Assessing the Sensitivity of Climate Risk Variables in Opposed to Climate Hazards

Abstract

Climate variability such as the El Nino-Southern Oscillation (ENSO), can increase the likelihood of hydroclimatic hazards such as flooding and flood inundation due to sea level rise, potentially causing losses and damage to vulnerable coastal areas. The city of Pekalongan is vulnerable to the danger of permanent inundation of sea level rise and pluvial flooding caused by climate variability, high tides, erosion, and land subsidence. Climate risk assessments to determine adaptation actions are carried out by taking into account risk components such as hazards, exposures, sensitivity and adaptation capacity that make up the vulnerability. The use of Principal Component Analysis (PCA), Factor Analysis for Mixed Data (FAMD), and Polychoric PCA methods was carried out to analyze the contribution of the importance of each climate risk building variable in various dimensions of climate risk data consisting of continuous, discrete, and binary data. Sensitivity analysis by removing one by one the variables that build climate risk is also used to determine the most sensitive variables in the climate risk model. The results of the analysis show that the sub-districts in North Pekalongan District such as Degayu, Krapyak, Panjang Wetan, Panjang Baru, Kandang Panjang, Bandari and Padukuhan Kraton have a risk index with a range of 0.7 to 0.9 so that they are classified as high to very high. Other areas of Pekalongan City generally have a very low to low risk level, with an index of 0.2 to 0.4. The results of the sensitivity analysis showed that the variables that had the most influence on climate risk consisted of inundation of sea level rise, slopes, fishermen, distance from the coast, and distance from the river which could reduce or increase the value of the climate risk index by 5 to 10% if these variables were eliminated in the multi-level PCA simulation.