8 research outputs found
リモートセンシング技術を使用したエチオピアのタナ湖におけるホテイアオイ (Eichhornia crassipes) の時空間的変動解析とその影響
No full text創価大学博士（工学）This study examined the spatiotemporal dynamics of water hyacinth (WH) and its impact on hydrology and water quality in Lake Tana, Ethiopia. Three non-parametric machine learning algorithms were evaluated for WH detection. All classifiers achieved >95% accuracy with Sentinel-2 and >90% with Landsat-8. Although the performance differences between the methods were small, Random Forest demonstrated the highest accuracy and was used to estimate the spatiotemporal variability of WH distribution. High WH populations were concentrated in Lake Tana’s northeastern sector, with spatial coverage increasing significantly from 2015 to 2023. Water loss due to WH evapotranspiration also increased significantly during this period. Lake surface water temperature (LSWT) decreased significantly across all seasons except the dry season. Turbidity declined significantly in all seasons except the pre-rainy season. Chlorophyll-a (Chl-a) decreased in pre-rainy and rainy seasons but showed a non-significant increasing trend during dry and post-rainy seasons. WH biomass had a non-significant positive correlation with LSWT (r = 0.18), while a significant negative correlation with turbidity (r = -0.33) and Chl-a (r = -0.35). This study identified RF as the most accurate method for WH detection and comprehensively quantified its spatiotemporal distribution and impacts on the ecosystem using remote sensing technology for the first time.doctoral thesi
Land use/cover classification using machine learning algorithms and their impacts on land surface temperature and soil moisture in the Alawuha Watershed, Ethiopia
No full textLand use/cover (LULC) mapping is vital for natural resource management and environmental monitoring in rapidly developing regions such as Ethiopia's Northern Highlands. This study pioneers the integration of Sentinel-1 Synthetic Aperture Radar and Sentinel-2 Level 2A MultiSpectral Instrument data via Google Earth Engine to achieve high-accuracy LULC classification in the Alawuha Watershed, evaluating Classification and Regression Trees (CART), Random Forest (RF), and Support Vector Machines (SVM). It also examined spatiotemporal variations in land surface temperature (LST) and the soil moisture index (SMI) across LULC types using Landsat 8. The Radial Basis Function (RBF) SVM outperformed RF and CART, achieving an average overall accuracy (OA) of 89.6 % and an F1 score of 89.5 % across 2019 and 2024, compared to 88.2 % OA and 88.1 % F1 for RF, and 83.8 % OA and 83.3 % F1 for CART. Spatiotemporal analysis revealed urban expansion, increased forest cover, and stable farmland, with farmland consistently dominant in the watershed. LST decreased significantly from 2014 to 2025, with built-up areas showing the highest values at 41.4 °C (2019) and 38.1 °C (2024) and forests the lowest at 30.4 °C (2019) and 27.8 °C (2024). SMI increased significantly (2014–2025), with forests recording the highest values at 0.59 (2019) and 0.66 (2024), and built-up and bare lands the lowest. These findings highlight LULC's role in regulating microclimates and water balance, offering key insights for sustainable land-use planning and environmental management
The Spatiotemporal Dynamics of Land Use Land Cover Change, and Its Impact on Soil Erosion in Tagaw Watershed, Blue Nile Basin, Ethiopia
No full textEvaluating InVEST model for simulating annual and seasonal water yield in data-scarce regions of the Abbay (Upper Blue Nile) Basin: implications for water resource planners and managers
No full textIn developing countries, hydrological data is one of the limiting factors for evidence-based water resources planning and management. Thus, evaluating the performance of hydrological models that require relatively simple inputs is imperative for using them in the decision-making process. This study aims to evaluate the performance of the Integrated Valuation of Ecosystem and Tradeoff (InVEST) models for simulating annual and seasonal water yields in data-scarce regions of the Abbay (Upper Blue Nile) Basin with a case study in the Gumara watershed. The input data required by the InVEST Water Yield and the InVEST Seasonal Water Yield models were prepared from primary and ancillary data sources. The two InVEST models were calibrated using the calibrated Soil and Water Assessment Tool (SWAT) model outputs, and NSE of 0.86 and 0.91, PBAIS of 2.5% and 20.8%, and RMSE of 16 mm and 7 mm were obtained for the InVEST Water Yield and Seasonal Water Yield models, respectively. The attained performance measures indicated that both models can be used for evidence-based water resources planning and management in data-scarce regions of the Upper Blue Nile Basin. However, due attention is needed for calibrating the models since they are sensitive to the calibrating parameters
Detection of Water Hyacinth (Eichhornia crassipes) in Lake Tana, Ethiopia, Using Machine Learning Algorithms
