Total: 1
The accuracy of a mass model in the strong lensing analysis is crucial for unbiased predictions of physical quantities such as magnifications and time delays. While the mass model is optimized by changing parameters of the mass model to match predicted positions of multiple images with observations, positional uncertainties of multiple images often need to be boosted to take account of the complex structure of dark matter in lens objects, making the interpretation of the chi-square value difficult. We introduce the Jackknife method as a new method to validate strong lens mass models, specifically focusing on cluster-scale mass modeling. In this approach, we remove multiple images of a source from the fitting and optimize the mass model using multiple images of the remaining sources. We then calculate the multiple images of the removed source and quantitatively evaluate how well they match the observed positions. We find that the Jackknife method performs effectively in simulations using a simple model. We also demonstrate our method with mass modeling of the galaxy cluster MACS J0647.7+7015. We discuss the potential of using the Jackknife method to validate the error estimation of the physical quantities by the Markov Chain Monte Carlo.