**8. Conclusions**

We propose a modified antecedent rainfall–susceptibility (*AR*-*S*) threshold approach that improves on the initial *AR-S* method of [35], being transferable to other data sets for landsliding and *S*. For its development and evaluation we exploit the most current and extensive landslide inventory for the western branch of the East African Rift comprising 184 dated landslide events from 2001 to 2019, satellite-based rainfall estimates from TMPA 3B42 RT, and two *S* models, i.e., the continental-scale model of [45] and the regional-scale *S* model of [49]. The main novelty in the modified *AR-S* approach is the stratified selection of data associated with the lowest *AR* values able to cause landsliding, allowing to deploy data sets that are not necessarily homogenously distributed over the *S* range. Furthermore, we highlight that the threshold procedure is more meaningful when no bootstrapping statistical technique is applied, as the uncertainties in the parameters that define the power-law threshold model are mainly introduced by the bootstrapping related random sampling in combination with the presence of outliers in the data set. We obtain improved *AR* thresholds with an increased susceptibility-dependent gradient, and *AR* threshold maps with a higher accuracy through the use of the regional-scale *S* model in the modified *AR-S* approach. The improved *AR* threshold values at the 5% exceedance probability range from 7 mm in the most susceptible areas (*S* = 0.72) to 62 mm in the lowest susceptible areas (*S* = 0.10) where landslides have been recorded (uncorrected for underestimation by TMPA). Our approach is foremost relevant in data-scarce regions, where the lack of abundant data from rain gauges and in particular on landslide occurrence hampers the use of homogenously distributed data sets. Moreover, we suggest that this modified method is transferable not only to other data sets for *S*, but to any parameter that might be considered as a possible cause for landsliding.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2073-4441/11/11/2202/s1, Figure S1: Distribution of the data (white bars) in the calibration data set over 10 logarithmic equidistant *S* classes for the (**a**) continental-scale [45] and (**b**) regional-scale [49] susceptibility models. "10%" and "20%" refer to the ratio of the data with the lowest *AR* values that are selected from the data set (presented here without random sampling) for the calibration of the 5% and 10% thresholds respectively; Figure S2: Log–log plot of antecedent rain (mm) vs. landslide susceptibility (regional-scale [49]) for the landslide events on the reported day and the days prior and after that date (with the point size relative to their attributed weights, i.e., 0.67 and 0.17 respectively). The green and red curves are the *AR* thresholds at 5% and 10% exceedance probability levels respectively, obtained with the modified *AR-S* method (Figure 4) without adopting the bootstrapping statistical technique, using the calibration (CAL) data set only. Ndata is the number of data in the expanded calibration and validation (VAL) data sets; Code S1: R code for the modified *AR-S* threshold approach (Figure 4) with the bootstrapping statistical technique; Code S2: R code for the modified *AR-S* threshold approach (Figure 4) without the bootstrapping statistical technique.

**Author Contributions:** Conceptualization, E.M., O.D. and A.D. (Alain Demoulin); Data curation, E.M.; Formal analysis, E.M.; Funding acquisition, E.M. and O.D.; Investigation, E.M.; Methodology, E.M.; Project administration, E.M. and O.D.; Resources, E.M., O.D. and A.D. (Arthur Depicker); Software, E.M.; Supervision, O.D. and A.D. (Alain Demoulin); Validation, E.M.; Visualization, E.M.; Writing—original draft, E.M.; Writing—review & editing, E.M., O.D., A.D. (Arthur Depicker) and A.D. (Alain Demoulin)

**Funding:** This study was supported by the Belgium Science Policy (BELSPO) through (1) the PAStECA project (BR/165/A3/PASTECA) entitled 'Historical Aerial Photographs and Archives to Assess Environmental Changes in Central Africa' (http://pasteca.africamuseum.be/), (2) the RESIST project (SR/00/305) entitled 'Remote Sensing and In Situ Detection and Tracking of Geohazards' (http://resist.africamuseum.be/), (3) the GeoRisCA project (SD/RI/02A), entitled 'Geo-Risk in Central Africa: integrating multi-hazards and vulnerability to support risk management' (http://georisca.africamuseum.be), and (4) the AfReSlide project (BR/121/A2/AfReSlide) entitled 'Landslides in Equatorial Africa: Identifying culturally, technically and economically feasible resilience strategies' (http://afreslide.africamuseum.be/). E.M. was funded by F.R.S.–FNRS.

**Acknowledgments:** The authors thank Jente Broeckx for providing the continental landslide susceptibility model. Special thanks go to our partners at Université Officielle de Bukavu (DR Congo) and Centre de Recherche en Sciences Naturelles de Lwiro (DR Congo), who facilitated fieldwork in the study area and provided information on the timing of the landslides. We acknowledge the NASA Goddard Earth Sciences Data and Information Services Center for providing full access to the precipitation data sets exploited in this study. We also thank the reviewers for their constructive feedback.

**Conflicts of Interest:** The authors declare no conflict of interest.
