Soil Salinity Mapping and Risk Assessment Using Remote Sensing and Gis: the Case of Wonji Sugar Cane Irrigation farm
| dc.contributor.advisor | Argaw Mekuria (PhD) | |
| dc.contributor.author | Asfaw Engdawork | |
| dc.date.accessioned | 2020-11-10T09:17:22Z | |
| dc.date.accessioned | 2023-11-09T14:11:38Z | |
| dc.date.available | 2020-11-10T09:17:22Z | |
| dc.date.available | 2023-11-09T14:11:38Z | |
| dc.date.issued | 2014-06-06 | |
| dc.description.abstract | Soil salinity is one of the most damaging environmental problems worldwide, especially in arid and semi-arid regions. The aim of this study was to assess the risk of soil salinity in Wonji sugarcane irrigation farm and to develop effective combined remote sensing based statistical regression models to predict and map spatial variation of soil salinity. Different spectral indices were calculated from original bands of landsat images. Statistical correlation between field measurements of Electrical Conductivity (EC), spectral indices and landsat original bands showed that the Salinity Index (SI) had the highest correlation with EC. Combining these remotely sensed variables into one model yielded the best fit with R2 = 0.78. The result obtained from SI was not only in area wise, but also the level of salinity. Out of the total area, 18.8 and 23 % was identified as moderately and slightly saline respectively. The prediction model that was obtained from the regression analysis was used to derive a salinity map and estimate EC level for the 1985, 1995 and 2012 landsat images. The result of SI shows moderately and slightly saline soils were increased by 10 and 14 ha/yr respectively. The Spatial Overlay model was also developed from ground water table, elevation, geology, soil texture and vegetation density. Three classes have been identified with varying degree of salinity and the result showed about 36% of the study area is non-saline where as 30% and 33.7% is moderately and slightly saline respectively. In spatial overlay salinity model, the class of moderately saline soils was found in the areas underlain by the lacustrine Sediments and shallow ground water level. It is evident that the areas highly vulnerable to salinization greatly related to the ground water level that normally occurred on the lacustrine sediment. The spatial distribution of salt affected area derived from SI and spatial overlay model were similar pattern but of in different extent. The validation of the two models has been carried out by the existing EC values referenced to the same locations by making linear regression to test their predication capability and hence remote sensing soil salinity prediction model has revealed better correlation coefficient of 63 % to the measured ECe. The spatial overlay analysis between salt affected areas and canal and water table was made to assess the spatial distribution as well as the relationship with these features. It was revealed that the spatial distribution was not highly influenced by the features considered except ground water table. The results demonstrate that modeling and mapping spatial variation of soil salinity based on remote sensing data is a promising approach. | en_US |
| dc.identifier.uri | http://10.90.10.223:4000/handle/123456789/23144 | |
| dc.language.iso | en | en_US |
| dc.publisher | Addis Ababa University | en_US |
| dc.subject | EC | en_US |
| dc.subject | Prediction Model | en_US |
| dc.subject | Spatial Overlay | en_US |
| dc.subject | Salinity Model | en_US |
| dc.subject | Si | en_US |
| dc.title | Soil Salinity Mapping and Risk Assessment Using Remote Sensing and Gis: the Case of Wonji Sugar Cane Irrigation farm | en_US |
| dc.type | Thesis | en_US |