin, Nigeria
    (2021) Abdulmajid, Aminu; Eckardt, Frank; Odera, Patroba
    This study examines the expansion and utilisation of Fadama irrigation in the Sokoto Rima River Basin using the ground yield and remote sensing data. Decadal land use land cover (LU/LC) change detection was conducted using remotely sensed data from Landsat 4,5,7 ETM for 1988, 1998 and Landsat 8 OLI for 2018 using a digital classification and a cloud-based classification provided by Google Earth Engine (GEE) API, with an overall accuracy of 97% in 1988, 92% in 1998 and 90% in 2018. Additionally, the Normalised Difference Vegetation Index (NDVI) derived from MODIS, GEE NDVI, and the Landsat was used to assess the crop yield patterns of the irrigation farming at the Fadama sites in conjunction with the limited ground yield data from 2000-2005, and was found to be a positive change over the years. The result generated from the classification was juxtaposed with observable field characteristics of the LU/LC identified. The decadal trend between 1988 to 1998 saw an increase of 114% in dam surface water and 166% of the natural vegetation, while the nonvegetated areas and the Fadama areas decreased in size by -100% and -65% respectively. Between 1998 and 2018, the reverse of the previous pattern was observed, with water and vegetated areas decreasing in their surface area in hectares by -80% and -23% respectively. Non-vegetated areas increased by 3% while Fadama areas increased in size by 112%, indicating that the Fadama areas were under-utilised by - 65% of the total hectarage in the past, and in recent years a 47% increase was recorded between the two periods. In addition, the Standardised Precipitation and Evapotranspiration Index (SPEI) data from 1950 was used to analyze the meteorological and hydrological drought pattern at four locations within the basin and was found to be a moderate drought. The SPEI data was then correlated with the Global Surface Water Explorer data to observe the surface water dynamics and to show the drought extent around Bakolori, Goronyo, Jibia, Wurno and Zobe dams. The result showed that the dams are shrinking in size due to risen temperature and consequent evaporation caused by moderate drought and water use for irrigation. Lastly, this research uncovers the utilization of the Fadama in recent years by 47%, but with mismanagement of the resources in the Sokoto Rima River floodplain because not all the irrigation fields are put into cultivation, with 39,907 ha put into use from the planned irrigation of 105,472 ha in the entire basin. Therefore, this research recommends a proper intervention of the government to promote and enhance sustainable management of the Fadama lands, water, and the vegetation resource. This should be done in consideration of the deteriorating climate to close yield gaps, with much emphasis on engaging the local farmers by monitoring yearly ground yield data and ensuring all incentives are distributed effectively to brace an active and sustainable management of the entire basin and its vast resource.
  • Thumbnail Image
    Item
    Open Access
    Assessment of Hand-held GPS Surveying in Land Adjudication: A Case Study of Ngoliba Settlement Scheme in Kenya
    (JKUAT, 2017-08-02) Odera, Patroba; Jatani, Abdulkadir
    Land adjudication is the process through which existing rights in a particular parcel of land are ascertained. Land ownership in Kenya is largely achieved through a land adjudication process, especially in the rural areas. The Preliminary Index Diagrams (PIDs) are the official map documents that together with adjudication records constitute the adjudication register required for determination and registration of interests and rights over land. The PIDs are produced from enlarged, marked and un-rectified aerial photographs; hence they contain distortions. This paper assesses the application of hand-held GPS surveying to determine plot boundaries in an adjudication area in Kenya. The study was conducted in Ngoliba Settlement schemes. Areas of 12 plots of similar sizes were obtained by three methods - PIDs, traditional ground based surveying using a station-theodolite and EDM (total station) and using a hand-held single GPS (in point positioning mode). Plot areas obtained from PIDs and GPS were compared against those determined from total station measurements, which are taken as the standard in this study. We also compared the coordinates and side lengths derived from total station and handheld GPS measurements. Results show that hand-held GPS can recover 98.1% of the combined area of the 12 plots compared to PIDs which can only recover 77.5%. Results from coordinates and side lengths differences show that hand-held GPS can determine side lengths and coordinates that are approximately close to the ones obtained by total station. Hence Hand-held GPS would significantly improve the accuracy of parcel areas for land adjudication when compared to those obtained from PIDs only.
