We introduce and analyze a class of neural network models motivated by the Drosophila central complex, designed to capture the emergence and dynamics of orientation-selective activity bumps. Starting from a biologically inspired ring connectivity model, we derive a simplofied reduced model of recurrent neural activity that supports stable localized patterns encoding angular position. We first study the deterministic dynamics and identify parameter regimes ensuring existence and global stability of bump solutions. We then extend the framework to a stochastic setting, incorporating both additive Brownian noise and a Markovian switching mechanism representing time-varying external cues. The resulting system is a switching diffusion with piecewise linear drift, for which we establish well-posedness, characterize the infinitesimal generator, and prove the existence of an invariant measure. Numerical simulations in low and high dimensions illustrate the robustness of the bump attractor under noise and switching stimuli, as well as the convergence toward the predicted stationary states. These results provide a mathematically tractable framework for understanding how population activity in the insect central complex encodes heading direction in the presence of variability.
Following crop growth through the vegetative cycle allows farmers to predict fruit setting and yield in early stages, but it is a laborious and non-scalable task if performed by a human who has to manually measure fruit sizes with a calliper or dendrometers. In recent years, computer vision has been used to automate several tasks in precision agriculture, such as counting fruits and estimating their size. However, the fundamental problem of matching fruits from one video to another, collected on a later date, which is needed to track fruits’ growth through time, remains to be solved. Few attempts were made, but either they assumed that the camera always starts from the same known position and that there are sufficiently distinct visual features to match, or they used other data sources, like GNSS. Here we propose a new paradigm to tackle this problem, based on constellations of 3D fruit centroids, and introduce a descriptor for very sparse 3D point clouds that can be used to match fruits across videos. Matching constellations instead of individual fruits is key to deal with non-rigidity, occlusions, and challenging imagery with few distinct visual features to track. The results show that the proposed method can be successfully used to match fruits across videos and through time, and also to build an orchard map and later use it to locate the camera pose in 6DoF, thus enabling, for example, autonomous navigation of robots in the orchard or selective fruit picking.
The air-sea CO2 flux FCO2 is an important component of the global carbon cycle and understanding its response to climate change is crucial to adjust mitigation pathways. Multi-linear regression supports the expectation that the balance between the CO2 partial pressures of air and the sea surface (pCO2) is the most important driver of temporal FCO2 variability. Discrepancies in FCO2 variability between state-of-the-art Earth System Models (ESMs) and gap-filled pCO2-products suggest that systematic biases exist across an ensemble of ESMs. In the high latitudes, the climate change induced trend towards lighter seawater is overestimated in ESMs, which yields - in contrast to observations - shallower mixed layers over the contemporary period and hence a suppressed carbon supply from depth. While mixed layer depth variability and trends appear biased throughout the global ocean, this is not a determining factor for pCO2 variability in the strongly stratified subtropical gyres. The results highlight the importance of accurately modeling circulation and hydrographic properties to obtain robust estimates of FCO2 and its variability.
Context. High-mass stars and star clusters form from the fragmentation of massive dense clumps driven by gravity, turbulence, and magnetic fields. The extent to which each of these agents impacts the fragmentation depending on the clump mass, density, and evolutionary stage is still largely unknown. Aims. The ALMA evolutionary study of high-mass protocluster formation in the GALaxy (ALMAGAL) project, with similar to 1000 clumps observed at similar to 1000 au resolution, allows a statistically significant characterization of the fragmentation process over a large range of clump physical parameters and evolutionary stages. Our goal is to characterize where and how the dense cores revealed by ALMA are distributed in massive potentially cluster-forming clumps to trace how fragmentation is initially set and how it proceeds before gas dispersal due to stellar feedback. Methods. We characterized the spatial distribution of dense cores in the 514 ALMAGAL clumps that host at least four cores, using a set of quantitative descriptors that we evaluated against the clump bolometric luminosity-to-mass ratio, which we adopted as an indicator of the evolution of the system. We measured the separations between cores with the minimum spanning tree (MST) method, which we compared with the predictions of gravitational fragmentation from Jeans theory. We investigated whether cores have specific arrangements using the Q parameter or variations due to their masses with the mass segregation ratio, Lambda(MSR). Results. ALMAGAL cores are distributed throughout the entire area of the clump, usually arranged in elliptical groups with an axis ratio e similar to 2.2, although high values with e >= 5 are also observed. We found a single characteristic core separation per clump in similar to 76% of cases, suggesting that multiple fragmentation lengths may be frequently present. Typical core separations are compatible with the clump-averaged thermal Jeans length,lambda(th)(J). However, we found an additional population of cores, typical of low-fragmented and young clumps, which are on average more widely separated with l approximate to 3 & times; lambda(th)(J). By stacking the distributions of the core separations in clumps of similar evolutionary stage, we also found that the separation decreases on average from l similar to 22 000 au in younger systems to l similar to 7000 au in more evolved ones. The ALMAGAL cores are typically distributed in fractal-type subclusters, while centrally concentrated patterns appear only at later stages, but we do not observe a progressive transition between these configurations with evolution. Finally, we also found 110 ALMAGAL systems with a signature of mass segregation, with an occurrence that increases with evolution.
Game species are valuable resources in many regions and contribute to a range of ecosystem services, yet they are often studied and managed individually despite responding to similar environmental conditions. This study examined spatial variation in 10 game species across a coastal-inland gradient in Norway by (1) examining regional variation in relative abundance, and (2) analysing spatial patterns of game species assemblages in relation to land-cover composition and elevation. We used harvest density as a proxy for relative abundance, and analysed data using negative binomial regression, self-organising maps, and principal component analysis. Significant regional differences were observed for several species, including moose (Alces alces), red deer (Cervus elaphus), roe deer (Capreolus capreolus), black grouse (Lyrurus tetrix), mountain hare (Lepus timidus), and capercaillie (Tetrao urogallus). The relative harvest density of red fox (Vulpes vulpes), pine marten (Martes martes), rock ptarmigan (Lagopus muta), and willow grouse (Lagopus lagopus) did not differ significantly among regions. Spatial clustering identified five game species assemblages (red deer, willow grouse, moose-forest grouse, roe deer-moose, willow grouse–ptarmigan) defined by distinct combinations of land-cover types, elevation and harvest densities across municipalities. Assemblages were distributed across gradients from coastal semi-natural landscapes to inland forest–agricultural mosaics and from lowland to higher-elevation environments. Our findings show that game species assemblages vary systematically with landscape composition and elevation across the coastal–inland gradient. This highlights the value of an assemblage-based perspective for understanding spatial patterns and integrating game species management with land-use planning.
