@article{fdi:010083979,
  title = {{C}limatic and morphometric explanatory variables of glacier changes in the {A}ndes (8-55 degrees {S}) : new insights from machine learning approaches},
  author = {{C}aro, {A}. and {C}ondom, {T}homas and {R}abatel, {A}.},
  editor = {},
  language = {{ENG}},
  abstract = {{O}ver the last decades, glaciers across the {A}ndes have been strongly affected by a loss of mass and surface areas. {T}his increases risks of water scarcity for the {A}ndean population and ecosystems. {H}owever, the factors controlling glacier changes in terms of surface area and mass loss remain poorly documented at watershed scale across the {A}ndes. {U}sing machine learning methods ({L}east {A}bsolute {S}hrinkage and {S}election {O}perator, known as {LASSO}), we explored climatic and morphometric variables that explain the spatial variance of glacier surface area variations in 35 watersheds (1980-2019), and of glacier mass balances in 110 watersheds (2000-2018), with data from 2,500 to 21,000 glaciers, respectively, distributed between 8 and 55 degrees {S} in the {A}ndes. {B}ased on these results and by applying the {P}artitioning {A}round {M}edoids ({PAM}) algorithm we identified new glacier clusters. {O}verall, spatial variability of climatic variables presents a higher explanatory power than morphometric variables with regards to spatial variance of glacier changes. {S}pecifically, the spatial variability of precipitation dominates spatial variance of glacier changes from the {O}uter {T}ropics to the {D}ry {A}ndes (8-37 degrees {S}) explaining between 49 and 93% of variances, whereas across the {W}et {A}ndes (40-55 degrees {S}) the spatial variability of temperature is the most important climatic variable and explains between 29 and 73% of glacier changes spatial variance. {H}owever, morphometric variables such as glacier surface area show a high explanatory power for spatial variance of glacier mass loss in some watersheds (e.g., {A}chacachi with r(2) = 0.6 in the {O}uter {T}ropics, {R}io del {C}armen with r(2) = 0.7 in the {D}ry {A}ndes). {T}hen, we identified a new spatial framework for hydro-glaciological analysis composed of 12 glaciological zones, derived from a clustering analysis, which includes 274 watersheds containing 32,000 glaciers. {T}hese new zones better take into account different seasonal climate and morphometric characteristics of glacier diversity. {O}ur study shows that the exploration of variables that control glacier changes, as well as the new glaciological zones calculated based on these variables, would be very useful for analyzing hydro-glaciological modelling results across the {A}ndes (8-55 degrees {S}).},
  keywords = {climate drivers ; morpho-topographic drivers ; glacier changes ; machine learning ; clustering ; andes ; {PEROU} ; {BOLIVIE} ; {CHILI} ; {ARGENTINE} ; {ANDES}},
  booktitle = {},
  journal = {{F}rontiers in {E}arth {S}cience},
  volume = {9},
  numero = {},
  pages = {713011 [21 p.]},
  year = {2021},
  DOI = {10.3389/feart.2021.713011},
  URL = {https://www.documentation.ird.fr/hor/fdi:010083979},
}