Reproducing Malcomb (2014)
I reproduced a study by Malcomb et al. (2014) that quantified climate vulnerability across Malawi by calculating a vulnerability score. This score was composed of factos in the categories of assets, access, livelihood sensitivity, and biophysical exposure (flood and drought risk).
Takeaways from the Reproduction Results
Adaptive Capacity Score
The reproduction tends to assign a lower Adaptive Capacity score to each TA than the original study (Figure 4). The reproduction was only higher than the original 4 times, and all were located in the south of Malawi. The reproduction assigned the same score 82 times, and underestimated the score 111 times. The Spearman’s Rho is 0.789, suggesting that the reproduction consistently made predictions when compared with the original scores. I wonder if this comes from the interpretation of the quintile scale as being from 0 to 4 instead of 1-5 (i.e. a larger bin count) which changed the normalization of the data.
Climate Change Vulnerability Score
The map comparing the original and reproduction vulnerability risk assessment (Figure 5) shows that the reproduction tended to assign lower vulnerability scores than the original study. However, the Spearman Rho coefficient of 0.197 suggests the values assigned have very little correlation to those of the original study. The vulnerability score is based on the adapative capacity score, so errors in the adaptive capacity could propagate into the uncertainty we see in the vulnerability reproduced score. However, the vulnerability score is an additive aggregation, so there must have been errors in the other factors in order to yield the random predictions.
My Contribution to the Reproduction
The Malcomb study only calculated vulnerability scores over a raster where each pixel was assigned a vulnerability score. The final climate change vulnerability map (Figure 5) doesn’t display the numerical score, instead displaying a color ramp from low vulnerability to high vulnerability. To quantify vulnerability for practical use (i.e. for stakeholders trying to allocate money and needing specific rankings and administrative areas), I calculated a vulnerability score for each TA, changing the analysis scale for this final output. To do this, I used the “aggregate” function to the mean vulnerability score of all pixels in a TA. I will visualized these scores on a map of the TA boundaries colored by vulnerability score and printed the TAs with the lowest and highest risk scores.
Related Links
The original study: Malcomb, D. W., E. A. Weaver, and A. R. Krakowka. 2014. Vulnerability modeling for sub-Saharan Africa: An operationalized approach in Malawi. Applied Geography 48:17–30. DOI:10.1016/j.apgeog.2014.01.004