Stics is a robust, multipurpose tool that has been developed since the late 1990s by scientists from a number of research institutes, for use by researchers, teachers and professionals in the agricultural sector. It is designed to improve our understanding of how agro-ecosystems function and to support the agro-ecological transition in the face of climatic, environmental and societal challenges.

This is a model used in agriculture to simulate and understand the processes involved in soil and crop functioning. It is a deterministic model (for the same set of input data, it always produces the same result), mechanistic (based on physical and biological processes) and dynamic, operating on a daily time scale.

The Stics model and its formalisms are fully described in a dedicated book (Beaudouin et al., 2023).  Readers are invited to consult the book (available free of charge in ebook and pdf formats) for more details.

What can be simulated with STICS?

Stics simulates energy, water, carbon and nitrogen balances throughout the crop cycle. The model is capable of representing the functioning of annual crops (wheat, maize, soya, etc.), bispecific associated crops (wheat-pea, etc.), perennial crops (grassland, vines, Miscanthus, etc.), intermediate crops (mustard, oats, etc.), as well as periods of bare soil.  A simulation with STICS corresponds to a homogeneous spatial unit (soil, climate, cropping practices). In terms of time, STICS can simulate a crop cycle (from sowing to harvesting) or a succession of crops over several years.

As input, the model needs data relating to the characteristics of the soil explored by the root system (for example, the thickness of the different horizons, their bulk density and their moisture content at the permanent wilting point and at field capacity) as well as daily climatic variables characterising the state of the atmosphere in the vicinity of the system (see Validity domain section below). The model can be used to simulate the effect of farming practices on the system, which need information on tillage, sowing operations (date, depth and density of sowing), inputs of organic products, mineral fertilisation (dates, doses and types of fertiliser), irrigation, and foliage management for perennial plants.  The initial state of the system must also be described, such as the water and mineral nitrogen content of the soil, or the initial state of the plant where applicable. Finally, it is essential to enter the characteristics of the species grown, or even the variety, in order to take into account the ecophysiology specific to the crop in question (for example, its phenology or nitrogen requirements). The model can be used to simulate plant growth, which may be limited by stresses of varying intensity and length, derived from the water and nitrogen balances. The necessary parameter sets are supplied with the model for a large number of species and varieties.

The outputs of Stics reflect the objectives for which it was built. The evolution of a wide range of agronomic variables (phenology, biomass growth, yield development, quality, stress, etc.) and outputs that provide an environmental balance (water balance, soil carbon and organic nitrogen stocks, quantity of nitrates leached, N₂O emissions, etc.). ) can be simulated by the model. The variables of interest are simulated on a daily time scale, over the entire soil profile and over the entire plant cycle, or even over several cycles.

Philosophy behind the development of the model

From the outset of its development, STICS has been designed to meet four main criteria:

• The balance in the complexity and description of the processes involved in the soil-plant-atmosphere continuum. This balance aims at proposing a model that can be applied in a wide variety of agricultural contexts.
• The genericity of the algorithms used to describe the plant function using non-specific eco-physiological concepts enabling a single model capable of simulating the growth of a wide variety of plants.
• Robust algorithms and parameterisation to provide realistic results under a wide range of agri-environmental conditions.
• The simplicity of the input data required to run the model as well as easily accessible (and if possible directly measurable) model parameters with low sensitivity to changes in scale.

The design of a dynamic and functional model with a strong agri-environmental focus has led to the emergence of a fifth quality: scalability. This is illustrated by the development of new formalisms over time to simulate perennial cropping or intercropping systems. This scalability also resulted in the model being used at different spatial scales, from the plot scale to the macro-regional scale.

Validity domain

Using a 1D description of the agroecosystem (2D for some specific row crops and for associated crops), Stics simulates the processes at work in the soil-plant system, including the plant cover, the surrounding microclimate and the soil (in particular the part of the soil colonised by the roots).

The upper limit of the simulated system is delimited by the lowest layer of the atmosphere (2 m high), where the standard meteorological variables are characterised (radiation, minimum and maximum temperatures, rainfall, reference evapotranspiration and, possibly, wind and humidity). The climatic data have a forcing function and are referred to as driving variables.

The lower limit of the simulated system corresponds to the soil/subsoil interface where water and nutrients become inaccessible to the crop root system. Soil properties are defined at the level of the soil horizon, but most soil-related processes are simulated at centimetre scale (elementary layers).

How the model works?