  • No Thumbnail Available
    Item
    Open Access
    Determination and analysis of land deformation in major landslide areas in Kenya using geospatial techniques: A case study of Murang a and Nyeri counties
    (2024-03) Yumbu, Alex; Odera, Patroba
    This thesis presents a meticulous investigation into determining and analysing land deformation in the prominent landslide-prone areas of Kenya, specifically within the central highlands encompassing Murang'a and Nyeri counties. The research employs a spectrum of areal deformation methods, including Small Baseline Sub-set (SBAS) Interferometric Synthetic Aperture RADAR (InSAR) from processed ALOS PALSAR and Sentinel-1 datasets, Gravity Recovery and Climate Experi-ment (GRACE/GRACE-FO) satellite data, and hydrological modelling data from the Global Land Data Assimilation System (GLDAS), to elucidate and dissect the time series deformation trends prevalent in the study area. A multi-faceted methodology was employed to investigate the deformation induced by landslides in the study area. Various datasets were harnessed to provide a compre-hensive understanding of the phenomena. Satellite gravimetry data from GRACE and GRACE-FO, in conjunction with hydrological modelled data from GLDAS, delivered valuable insights into vertical deformation rates at selected points of inter-est. These points were identified through InSAR processing using RADAR imagery from Sentinel-1 and ALOS PALSAR, which offered finer spatial resolution. While Sentinel-1 imagery was acquired and processed in both the ascending and descend-ing nodes, ALOS PALSAR imagery was solely processed in the ascending node due to dataset availability constraints. In the Sentinel-1 SBAS analysis, temporal and perpendicular baselines were constrained to 60 days and approximately 200 metres, respectively. For ALOS PALSAR imagery, the temporal and perpendicular baselines were constrained to 875 days and 2500 metres, respectively. Furthermore, to discern the most probable causes of deformation, a range of factors were considered. These encompassed precipitation data, which was critical in as- sessing the role of rainfall in triggering landslides. Additionally, geomorphological factors, such as slope, aspect, and surface roughness, extracted from SRTM Arcsec- ond DEM were incorporated. Land Use Land Cover data spanning from 1973 to 2022 were analysed to understand the impact of human activities and land cover changes on the occurrence of landslides. The amalgamation of these datasets and analyses facilitated a comprehensive examination of the factors contributing to deformation and landslides in the study area. Results reveal distinctive deformation patterns over the study area from various datasets. Sentinel-1 Ascending imagery indicates varying uplift rates from 19.8 to 189.3 mm/yr and varying subsidence rates from -169.4 to -63.8 mm/yr over the area Iv of study. Sentinel-1 Descending data corroborates these trends, with varying uplift rates from 59.5 to 209.9 mm/yr, and varying subsidence rates from -200.8 to -90.7 mm/yr. Results from the ALOS PALSAR dataset contributes additional complexity but with lower values than Sentinel -1 data, probably due to data gaps. The rates of uplift observed from ALOS PALSAR dataset vary from 16.5 to 100.8 mm/yr, while the rate of subsidence vary from -715.9 to -60.8 mm/yr. GRACE/GRACE-FO and GLDAS datasets provide smaller values of uplift and subsidence at a larger scale, devoid of large deformations at smaller scales captured by InSAR. Among investigated possible causes of landslides, geomorphological factors emerge as prominent influencers of deformation patterns, while changes in land use and land cover, high and extreme rainfall events and lithology and soil properties also play pivotal roles. Crucially, this thesis underscores that landslides and deformation are rarely attributed to single causative agents; rather, they result from an intricate interplay of various factors, often occurring simultaneously.
