Socioeconomic Drivers of Land Use and Land Cover Change in Western Ethiopia
Table 2
Accuracy assessment for 1990, 2000, 2010, and 2020.
LULC classes
Bare land
Farmland
Forest
Settlement
Water body
Total user
User’s accuracy (%)
Accuracy assessment in 1990
Bare land
4
0
0
0
0
4
100
Farmland
0
8
0
0
0
8
100
Forest
0
2
4
0
0
6
67.7
Settlement
0
1
2
6
0
9
67.7
Water body
0
0
0
0
3
3
100
Total producer
4
11
6
6
3
30
Producer accuracy (%)
100
72.7
66.7
100
100
Accuracy assessment in 2000
Bare land
4
0
0
0
0
4
100
Farmland
0
7
0
1
0
8
87.5
Forest
0
0
7
0
0
7
100
Settlement
0
1
1
6
0
8
75
Water body
1
0
0
0
2
3
66.6
Total producer
5
8
8
7
2
30
Producer accuracy (%)
80
87.5
87.5
85.7
100
Accuracy assessment in 2010
Bare land
4
0
0
0
0
4
100
Farmland
0
6
1
0
0
7
85.7
Forest
1
1
6
0
1
8
75
Settlement
1
1
1
5
0
8
62.5
Water body
0
0
0
0
3
3
100
Total producer
5
8
8
5
4
30
Producer accuracy (%)
80
75
75
100
75
Accuracy assessment in 2020
Bare land
3
0
0
0
1
4
75
Farmland
0
7
0
0
0
7
100
Forest
0
1
6
0
0
7
85.7
Settlement
0
2
0
6
0
8
75
Water body
1
0
0
0
3
4
75
Total producer
10
10
6
6
4
30
Producer accuracy (%)
75
70
100
100
75
Overall accuracy
For 1990=83.3
For 2000=86.6
For 2010=80
For 2020=83.3
Overall kappa statistics
0.795 (79.5%)
0.826 (82.6%)
0.746 (74.6%)
0.788 (78.8%)
Sources: own summary of Landsat image results analysis via Arc GIS 10.3 (2023). The diagonal bold values show that the areas correctly classified for each year (1990–2020), the bold values of the last two rows for each year (1990–2020) represent total producer and producer accuracy (%), where as the last two columns for each year bold number represent total user and user’s accuracy (%), and finally, the last two rows bold values indicate that to focuses overall accuracy and overall kappa statistics.
