Page 162 - AJWEP-22-4

Deresse, et al.

suitability. ArcGIS 10.8 was utilized for mapping, to Saleem et al. Likewise, elevation and slope maps
35
28
while ERDAS Imagine 2015 was used to process the were also adjusted according to FAO guidelines. 36
Landsat 8 images. The MCE model integrated various Landsat images (Operational Land Imager [OLI],
factors, including soil, topography, climate, proximity, Enhanced Thematic Map [ETM+], and Thematic Mapper
and land use land cover (LULC) and assigned weights [TM]) from 2023 were obtained from the United States
to evaluate their impact on coffee production. Geological Survey (USGS; https://earthexplorer.usgs.
gov/). ERDAS Imagine 2015 was used to create a LULC
2.3. Data analysis map through supervised classification. The maximum
2.3.1. Survey data analysis likelihood algorithm categorized pixels into cultivated
Survey data were analyzed using SPSS V.26, applying lands, forests, grasslands, settlements, bare lands,
descriptive statistics (frequencies and percentages) to woodlands, and water bodies (Table A1). Accuracy
37
outline socioeconomic characteristics. Cross-tabulations assessments of the LULC classes were conducted
and Chi-square (χ²) tests were employed to examine following standard procedures.
the impact of various factors, such as soil, topography, Proximity is a key factor in assessing land suitability
climate, and proximity, on coffee production in the for coffee production. A close proximity to water sources
Abaya and Gelana Districts. ensures consistent moisture for coffee plants, reducing
irrigation costs and enhancing yields. Areas within 2 km
12
2.3.2. Spatial data processing and analysis of towns were excluded to prevent shadowing, as proximity
In this study, we used MCE and GIS tools to assess land to markets and transportation infrastructure reduces
suitability for coffee production in accordance with distribution costs and increases profitability. Good road
38
Akpoti et al. and Food and Agriculture Organization access improves transportation efficiency for both coffee
29
(FAO) guidelines, evaluating factors such as and agricultural inputs and facilitates extension services,
30
topography, soil, land use, climate, and accessibility. supporting better farming practices and productivity. 39
MCE assessed both quantitative and qualitative data,
providing a comprehensive overview of temperature, 2.3.3. Accuracy assessment
precipitation, soil condition (texture, depth, and pH), The classifications were compared with reference data
slope, land cover types, and infrastructural accessibility to measure differences. Random pixel selection and
(proximity). GIS enabled spatial analysis, with ArcGIS standardized methods were used to reduce bias. 40
31
10.8 used for mapping and ERDAS Imagine 2015 for
processing Landsat 8 images. The overall accuracy measures the percentage of
Soil data (pH, depth, and texture) of the study correctly classified pixels across all classes, providing a
41
area were sourced from the International Soil metric for the map’s overall correctness:
Reference and Information Centre (ISRIC) website Total numberof
(https://www.isric.org/explore/isric-soil-data- correctly classifiedpixels
hub). Raster-formatted data were acquired, and the Overallaccuracy Totall numberof 100%
corresponding soil data for each study area were
extracted and resampled in ArcGIS 10.8 to a resolution referencepixel
of 30 m. Soil maps were generated based on coffee (II)
crop requirements using land suitability categories The Kappa coefficient evaluates classification
from the Belay and Assen. 32 accuracy by comparing reference data (ground truth
Climate grid data were collected from the Hawassa data) and the observed classification results from
Station and processed in ArcGIS 10.8 to estimate classified map (e.g., coffee suitability classes derived
rainfall and temperature distributions. The temperature from Landsat TM/OLI). A value > 0.8 indicates strong
map was classified using Nagashree et al., and the agreement, 0.4 – 0.8 indicates moderate agreement, and
33
same classification was applied to the rainfall map for <0.4 reflects poor agreement. 42
assessing land suitability for coffee production. A digital Total sum corrected thesum
elevation model (DEM) was used to analyze elevation of allthe rowtotal ccolumn total

and slope for Arabica coffee production. Elevation Kappa coeffcient
34
ranged from 1104 to 2305 m in the Abaya District Totalsquared sumofall the
and 1075 to 2511 m in the Gelana District. Slope was ( column total)
calculated from the DEM data and reclassified according (III)

Volume 22 Issue 4 (2025) 154 doi: 10.36922/AJWEP025190143