Soil Model Brings Hidden Springtail Worlds Into Digital Landscapes

Beneath every agricultural field lies a bustling miniature metropolis, home to billions of tiny creatures that most people never see. Among them is Folsomia candida, a small white springtail that has quietly served science for over six decades as a standard test organism in ecotoxicology. Now, researchers have built something remarkable: a formal population model that brings this soil-dwelling creature to life inside a sophisticated digital landscape, complete with hourly weather, daily soil moisture, and spatially explicit environmental stressors. The work, published in Agricultural and Environmental Modelling, represents a significant step forward in understanding how soil organisms respond to the complex, patchy realities of agricultural environments.

Collembola, commonly known as springtails, are among the most abundant soil invertebrates on Earth, with an estimated 100,000 individuals per square metre in the soil and leaf litter layers of most terrestrial ecosystems. They maintain soil health by fragmenting organic matter, regulating microbial communities, and serving as prey for spiders, mites, millipedes, and ants. The European Union has identified soil health as a main priority in its Soil Strategy for 2030, making the fate of these tiny decomposers a matter of continental policy concern. Yet despite their ecological importance, how springtail populations respond to environmental stressors across real, spatially structured landscapes has remained poorly understood.

The new model, developed within the Animal, Landscape and Man Simulation System known as ALMaSS, uses a subpopulation approach to represent F. candida life cycles under varying environmental conditions in European agricultural landscapes. It integrates stage-structured population dynamics with spatially explicit landscape processes at 10 by 10 metre resolution. The model explicitly represents egg, juvenile, and adult life stages, with transitions between stages and vital rates responding non-linearly to temperature, soil moisture, and food quality through empirically parameterised thermal performance curves and dose-response functions fitted to published experimental data.

What makes this model genuinely novel is its treatment of soil moisture. Previous population models for springtails have typically represented life cycles under constant temperature or without moisture effects at all. Only one prior model, a stage-structured model for the springtail Paronychiurus kimi, explicitly incorporated both temperature and soil moisture, but it relied on degree-day theory and lacked spatial interaction. An individual-based model for F. candida developed by Meli and colleagues in 2014 represented spatial interactions but did not incorporate soil moisture into life-history processes. The new work addresses both limitations simultaneously.

The modelling of surface soil moisture is described as a novel addition to ALMaSS. Rather than using water content directly, the researchers model soil water potential, which quantifies the energy status of water retained by the soil matrix and reflects the effort required by soil organisms to extract water. Soil water potential is considered a more universally comparable measure than water content because it allows comparisons across different soil types. To estimate it, the team converts hourly volumetric water content from the ERA5 reanalysis dataset using soil water retention curves generated by a physics-informed pedotransfer function trained on 4,200 soil samples well distributed across Denmark.

Because the native spatial resolution of the ERA5 dataset, approximately 28 kilometres, far exceeds the typical ALMaSS landscape window of 10 by 10 kilometres, the researchers redistribute soil moisture to each landscape polygon based on slope, elevation, and evapotranspiration. Water accumulates in flatter and lower-lying areas, so terrain characteristics are used to refine the moisture estimates accordingly. Evapotranspiration itself depends on crop types and weather, estimated through a crop coefficient approach that relates the coefficient to leaf area index, with reference evapotranspiration values assigned by agroclimatic region and temperature band.

The life-history components of the model are driven by thermal performance curves based on the Lobry-Rosso-Flandrois model, which characterises the relationship between temperature and biological development rate using four parameters: thermal optimum, maximum temperature, minimum temperature, and maximum performance rate. Hatching time data from 4 to 24 degrees Celsius were compiled from multiple published studies, with eggs failing to hatch above 28 degrees Celsius and optimal hatching success at 21 degrees Celsius. Time to first oviposition, marking the transition from juvenile to adult, was collected across 4 to 26 degrees Celsius, with no reproduction occurring at 0 or 30 degrees Celsius.

Soil moisture effects are captured using a three-phase model for egg development, reparameterised for both egg-to-juvenile and juvenile-to-adult transitions. Development generally ceases below a moisture threshold, and the model assumes a constant rate when soil water potential ranges from 0 to 0.11 megapascals, followed by a linear decline until development stops entirely at drier conditions. For reproduction, the total rate is determined by taking the minimum of the temperature-driven and moisture-driven rates, reflecting the physiological principle that favourable moisture cannot compensate for temperatures that already impose strong constraints, and vice versa.

Mortality in the model arises from three components: temperature stress, soil-moisture stress, and predation. Temperature-induced mortality differs among life stages, with hatchability modelled using a three-parameter log-logistic function and juvenile mortality fitted separately for temperatures below and above 20 degrees Celsius. The model predicts complete mortality when soil temperature exceeds approximately 32 degrees Celsius, though the researchers note that natural soils exhibit substantial microclimatic heterogeneity, and individual springtails may occupy cooler or moister microsites within soil aggregates or pore spaces that buffer exposure to lethal extremes.

The formal model documentation establishes the theoretical foundation and empirical basis for subsequent implementation, evaluation, and application to questions in landscape ecology and environmental risk assessment. The researchers are explicit about the model’s boundaries: it does not incorporate physiological acclimation, evolutionary adaptation, or genetic variation, and food quality and availability are simplified as a score rather than mechanistic diet dynamics. Outputs such as population growth rates, spatial distribution, and recovery should be interpreted as indicators of relative risk rather than absolute estimates of abundance. While the European Food Safety Authority guidelines caution against using F. candida for regulatory environmental risk assessment of plant protection products, the species remains highly relevant within ecotoxicological testing frameworks, and this model could improve the interpretation and prediction of ecotoxicological experiment results while serving as a template adaptable to other soil species with comparable ecology.

Subject of Research: Integrating spatial and environmental stressors in a population model of Folsomia candida ( Collembola , Isotomidae ): a Formal Model within ALMaSS framework

Article Title: Integrating spatial and environmental stressors in a population model of Folsomia candida ( Collembola , Isotomidae ): a Formal Model within ALMaSS framework

Article References: Xie, L., Duan, X., Norouzi, S., de Jonge, L. W., & Topping, C. J. (2026). Integrating spatial and environmental stressors in a population model of Folsomia candida (Collembola, Isotomidae): a Formal Model within ALMaSS framework. Agricultural and Environmental Modelling, 8, Article e184962. https://doi.org/10.3897/aem.8.184962

Image Credits: AI Generated

DOI: 10.3897/aem.8.184962

Keywords: Integrating, spatial, environmental, stressors, population, model, Folsomia, candida, Collembola, Isotomidae, Formal, within