  • No Thumbnail Available
    Item
    Open Access
    Determination and analysis of land deformation in major landslide areas in Kenya using geospatial techniques: A case study of Murang a and Nyeri counties
    (2024) Yumbu, Alex; Odera, Patroba
    This thesis presents a meticulous investigation into determining and analysing land deformation in the prominent landslide-prone areas of Kenya, specifically within the central highlands encompassing Murang'a and Nyeri counties. The research employs a spectrum of areal deformation methods, including Small Baseline Sub-set (SBAS) Interferometric Synthetic Aperture RADAR (InSAR) from processed ALOS PALSAR and Sentinel-1 datasets, Gravity Recovery and Climate Experi-ment (GRACE/GRACE-FO) satellite data, and hydrological modelling data from the Global Land Data Assimilation System (GLDAS), to elucidate and dissect the time series deformation trends prevalent in the study area. A multi-faceted methodology was employed to investigate the deformation induced by landslides in the study area. Various datasets were harnessed to provide a compre-hensive understanding of the phenomena. Satellite gravimetry data from GRACE and GRACE-FO, in conjunction with hydrological modelled data from GLDAS, delivered valuable insights into vertical deformation rates at selected points of inter-est. These points were identified through InSAR processing using RADAR imagery from Sentinel-1 and ALOS PALSAR, which offered finer spatial resolution. While Sentinel-1 imagery was acquired and processed in both the ascending and descend-ing nodes, ALOS PALSAR imagery was solely processed in the ascending node due to dataset availability constraints. In the Sentinel-1 SBAS analysis, temporal and perpendicular baselines were constrained to 60 days and approximately 200 metres, respectively. For ALOS PALSAR imagery, the temporal and perpendicular baselines were constrained to 875 days and 2500 metres, respectively. Furthermore, to discern the most probable causes of deformation, a range of factors were considered. These encompassed precipitation data, which was critical in as- sessing the role of rainfall in triggering landslides. Additionally, geomorphological factors, such as slope, aspect, and surface roughness, extracted from SRTM Arcsec- ond DEM were incorporated. Land Use Land Cover data spanning from 1973 to 2022 were analysed to understand the impact of human activities and land cover changes on the occurrence of landslides. The amalgamation of these datasets and analyses facilitated a comprehensive examination of the factors contributing to deformation and landslides in the study area. Results reveal distinctive deformation patterns over the study area from various datasets. Sentinel-1 Ascending imagery indicates varying uplift rates from 19.8 to 189.3 mm/yr and varying subsidence rates from -169.4 to -63.8 mm/yr over the area Iv of study. Sentinel-1 Descending data corroborates these trends, with varying uplift rates from 59.5 to 209.9 mm/yr, and varying subsidence rates from -200.8 to -90.7 mm/yr. Results from the ALOS PALSAR dataset contributes additional complexity but with lower values than Sentinel -1 data, probably due to data gaps. The rates of uplift observed from ALOS PALSAR dataset vary from 16.5 to 100.8 mm/yr, while the rate of subsidence vary from -715.9 to -60.8 mm/yr. GRACE/GRACE-FO and GLDAS datasets provide smaller values of uplift and subsidence at a larger scale, devoid of large deformations at smaller scales captured by InSAR. Among investigated possible causes of landslides, geomorphological factors emerge as prominent influencers of deformation patterns, while changes in land use and land cover, high and extreme rainfall events and lithology and soil properties also play pivotal roles. Crucially, this thesis underscores that landslides and deformation are rarely attributed to single causative agents; rather, they result from an intricate interplay of various factors, often occurring simultaneously.
  • Thumbnail Image
    Item
    Open Access
    Determination of Suitable Sites for Establishment of Large-Scale Concentrated Solar Power Plants in Kenya
    (Commercial, 2017-01-06) Gathu, Joan; Odera, Patroba; Waithaka, Edward
    The demand for energy in Kenya, especially for electricity, is increasing rapidly due to population growth, decentralization of governance, and technological and industrial development. Hydroelectricity, the core source of power, has proved unreliable due to the rapid climate change. In response, the country has ventured into other renewable sources to counter the issues posed by the alternative nonrenewable sources such as unreliability, high costs, and environmental degradation as seen with the use of diesel and kerosene. The purpose of this research is to determine the viability of setting up a large-scale concentrated solar power plantation in Kenya that will assist in stabilizing Kenya’s energy demand and supply as well as increase its affordability. The project is divided into three phases. The first phase conducts an overlay analysis to determine the Kenya’s solar energy potential. The results show that the northern region has the highest potential. The second step involves the creation of an exclusion mask which eliminates the unsuitable land forms and Land Use Land Cover. Based on the results, the best ten sites are situated in Turkana and Marsabit counties. The final phase involves the evaluation of the potential capacity of power that could be generated per square kilometer. The study finds out that the potential varies based on the technologies: parabolic trough, linear Fresnel reflector, or dish systems.
