Figures & data
Table 1. Land change drivers & constraints.
Table 2. LULC trajectory used in creating probability/transition potential Surfaces.
Figure 2. Projected land use data model construction.
Note: Dse socioeconomic drivers, Dpr proximity drivers, Dpb probability driver, 1–3 perceptron, w weights, Ci first constraint (e.g. forest), Cn last constraint (e.g. urban growth boundary).
![Figure 2. Projected land use data model construction.Note: Dse socioeconomic drivers, Dpr proximity drivers, Dpb probability driver, 1–3 perceptron, w weights, Ci first constraint (e.g. forest), Cn last constraint (e.g. urban growth boundary).](/cms/asset/a5c66c32-0d0a-4214-b88e-93bfbd157f0d/tgrs_a_1533158_f0002_b.gif)
Table 3. Producers and users accuracy for pixel-based hybrid classified images.
Table 4. Producers and users accuracy for object-based hybrid classified images.
Table 5. Land use/land cover net gain transition matrix (hectares).
Figure 4. Historical and contemporary object-based hybrid classified images for 1990 (a), 2000 (b), and 2011 (c).
![Figure 4. Historical and contemporary object-based hybrid classified images for 1990 (a), 2000 (b), and 2011 (c).](/cms/asset/454c09a1-d0cd-4c12-bdb4-4acc7bf65e14/tgrs_a_1533158_f0004_oc.jpg)
Figure 6. Snippet of 2011 pixel-based hybrid projected image (a) compared to object-based hybrid (b).
![Figure 6. Snippet of 2011 pixel-based hybrid projected image (a) compared to object-based hybrid (b).](/cms/asset/fed13ce1-ba60-4ef8-b2ed-a4245d227f1e/tgrs_a_1533158_f0006_oc.jpg)
Figure 7. Average ROC performance of model for all projected 2011 LULC classes.
Note: AUCO is object-based hybrid, AUCP is pixel-based hybrid.
![Figure 7. Average ROC performance of model for all projected 2011 LULC classes.Note: AUCO is object-based hybrid, AUCP is pixel-based hybrid.](/cms/asset/3afb98f4-35d3-49ed-81df-6d794ea159fa/tgrs_a_1533158_f0007_oc.jpg)