@article{fdi:010065372,
  title = {{L}arge topsoil organic carbon variability is controlled by {A}ndisol properties and effectively assessed by {VNIR} spectroscopy in a coffee agroforestry system of {C}osta {R}ica},
  author = {{K}inoshita, {R}. and {R}oupsard, {O}. and {C}hevallier, {T}iphaine and {A}lbrecht, {A}lain and {T}augourdeau, {S}. and {A}hmed, {Z}. and van {E}s, {H}. {M}.},
  editor = {},
  language = {{ENG}},
  abstract = {{A}ssessing the spatial variability of soil organic carbon ({SOC}) is crucial for {SOC} monitoring and comparing management options. {T}opsoil (0-5 cm) {SOC} concentrations were surveyed in a coffee agroforestry watershed (0.9 km(2)) on {A}ndisols in {C}osta {R}ica with uniform farm management. {W}e encountered high values and large spatial variations of {SOC}, from 48.1 to 172 g kg(-1) in the dry combustion set ({SOC}ref; n = 72) used for calibrating the visible-near-infrared reflectance spectroscopy ({VNIRS}) samples ({SOCVNIRS}; 350-2500 nm; n = 520). {VNIRS} using partial least squares regression was effective in predicting {SOC} ({R}-2 = 0.85; a root mean square error ({RMSE}) = 12.3 g kg(-1)) and proved an effective proxy measurement. {W}e assessed several topographic, vegetation and andic soil property variables, of which only the latter (metal-humus complexes and allophanes) displayed strong correlations with {SOC}ref concentrations. {W}e compared {R}andom {F}orest and three geostatistical approaches for the interpolation of {SOC} in unsampled locations. {O}rdinary kriging with {SOC}ref yielded an {RMSE} of 28.0 g kg(-1). {R}andom {F}orest was successful in incorporating many weakly and non-linearly correlated covariates with {SOC} ({RMSE} = 14.7 g kg(-1)), provided {A}l-p (the sodium pyrophosphate extractable aluminum), the best predictor of {SOC} (r = 0.85) but also the most costly variable to acquire. {C}o-lcriging with {A}lp also showed high reduction in {RMSE} (16.0 g kg(-1)). {C}o-kriging with {SOCVNIRS} only showed marginal reduction in {RMSE} to 242 g kg(-1) due to the presence of a high nugget effect. {L}ocal variability of {SOC} in this volcanic agroforestry watershed was dominated by andic properties whereas topographic or vegetation variables had very little impact. {E}stimation of {SOC} variability is recommended using inexpensive proxy measurements like {VNIRS} ({RMSE} = 12.3 g kg-1) rather than spatial interpolation techniques.},
  keywords = {{S}oil organic carbon ; {VNIR} spectroscopy ; {R}andom {F}orest ; {C}o-kriging ; {A}ndisols ; {A}llophane ; {A}groforestry ; {COSTA} {RICA}},
  booktitle = {},
  journal = {{G}eoderma},
  volume = {262},
  numero = {},
  pages = {254--265},
  ISSN = {0016-7061},
  year = {2016},
  DOI = {10.1016/j.geoderma.2015.08.026},
  URL = {https://www.documentation.ird.fr/hor/fdi:010065372},
}