Recent observations demonstrating that auroral beads are observed prior to the majority of substorm onsets (Kalmoni et al., 2017, https://doi.org/10.1002/2016GL071826) have reinforced the potential importance of ballooning instabilities for near‐Earth magnetospheric substorm onset. Here we examine pressure anisotropic ballooning instabilities in stretched magnetotail geometries. Our results show that transition from an initially perpendicular anisotropy toward parallel anisotropy reduces the plasma β threshold for triggering a ballooning instability. Such increasingly parallel anisotropies can form as a direct consequence of tail stretching that occurs during the late substorm growth phase through the well‐known effects of drift shell splitting and the competition between betatron and Fermi processes in the tail. We propose such increasingly parallel ballooning triggers auroral substorm onset on field lines in the transition region between dipolar and tail‐like fields, consistent with observational constraints on the location of the onset arc with respect to the ion isotropy boundary in the magnetotail.

<p>The SPITFIRE fire model is used with several Dynamic Global Vegetation Models to model fire-vegetation interactions on a global scale. Since its development in 2010 it has been used for multiple studies in this field. The model consists of components that calculate ignitions, fire spread, post-fire mortality, and emissions. We find that the fire spread component of this model contains errors that introduce significant biases to its results. In particular, errors in the application of the Rothermel equation result in fires that are significantly too large and intense. Further, unphysically low live grass moistures in the model result in excessively fire-prone grasslands, and therefore a strong link between the presence of grasslands and the presence of fire. This combination results in areas where the SPITFIRE model calculates excessive tree mortality and consequent grassland formation. We perform a detailed analysis of these errors and examine the impact that corrections to them have on SPITFIRE model results.</p>

<p>Fire modelling incorporated into global dynamic vegetation models (DGVMs) allows for the projection of changes to fire-related biogeophysical and biogechemical processes under future climate scenarios, including anthropogenic climate change. Due to the large grid sizes often required to efficiently model fire and vegetation dynamics in a global manner, fire-enabled DGVMs generally neglect some finer-scale effects, including slope. However, slope can have a significant impact on the spread of individual fires and, therefore, the global area burned. As a fire moves uphill, the angle of flames is better suited to heating nearby fuel, thus increasing the rate of spread relative to fires on level ground. In this study, we apply a function to account for the impact of slope on fire spread in the SPITFIRE model incorporated into the LPJmL5.3 DGVM to improve the calculation of fire-related processes, including burnt area. We aggregate slope data across a grid cell to account for the impact of slope in a general way appropriate to the  grid size used in SPITFIRE. Our approach, while focused on the SPITFIRE model, may also be applicable to other DGVM-based fire models.</p>

Recent work by Oberhagemann and Mann (2020, https://doi.org/10.1029/2019GL085271) has shown that ballooning instabilities in the transition region between dipolar and taillike fields associated with substorm onset may be triggered by an anisotropic pressure distribution that moves from an initially mild perpendicular anisotropy toward more parallel anisotropy. Here we examine the mechanism which can result in this transition toward more parallel pressure during growth phase tail stretching. We trace particles through a two-dimensional model magnetic field to examine the evolution of particle distributions as they periodically drift through a tail which changes from a less stretched, quiet time to a more stretched late growth phase configuration. Our findings show that these particle pitch angle distributions become more parallel anisotropic through a combination of drift shell splitting and preferential de-energization in the perpendicular direction. Drift shell splitting causes low pitch angle particles from a high-pressure region to mix with high pitch angle particles from a low-pressure region in a radially localized area in the tail transition region where high earthward pressure gradients exist. Preferential de-energization occurs due to a transfer of particles from less stretched to more stretched field lines that reverses the usual balance between Fermi and betatron acceleration and causes particle pitch angle distributions to become more field aligned. Our model predicts onset characteristics and location in good agreement with observations, with the most unstable region consistent with the location of the auroral beads, which accompany the onset of the expansion phase of many substorms.

Fire-enabled Dynamic Global Vegetation Models (DGVMs) are used to model fire-vegetation interactions and their impacts on global vegetation dynamics. The challenge of modelling fires in boreal zones has been addressed by several DGVMs by allowing for longer modelled fire durations. However, DGVMs generally do not account for the impact of slope on increasing fire spread and only a few account for terrain-related landscape fragmentation. We investigate improvements to modelling boreal fires in the LPJmL-SPITFIRE DGVM by accounting for the impacts of terrain and incorporating longer fire durations. Our work is conducted in two parts: first, we investigate the impact of slope and terrain ruggedness on burnt area at the 0.5° by 0.5° resolution typical of DGVMs using satellite data, and with a particular focus on boreal regions. We demonstrate that terrain fragmentation acts as a limit on burnt area for the largest fires in a grid cell and that slope-driven wildfire spread can increase burnt area up to this limit. Therefore, these terrain effects are important for inclusion in fire modelling in general, and in boreal regions in particular. The second part of our work consists of the development of a function based on these results for implementation in LPJmL SPITFIRE and the evaluation of model improvements in boreal regions when this function is combined with longer fire durations. The results of this work represent a useful addition to LPJmL-SPITFIRE as well as DGVMs in general that do not incorporate the effects of terrain-based landscape fragmentation and slope-driven wildfire spread.