Mapping existing landslides is a fundamental prerequisite to build any reliable susceptibility model. From a series of landslide presence/absence conditions and associated landscape characteristics, a binary classifier learns how to distinguish potentially stable and unstable slopes. In data rich areas where landslide inventories are available, addressing the collection of these can already be a challenging task. However, in data scarce contexts, where geoscientists do not get access to pre-existing inventories, the only solution is to map landslides from scratch. This operation can be extremely time-consuming if manually performed or prone to type I errors if done automatically. This is even more exacerbated if done over large geographic regions. In this manuscript we examine the issue of mapping requirements for west Tajikistan where no complete landslide inventory is available. The key question is: How many landslides should be required to develop reliable landslide susceptibility models based on statistical modeling? In fact, for such a wide and extremely complex territory, the collection of an inventory that is sufficiently detailed requires a large investment in time and human resources. However, at which point of the mapping procedure, would the resulting susceptibility model produce significantly better results as compared to a model built with less information? We addressed this question by implementing a binomial Generalized Additive Model trained and validated with different landslide proportions and measured the induced variability in the resulting susceptibility model. The results of this study are very site-specific but we proposed a very functional protocol to investigate a problem which is underestimated in the literature.

In this study, a new tool for quantitative, data-driven susceptibility zoning (SZ) is presented. The SZ plugin has been implemented as a QGIS plugin to maximize its operational use within the geoscientific community. QGIS is in fact a commonly used open-source geographic information system. We have scripted the plugin in Python, and developed it as a collection of functions that allow one to pre-process the input data, calculate the susceptibility, and then estimate the quality of the classification results. The susceptibility zoning can be carried out via a number of classifiers including weight of evidence, frequency ratio, logistic regression, random forest, support vector machine, and decision tree. The plugin allows one to use any kind of mapping units, to fit the model, to test it via a k-fold cross-validation, and to visualize the relative receiving operating characteristic (ROC) curves. Moreover, a new classification method of the susceptibility index (SI) has been implemented in the SZ plugin. A typical workflow of the SZ plugin is described, and its application for landslide susceptibility zoning in Northeast India is reported. The data of the predisposing factors used are open, and the analysis has been carried out using a logistic regression and weight of evidence models. The corresponding area under the curve of the relative ROC curves reflects an optimal model prediction capacity. The user-friendly graphical interface of the plugin has allowed us to perform the analysis efficiently in few steps.