Fig. 3: Causal graphs at different time scales.
From: Spatiotemporal heterogeneity reveals urban-rural differences in post-disaster recovery
a A causal graph union aggregated from b to f. b A causal graph for July 1st, 2017 - December 31st, 2017 (tp1). c A causal graph for July 1st, 2017 - September 10th, 2017 (tp2). d A causal graph for September 11th, 2017 - September 30th, 2017 (tp3). e A causal graph for October 1st, 2017 - December 31st, 2017 (tp4). f A causal graph for September 11th, 2017 - December 31st, 2017 (tp5). Panels (b) to (f) in this figure have two parts. The upper part shows the county-level aggregation results, and the lower part describes the POI-level aggregation results. g The proportion of types of causal links of causal graphs from panels (a) to (f). Intra-regional links represent the links within a certain county. On the other hand, inter-regional links represent the links between two counties.