  • Thumbnail Image
    Item
    Open Access
    Evaluation of GOCE-based global gravity field models over Japan after the full mission using free-air gravity anomalies and geoid undulations
    (Springer, 2017-09-26) Odera, Patroba; Yoichi, Fukuda
    The performance of Gravity field and steady-state Ocean Circulation Explorer (GOCE) global gravity field models (GGMs), at the end of GOCE mission covering 42 months, is evaluated using geoid undulations and free-air gravity anomalies over Japan, including six sub-regions (Hokkaido, north Honshu, central Honshu, west Honshu, Shikoku and Kyushu). Seventeen GOCE-based GGMs are evaluated and compared with EGM2008. The evaluations are carried out at 150, 180, 210, 240 and 270 spherical harmonics degrees. Results show that EGM2008 performs better than GOCE and related GGMs in Japan and three sub-regions (Hokkaido, central Honshu and Kyushu). However, GOCE and related GGMs perform better than EGM2008 in north Honshu, west Honshu and Shikoku up to degree 240. This means that GOCE data can improve geoid model over half of Japan. The improvement is only evident between degrees 150 and 240 beyond which EGM2008 performs better than GOCE GGMs in all the six regions. In general, the latest GOCE GGMs (releases 4 and 5) perform better than the earlier GOCE GGMs (releases 1, 2 and 3), indicating the contribution of data collected by GOCE in the last months before the mission ended on 11 November 2013. The results indicate that a more accurate geoid model over Japan is achievable, based on a combination of GOCE, EGM2008 and terrestrial gravity data sets.
  • No Thumbnail Available
    Item
    Open Access
    Investigating flood susceptibility and vulnerability through geospatial techniques in Kerala State, India
    (2023) George, Mohan Navya; Odera, Patroba
    A flood is an overflow of water that submerges land that is usually dry. Flooding is a complex phenomenon caused by various meteorological, hydrological, geomorphological, and anthropogenic factors. Flood susceptibility identifies the most vulnerable areas that are flood prone based on the region's physical features. Flood vulnerability is the degree to which people or man-made structures are affected by, or unable to cope with the effects of flooding. One of the key factors in risk management and flood damage assessment is determining flood vulnerability. There has been an increased need to understand floods as they are recurring more frequently. The objectives of this study were to determine the land use/cover changes in Kerala State over the last 45 years, model flood susceptibility for the years 1973, 2001 and 2018 and finally determine flood vulnerability in Kerala for the year 2018. These objectives were achieved through Remote Sensing and GIS techniques. Image classification of Landsat images helped identify the change in land use/cover. The flood susceptibility model was attained by processing the physical characteristics and using the multi-criteria weighted analysis. The flood vulnerability model was achieved by processing the copying, social, infrastructure and physical vulnerability. Again, multi-criteria weighted analysis was used. Results show that in the year 2018, 50.31 % of the area had a medium flood susceptibility and 28.74% of the area had a high susceptibility. The areas of high flood susceptibility and medium flood susceptibility are visually seen to be around the high concentration of built-up area. The high susceptibility regions have only increased since 1973 as the built-up area increased. Regarding flood vulnerability, most of the State (60.59%) had a medium flood vulnerability in 2018, followed by high vulnerability areas taking up 28.48% of the land. The high vulnerability areas are found along the west coast of the State. This study brings awareness that the flood susceptibility regions are increasing year by year and that the State has areas that are highly prone to flood vulnerability. The driving factors behind the increase in flood susceptibility and high flood prone areas are the increase in rainfall, population and the built environment.