Modeling of resource acquisition processes in wheat and the coupling of SiriusQuality - Archisimple - Min3P

MUJICA CLAUDIO RAMÓN

Louvain-la-Neuve)

Workshop; SolACE. Final consortium meeting; 2022

SolACE

Crop models (CM) have been a useful tool in order to analyze the crop performance under different environments. Also, useful to integrate all the processes that determines the crop performance, to test different scenarios, to identify the key processes, testing hypotheses on interacting mechanisms derived from experimental results and to validate it simultaneously (Clausnitzer and Hopmans, 1994). CM involve simulation of the above-ground growth, crop yield, below-ground root system formation, root function for supplying water and nutrients to the plant and also the water storage in soil (Wang and Smith, 2004). These models have a lot of years of development (Hopmans and Bristow, 2002; Wang and Smith, 2004), but yet the main challenge is to reproduce the ability of plants to react to the different environmental changing conditions of availability of water, nutrients, temperatures, radiation, etc. To achieve this, Crop Models should represent the main physiological processes in the entire plant, as well as water, heat and chemical processes in the soil, as much mechanistically as possible. Otherwise, the results of modelling show dependence on the conditions where their parameters were derived.Most of the time CM has been devoted to the above-ground part of the crops, mainly due to the difficulties of the below-ground measurements. Therefore, CM have a better conceptualization and reproduction of the above-ground processes and a simplified modeling of the below-ground processes (Wang and Smith, 2004). Depending on the specific aim of each particular CM they conceptualize root system with different levels of complexity. In most of CM, the root functioning, is reduced to a simple variable to regulate the water and nutrients use of controlled by the above-ground processes and in biomass accumulation budget to conceptualize the sink of carbon. The simplest approach considers the root growth as a daily rate of increase in depth, controlled by specific crop and soil parameters (e.g. WOFOST, GECROS, LINTUL and DAISI models, Boogaard et al., 2011). Instead of a daily rate the root front could be also calculated by an exponential root deepening function (e.g in AQUACROP model).To calculate more realistic scenarios, the CM takes into account more processes and calculates these processes increasingly mechanistic way. Even with a level of empiricism, in how the functions were determined, it tends to include more physical, chemical and physiological mechanisms. Finally, in this increasing amount of processes to be considered, a representation of the root structure becomes necessary. For example, reproducing the root-soil interaction, the structural properties of root and the soil physical properties and water/nutrients availability are needed. The root system architecture (RSA) models aim for reproduce the 2 or 3 dimensional root architecture reproducing the emergence, the branching, growth and aging of roots constrained by different factors (Guo and Li, 2001; Pages et al., 2012; Wang and Smith, 2004). These RSA models could be integrated a crop model but they are mostly independent and externally coupled, e.g. OpenSimRoot, R-SWM or PlaNet-Maize (Javaux et al., 2008; Lobet et al., 2012; Postma et al., 2017) includes a crop and soil model and CRootBox, ArchiSimple or DigiR which are not coupled itself with crop or soil models (Barczi et al., 2018; Pages et al., 2012; Schnepf et al., 2018).The principal limitations of these RSA models is being too specific in terms of plant species, stages or processes considered and too complex, with many parameters. These issues where addressed by Pages et al., 2012, in the ArchiSimple RSA model, wich at the same time was provided with the capabilitie of simulate the relation to the soil medium and carbon availability from the aerial part. ArchiSimple represents three-dimensionally the root structure and function by means segments of predefined length, usually few millimeters. Every time step the main physiological processes involved in root growth are calculated to determine the amount and position of these segments. This allows to ArchiSimple to reproduce the emission of new adventitious roots from the shoot, elongation of existing roots, branching, radial growth and aging (Pages et al., 2012; Pagès et al., 2014).In CM the conceptualization of the soil available water is usually simple and most of the times calculated by the ‘tipping-bucket’ approach (Ritchie, 1998). The total extractable soil water is considered as the difference between drained upper limit soil water content parameter (DUL, is the limit where capillary forces are greater than gravity forces) and lower limit of plant-extractable soil water (LL, which corresponds with the wilting point) for each soil layer. On the other hand, by the same approach the actual extractable water is calculated as the difference between the soil water content (SW) and lower limit of plant-extractable soil water (LL) for each soil layer. Some CM incorporates other concepts (e.g. SAMARA,, conceptualizes the soil as two compartments. One of those, in the surface, is affected by evaporation and have a fixed depth. The other one, have a reduction in evaporation, simulating a self-mulching effect, and its extension depends on the root front. Both layers have 4 functional reservoirs simulating evaporation, transpiration and drainage processes. In AQUACROP the soil is divided in quarters and the water extraction is calculated in each one.As in RSA models for roots, there is more accurate and complex models for soil modelling. In order to reproduce the interaction between soil and roots is necessary to simulate water and nutrients movement across the soil profile. Commonly the water transport under unsaturated conditions is the main feature to reproduce soil functioning, but depending on the particular crop specie, soil or water status of the site, more processes are needed. For example, water fluxes in aquifers, gases and heat fluxes, nutrients transport, surface interactions between particles, microbiological and chemical reactions. Currently exists a big field in hydrological models (HM) with very detailed and accurate and mechanistic representation of the processes, but most of them simplify the root and plant, accounting them as sink terms (Steefel et al., 2015). The simplified sink approach was adequate for non-stress-plant-growth conditions and may work adequately for uniform soil conditions. However, it has become increasingly clear that a different approach is required if water and/or nutrient resources become limited in part of the root zone (Hopmans and Bristow, 2002).The aim of the present work is to couple the CM model Sirius Quality (SQ) and RSA model ArchiSimple (AS) with the hydrological model MIN3P.