Matthew J. Smith

Artificial intelligence (AI) is beginning to live up to its promise of delivering real value, driven by recent advances in the availability of relevant data, computation and algorithms. In the present paper, I discuss the value to agriculture from AI over the next decade. The more immediate applications will be to improve precision information about what is happening on the farm by improving what is being detected and measured. A consequence of this are more accurate alerts to farmers. Another is an increased ability to understand why phenomena occur in farm systems, so as to improve their management. From improved data and understanding come improved predictions, enabling more optimal decisions about how to manage farm systems and stimulating the development of decision support and recommender systems. In many cases, robotics and automated systems will remove much of the need for human decision-making and improve farm efficiencies and farm health. Artificial intelligence will also be needed to enable organisations to harness the value of information distributed throughout supply chains, including farm data. Digital twins will also emerge as an important paradigm to improve how information about farm entities is organised to support decision-making. There are also likely to be negative impacts from AI, such as disruption to the roles and skills needed from farm workers, indicating the need to consider the social and ethical impacts of AI each time a new capability is introduced. I conclude that understanding these challenges more deeply tends to highlight new opportunities for positive change.

Improving international food security under a changing climate and increasing human population will be greatly aided by improving our ability to modify, understand and predict crop growth. What we predominantly have at our disposal are either process-based models of crop physiology or statistical analyses of yield datasets, both of which suffer from various sources of error. In this paper, we present a generic process-based crop model (PeakN-crop v1.0) which we parametrise using a Bayesian model-fitting algorithm to three different sources: data–space-based vegetation indices, eddy covariance productivity measurements and regional crop yields. We show that the model parametrised without data, based on prior knowledge of the parameters, can largely capture the observed behaviour but the data-constrained model greatly improves both the model fit and reduces prediction uncertainty. We investigate the extent to which each dataset contributes to the model performance and show that while all data improve on the prior model fit, the satellite-based data and crop yield estimates are particularly important for reducing model error and uncertainty. Despite these improvements, we conclude that there are still significant knowledge gaps, in terms of available data for model parametrisation, but our study can help indicate the necessary data collection to improve our predictions of crop yields and crop responses to environmental changes.

Terrestrial ecosystems have absorbed roughly 30 % of anthropogenic CO₂ emissions over the past decades, but it is unclear whether this carbon (C) sink will endure into the future. Despite extensive modeling and experimental and observational studies, what fundamentally determines transient dynamics of terrestrial C storage under global change is still not very clear. Here we develop a new framework for understanding transient dynamics of terrestrial C storage through mathematical analysis and numerical experiments. Our analysis indicates that the ultimate force driving ecosystem C storage change is the C storage capacity, which is jointly determined by ecosystem C input (e.g., net primary production, NPP) and residence time. Since both C input and residence time vary with time, the C storage capacity is time-dependent and acts as a moving attractor that actual C storage chases. The rate of change in C storage is proportional to the C storage potential, which is the difference between the current storage and the storage capacity. The C storage capacity represents instantaneous responses of the land C cycle to external forcing, whereas the C storage potential represents the internal capability of the land C cycle to influence the C change trajectory in the next time step. The influence happens through redistribution of net C pool changes in a network of pools with different residence times. Moreover, this and our other studies have demonstrated that one matrix equation can replicate simulations of most land C cycle models (i.e., physical emulators). As a result, simulation outputs of those models can be placed into a three-dimensional (3-D) parameter space to measure their differences. The latter can be decomposed into traceable components to track the origins of model uncertainty. In addition, the physical emulators make data assimilation computationally feasible so that both C flux- and pool-related datasets can be used to better constrain model predictions of land C sequestration. Overall, this new mathematical framework offers new approaches to understanding, evaluating, diagnosing, and improving land C cycle models.

Invasive species are a serious threat to biodiversity worldwide and predicting whether an introduced species will first establish and then become invasive can be useful to preserve ecosystem services. Establishment is influenced by multiple factors, such as the interactions between the introduced individuals and the resident community, and demographic and environmental stochasticity. Field observations are often incomplete or biased. This, together with an imperfect knowledge of the ecological traits of the introduced species, makes the prediction of establishment challenging. Methods that consider the combined effects of these factors on our ability to predict the establishment of an introduced species are currently lacking. We develop an inference framework to assess the combined effects of demographic stochasticity and parameter uncertainty on our ability to predict the probability of establishment following the introduction of a small number of individuals. We find that even moderate levels of demographic stochasticity influence both the probability of establishment, and, crucially, our ability to correctly predict that probability. We also find that estimation of the demographic parameters of an introduced species is fundamental to obtain precise estimates of the interaction parameters. For typical values of demographic stochasticity, the drop in our ability to predict an establishment can be 30% when having priors on the demographic parameters compared to having their accurate values. The results from our study illustrate how demographic stochasticity may bias the prediction of the probability of establishment. Our method can be applied to estimate probability of establishment of introduced species in field scenarios, where time series data and prior information on the demographic traits of the introduced species are available.

Ecology studies the interactions between individuals, species and the environment. The ability to predict the dynamics of ecological systems would support the design and monitoring of control strategies and would help to address pressing global environmental issues. It is also important to plan for efficient use of natural resources and maintenance of critical ecosystem services. The mathematical modelling of ecological systems often includes nontrivial specifications of processes that influence the birth, death, development and movement of individuals in the environment, that take into account both biotic and abiotic interactions. To assist in the specification of such models, we introduce MELA, a process algebra for Modelling in Ecology with Location Attributes. Process algebras allow the modeller to describe concurrent systems in a high-level language. A key feature of concurrent systems is that they are composed of agents that can progress simultaneously but also interact - a good match to ecological systems. MELA aims to provide ecologists with a straightforward yet flexible tool for modelling ecological systems, with particular emphasis on the description of space and the environment. Here we present four example MELA models, illustrating the different spatial arrangements which can be accommodated and demonstrating the use of MELA in epidemiological and predator-prey scenarios.Comment: In Proceedings QAPL'16, arXiv:1610.0769

Improving international food security under a changing climate and increasing human population will be greatly aided by improving our ability to modify, understand and predict crop growth. What we predominantly have at our disposal are either process based models of crop physiology or statistical analyses of yield datasets, both of which suffer from various sources of error. In the current paper we present a generic process based crop model which we parametrise using a Bayesian model fitting algorithm to three different sources of data – space based vegetation indices, eddy covariance productivity measurements and regional crop yields. We show that the model parametrised without data, based on prior knowledge of the parameters can largely capture the observed behaviour but the data constrained model greatly improves both the model fit and reduces prediction uncertainty. We investigate the extent to which each dataset contributes to the model performance and show that while all data improves on the prior model fit, the satellite based data and crop yield estimates are particularly important for reducing model error and uncertainty. Despite these improvements, we conclude that there are still significant knowledge10 gaps, in terms of available data for model parametrisation, but our study can help indicate the necessary data collection steps for improvement in our predictions of crop yields and crop responses to environmental changes.