1. As nom points out, the Methodology section just below the graphs states clearly "The colored temperature lines are the modeled estimates that each climate fa

1. As nom points out, the Methodology section just below the graphs states clearly "The colored temperature lines are the modeled estimates that each climate factor contributes to the overall temperature." So the black line and the various lines of the hypothesized contributing factors are on the same scale. 2. Yes, the graph shows the increase in aerosols having a cooling effect over time. The text states, "Some pollutants cool the atmosphere, like sulfate aerosols from coal-burning. These aerosols offset some of the warming. (Unfortunately, they also cause acid rain.)" Perhaps you thought the graph was a graph of aerosol levels, rather than a graph of modeled impact of aerosols on temperature? 3. WRT "Correlation is not causation": this phrase is quite abused. As you note, non-correlation can be evidence of non-causation, and correlation can be indicative of causation. Since we have straightforward causal theories for the relationship between atmospheric CO2 and increased temperature, and experimental evidence, and historical evidence, its a rather silly point to dispute in this simple representation of various factors that get offered as alternatives to atmospheric greenhouse gasses.