Optimal spatial monitoring of populations described by reaction-diffusion models.

Using spatialised population measurements and related geographic habitat data, it is feasible nowadays to derive parsimonious spatially explicit population models and to carry on their parameter estimation. To achieve such goal, reaction-diffusion models are common in conservation biology and agricultural plant health where they are used, for example, for landscape planning or epidemiological surveillance. Unfortunately, if the mathematical methods and computational power are readily available, biological measurements are not. Despite the high throughput of some habitat related remote sensors, the experimental cost of biological measurements are one of the worst bottleneck against a widespread usage of reaction-diffusion models. Hence we will recall some classical methods for optimal experimental design that we deem useful to spatial ecologist. Using two case studies, one in landscape ecology and one in conservation biology, we will show how to construct a priori experimental design minimizing variance of parameter estimates, enabling optimal experimental setup under constraints.

Gene Introgression in Weeds Depends on Initial Gene Location in the Crop: Brassica napus-Raphanus raphanistrum Model.

The effect of gene location within a crop genome on its transfer to a weed genome remains an open question for gene flow assessment. To elucidate this question, we analyzed advanced generations of intergeneric hybrids, derived from an initial pollination of known oilseed rape varieties (Brassica napus, AACC, 2n = 38) by a local population of wild radish (Raphanus raphanistrum, RrRr, 2n = 18). After five generations of recurrent pollination, 307 G5 plants with a chromosome number similar to wild radish were genotyped using 105 B. napus specific markers well distributed along the chromosomes. They revealed that 49.8% of G5 plants carried at least one B. napus genomic region. According to the frequency of B. napus markers (0-28%), four classes were defined: Class 1 (near zero frequency), with 75 markers covering ∼70% of oilseed rape genome; Class 2 (low frequency), with 20 markers located on 11 genomic regions; Class 3 (high frequency), with eight markers on three genomic regions; and Class 4 (higher frequency), with two adjacent markers detected on A10. Therefore, some regions of the oilseed rape genome are more prone than others to be introgressed into wild radish. Inheritance and growth of plant progeny revealed that genomic regions of oilseed rape could be stably introduced into wild radish and variably impact the plant fitness (plant height and seed number). Our results pinpoint that novel technologies enabling the targeted insertion of transgenes should select genomic regions that are less likely to be introgressed into the weed genome, thereby reducing gene flow.

Pathogen population dynamics in agricultural landscapes: the Ddal modelling framework.

Modelling processes that occur at the landscape scale is gaining more and more attention from theoretical ecologists to agricultural managers. Most of the approaches found in the literature lack applicability for managers or, on the opposite, lack a sound theoretical basis. Based on the metapopulation concept, we propose here a modelling approach for landscape epidemiology that takes advantage of theoretical results developed in the metapopulation context while considering realistic landscapes structures. A landscape simulator makes it possible to represent both the field pattern and the spatial distribution of crops. The pathogen population dynamics are then described through a matrix population model both stage- and space-structured. In addition to a classical invasion analysis we present a stochastic simulation experiment and provide a complete framework for performing a sensitivity analysis integrating the landscape as an input factor. We illustrate our approach using an example to evaluate whether the agricultural landscape composition and structure may prevent and mitigate the development of an epidemic. Although designed for a fungal foliar disease, our modelling approach is easily adaptable to other organisms.