Inverse meta-modelling to estimate soil available water capacity at high spatial resolution across a farm

Abstract

Geo-referenced information on crop production that is both spatially-and temporally-dense would be useful for management in precision agriculture (PA). Crop yield monitors provide spatially but not temporally dense information. Crop growth simulation modelling can provide temporal density, but traditionally fail on the spatial issue. The research described was motivated by the challenge of satisfying both the spatial and temporal data needs of PA. The methods presented depart from current crop modelling within PA by introducing meta-modelling in combination with inverse modelling to estimate site-specific soil properties. The soil properties are used to predict spatially-and temporally-dense crop yields. An inverse meta-model was derived from the agricultural production simulator (APSIM) using neural networks to estimate soil available water capacity (AWC) from available yield data. Maps of AWC with a resolution of 10 m were produced across a dryland grain farm in Australia. For certain years and fields, the estimates were useful for yield prediction with APSIM and multiple regression, whereas for others the results were disappointing. The estimates contain 'implicit information' about climate interactions with soil, crop and landscape that needs to be identified. Improvement of the meta-model with more AWC scenarios, more years of yield data, inclusion of additional variables and accounting for uncertainty are discussed. We concluded that it is worthwhile to pursue this approach as an efficient way of extracting soil physical information that exists within crop yield maps to create spatially-and temporally-dense datasets.

Meta‐modeling soil organic carbon sequestration potential and its application at regional scale

“…They can be used as abstractions and simplifications of the simulation model and have been frequently applied in simulation studies Sargent 2000, Kleijnen et al 2005). Based on simulations of the Agricultural Production Systems sIMulator (APSIM), for example, Florin et al (2011) derived an inverse meta-model using crop yield data to estimate soil available water capacity at high spatial resolution. They found that the meta-model could explain ;70% of the variation in crop yield predicted by APSIM.…”

Available water capacity from a multidisciplinary and multiscale viewpoint. A review

“…4.2. Nevertheless, the matric potential at PWP can vary greatly among plant species, ranging from as low as − (Ridler et al 2012) Noah (Santanello et al 2007;Small 2007, 2010) SEtHyS (Coudert et al 2008) SWAP (Jhorar et al 2002(Jhorar et al , 2004Das et al 2008;Ines and Mohanty 2008;Singh et al 2010;Charoenhirunyingyos et al 2011;Shin et al 2013) WAVE (Ritter et al 2003) Forest 3-PG (Coops et al 2012) Crop models APSIM (Florin et al 2011) CERES (Link et al 2006;Braga and Jones 2004;Dente et al 2008) CropGro (Irmak et al 2001;Ferreyra et al 2006) STICS (Guérif et al 2006;Varella et al 2010aVarella et al , 2010bJégo et al 2012Jégo et al , 2015Sreelash et al 2012Sreelash et al , 2017Yemadje-Lammoglia et al 2018) Coupled soil-vegetation models Agro-hydrological models TNT2 (Ferrant et al 2016) 16,000 to -3500 kPa for xerophilic species to higher than -1000 kPa for hydrophilic species (Gobat et al 2004). The water content at -1500 kPa can thus differ from that "at the lowest limit in the field," as used in the definition of TTSW (Ratliff et al 1983).…”

“…Evapotranspiration/surface temperature Burke et al 1998;Jhorar et al 2002Jhorar et al , 2004Small 2007, 2010;Coudert et al 2008;Singh et al 2010;Charoenhirunyingyos et al 2011;Ridler et al 2012;Shin et al 2013LAI Guérif et al 2006Dente et al 2008;Varella et al 2010aVarella et al , 2010bCharoenhirunyingyos et al 2011;Coops et al 2012;Jégo et al 2012Jégo et al , 2015Ferrant et al 2016;Yemadje-Lammoglia et al, 2018Yield He et al 2001Timlin et al 2001;Morgan et al 2003;Braga and Jones 2004;Ferreyra et al 2006;Link et al 2006;Jiang et al 2008;Florin et al 2011;Yemadje-Lammoglia et al, 2018 Plant nitrogen content Varella et al 2010aVarella et al , 2010b…”

“…In comparison, spatial applications for areas of a few hundred m 2 to a few km 2 can benefit from adequate knowledge of the modeled situation, especially the availability of in situ measurements and prior information, low uncertainty in model inputs, and a model that is adequate for the situations considered. Most applications described in the literature focus on precision agriculture (e.g., Braga and Jones 2004;Florin et al 2011;Link et al 2006;Morgan et al 2003;Timlin et al 2001). Their objective was to assess variability in soil properties to optimize crop management (e.g., irrigation, fertilization) at the intra-field scale.…”

Estimating soil available water capacity within a Mediterranean vineyard watershed using satellite imagery and crop model inversion

“…Several studies were devoted to estimating soil hydrological properties or soil depth, by using different types of models devoted to subsurface water flows (Šimůnek et al, 2016, Galleguillos et al, 2017, Javaux et al, 2008, crop functioning (Florin et al, 2011, Dente et al, 2008Sreelash et al, 2017) or Soil -Vegetation -Atmosphere Transfer (Olioso et al, 2005, Gutmann et al, 2010, Bandara et al, 2015. Follow on from the literature review we discuss in introduction, we considered in the current study the STICS crop model for estimating SAWC components by inversion, and we selected three constraint variables for the fitting process, either alone or in combination, namely leaf area index (LAI), evapotranspiration (ET) and surface soil moisture (SSM).…”