No full textLake Tana is Ethiopia’s largest lake and is infested with invasive water hyacinth (E. crassipes), which endangers the lake’s biodiversity and habitat. Using appropriate remote sensing detection methods and determining the seasonal distribution of the weed is important for decision-making, water resource management, and environmental protection. As the demand for the reliable estimation of E. crassipes mapping from satellite data grows, comparing the performance of different machine learning algorithms could help in identifying the most effective method for E. crassipes detection in the lake. Therefore, this study aimed to examine the ability of random forest (RF), support vector machine (SVM), and classification and regression tree (CART) machine learning algorithms to detect E. crassipes and estimating seasonal spatial coverage of the weed on the Google Earth Engine (GEE) platform using Landsat 8 and Sentinel 2 images. Cloud-masked monthly median composite Landsat 8 and Sentinel 2 data from October 2021 and 2022, January 2022 and 2023, March 2022, and June 2022 were used to represent autumn, winter, spring, and summer, respectively. Four spectral indices were derived and used in combination with spectral bands to improve the E. crassipes detection accuracy. All methods achieved greater than 95% and 90% overall accuracy when using Sentinel 2 and Landsat 8 images, respectively. Using both data sets, all methods achieved a greater than 93% F1 score for E. crassipes detection. Though the difference in performance between the methods was small, the RF was the most accurate, while the SVM and CART methods had the same accuracy. The maximum E. crassipes coverage area was observed in autumn (22.4 km2), while the minimum (2.2 km2) was observed in summer. Based on Sentinel 2 data, the E. crassipes area coverage decreased significantly by 62.5% from winter to spring and increased significantly by 81.7% from summer to autumn. The findings suggested that the RF classifier was the most accurate E. crassipes detection algorithm, and autumn was an appropriate season for E. crassipes detection in Lake Tana
Biomechanical Properties and Agro-Morphological Traits for Improved Lodging Resistance in Ethiopian Teff (Eragrostis tef (Zucc.) Trottor) Accessions
No full textBiomechanical Properties and Agro-Morphological Traits for Improved Lodging Resistance in Ethiopian Teff (Eragrostis tef (Zucc.) Trottor) Accessions
No full textSusceptibility to lodging is a major constraint on teff production in Ethiopia, but efforts to develop lodging-resistant cultivars have not been successful. We studied the mechanical properties of teff culms and associated agro-morphological traits in field experiments with 320 teff accessions at two sites in northwestern Ethiopia during the 2018 and 2019 growing seasons. The results showed significant variability in both mechanical properties and agro-morphological traits among accessions. Traits contributing to lodging resistance, such as internode diameter, pushing resistance, and base failure moment, were significantly positively correlated with each other and with plant height. Similarly, the correlation of those traits with lodging index was significant and positive. In contrast, tiller number showed a significant negative correlation with lodging index. The peduncle–panicle length, which generally accounted for 59% of the plant height, should be a target when breeding for semi-dwarfism. Root system development, which reached a depth of more than 1 m in tall and 57 cm in dwarf teff accessions, signifies the presence of genetic variabilities for future root lodging studies in teff, and it may also explain why teff performs well in drought-prone areas of Ethiopia. Breeding programs for lodging resistance might focus on accessions with good standing ability (high base failure moment) and introgression of stem strength with a semi-dwarf phenotype. Alternatively, selection for a large internode diameter, increased pushing resistance and base failure moment, and a reduced tiller number should be considered
Yield Potential and Variability of Teff (<i>Eragrostis tef</i> (Zucc.) Trotter) Germplasms under Intensive and Conventional Management Conditions
No full textTeff is the most strategic cereal crop grown from high rainfall to drought prone areas of Ethiopia, where it covers nearly 30% of the land allotted for cereals. However, its productivity remains very low due to lack of knowledge and research interventions. To investigate the grain yield potential, estimate the genetic parameters, and the diversity, a pot experiment with intensive management and a field experiment with conventional management at two contrasting locations for two seasons using the same 317 genotypes and additional 3 improved cultivars in the field experiment were carried out. The results showed highly significant variation among the genotypes for grain yield, biomass, harvest index, and phenological traits under both experiments. The best linear unbiased predictor (BLUP)-adjusted grain yield performance of the genotypes ranged from 4.2 to 8.8 g/plant in the intensive management and 1.8 to 4.3 g/plant in the field growing condition with conventional management. Coefficient of genetic variation, heritability, and expected genetic advance for grain yield were the highest in both experiments. Among the phenological traits, the grain filling period in the intensive growing condition exceptionally showed the highest genetic coefficient of variation and genetic advance. The high grain yield performance and wider range of the harvest index observed under the intensive management condition with moderate to high heritability signifies the genetic potential of teff for further improvement through trait recombination