Stics is structured into computer ‘modules’, each module corresponding to a set of processes (phenological development, leaf growth, radiation interception and photosynthesis, yield and quality development, root growth, water balance, soil nitrogen transformations, microclimate, heat, water and nitrate transfers).

In functional terms, STICS is structured around a dynamic scheme for the daily growth of the plant cover based on the carbon function of plants: the radiation intercepted by the photosynthetically active apparatus, characterised by the leaf index, is transformed into biomass distributed to the various organs. This distribution of assimilates is based on source-sink balances.

The development module controls the growth of cultivated plants by organising the opening and closing of sinks and their intensity throughout the cycle. It also acts on the sources by controlling the setting up of the photosynthetically active apparatus and by activating remobilisation towards the storage organs.

The model simulates water and nitrogen balances in the soil, in order to assess their availability to plants and potential losses to the environment. Their descriptions are based on the classic compartment approach, which defines the different reservoirs in the system, their evolution and their relationships. The functions cover the physical, physico-chemical and biological processes used to simulate soil transfers and plant uptake, as well as transformations between the organic and mineral pools.

Based on the water, nitrogen and carbon balances, trophic stresses (water and nitrogen) of varying intensity and length are simulated. Other abiotic stresses, such as thermal stress (frost or high temperatures) or soil waterlogging, can also be taken into account. Response functions to environmental stresses can also include enzymatic activities, such as nitrate uptake. These stresses are considered in STICS as constraints on the potential functioning of the canopy, slowing down or stopping the development of cultivated plants and/or the development of yield. Work has also made it possible to take into account certain biotic stresses (fungal diseases) by coupling STICS with the MILA model (Caubel et al., 2014).

Genesis and development of the model

The STICS model was developed in the mid-1990’s by a group of INRA agronomists and soil scientists from various research units in France. This acronym stands for “Scientific, Technical and Interdisciplinary simulator of soil-Crop System functioning” (Simulateur mulTIdisciplinaire pour les Cultures Standard, in French). Behind STICS lies a real success story, based on a collective project that owes much to the first coordinator of the STICS “adventure”, the late Nadine Brisson. The STICS model was born out of the ECOSPACE project - 'heterogeneity of cultivated environments' - funded by INRA, which provided an opportunity to combine the expertise of three former departments of INRA: Bioclimatology, Agronomy and Soil Sciences. At that time, there were already agronomic models available at international level with different orientations: ecophysiological models of photosynthetic production (e.g. SUCROS), biogeochemical models (e.g. PASTIS, CENTURY), and more general soil-crop models (e.g. CERES, EPIC, APSIM, CROPSYS). While some of these models were being adapted to the French context, a group of French researchers sought to mergethe most recent knowledge in crop and soil sciences to create a new agroecosystem model: STICS.

The first version of STICS results from the merging of three dynamic models, namely GOA - a plant growth model (Ruget et al., 1993),  BYM  - a soil water balance model (Brisson et al., 1992),  and LIXIM - a model dedicated to simulation of water and nitrogen transfers in the soil (Mary et al., 1999).

Stics was initially applied to simulate the growth of wheat and maize crops (Brisson et al., 1998, Brisson et al., 2002). The model then evolved, driven by a wide range of new issues: estimation of regional production potential; crop water and nitrogen requirements; effect of farming practices on nitrate leaching; precision agriculture, including assimilation of remote sensing data; monitoring of French forage production and forecast projections; climate change impacts and adaptation; and added value of cover crops. In recent years, the themes covered by the model have expanded to include the assessment of agroecological cropping systems, intermediate crops, perennial crops, bioenergy crops, greenhouse gas (N₂O) emissions and long-term soil carbon sequestration.

Very early on, the idea emerged that STICS should not become a fixed model, but rather an interactive modelling platform. Since then, it has always been developed collaboratively, at the intersection of scientific and practical concerns. The genesis and early evolution of the model were made possible by the exceptional leadership of Nadine Brisson and her ability to establish collaborations with researchers of diverse skills and affiliations, as well as with technical experts. After Nadine's death, the Stics project team (SPT) took over the maintenance and development of the model.

Future development and maintenance of the model

Regarding future developments of the model, two entities ensure complementary functions:

• The STICS users’ group, which meets every two years for a seminar, performs the ultimate assessment of the standard version and expresses needs for further development of the model. It shares its propositions with the Project Stics Team.  In 2014, it becomes a scientific network of the INRAE's AgroEcoSystem Department .
• The Stics Project Team is responsible for the governance of the STICS network, setting priorities and developing the model.

The model is managed by the AgroClim Service Unit at the INRAE PACA Centre (Avignon).