@article{fdi:010069424,
  title = {{E}stimation of available water capacity components of two-layered soils using crop model inversion : effect of crop type and water regime},
  author = {{S}reelash, {K}. and {B}uis, {S}. and {S}ekhar, {M}. and {R}uiz, {L}aurent and {T}omer, {S}. {K}. and {G}uerif, {M}.},
  editor = {},
  language = {{ENG}},
  abstract = {{C}haracterization of the soil water reservoir is critical for understanding the interactions between crops and their environment and the impacts of land use and environmental changes on the hydrology of agricultural catchments especially in tropical context. {R}ecent studies have shown that inversion of crop models is a powerful tool for retrieving information on root zone properties. {I}ncreasing availability of remotely sensed soil and vegetation observations makes it well suited for large scale applications. {T}he potential of this methodology has however never been properly evaluated on extensive experimental datasets and previous studies suggested that the quality of estimation of soil hydraulic properties may vary depending on agro-environmental situations. {T}he objective of this study was to evaluate this approach on an extensive field experiment. {T}he dataset covered four crops (sunflower, sorghum, turmeric, maize) grown on different soils and several years in {S}outh {I}ndia. {T}he components of {AWC} (available water capacity) namely soil water content at field capacity and wilting point, and soil depth of two layered soils were estimated by inversion of the crop model {STICS} with the {GLUE} (generalized likelihood uncertainty estimation) approach using observations of surface soil moisture ({SSM}; typically from 0 to 10 cm deep) and leaf area index ({LAI}), which are attainable from radar remote sensing in tropical regions with frequent cloudy conditions. {T}he results showed that the quality of parameter estimation largely depends on the hydric regime and its interaction with crop type. {A} mean relative absolute error of 5% for field capacity of surface layer, 10% for field capacity of root zone, 15% for wilting point of surface layer and root zone, and 20% for soil depth can be obtained in favorable conditions. {A} few observations of {SSM} (during wet and dry soil moisture periods) and {LAI} (within water stress periods) were sufficient to significantly improve the estimation of {AWC} components. {T}hese results show the potential of crop model inversion for estimating the {AWC} components of two-layered soils and may guide the sampling of representative years and fields to use this technique for mapping soil properties that are relevant for distributed hydrological modelling.},
  keywords = {{S}oil hydraulic properties ; {A}vailable water capacity ; {STICS} ; {S}oil water ; content ; {GLUE} ; {I}nverse modelling ; {INDE}},
  booktitle = {},
  journal = {{J}ournal of {H}ydrology},
  volume = {546},
  numero = {},
  pages = {166--178},
  ISSN = {0022-1694},
  year = {2017},
  DOI = {10.1016/j.jhydrol.2016.12.049},
  URL = {https://www.documentation.ird.fr/hor/fdi:010069424},
}