Publications

title: Publications

Assessing climate resilience in rice production: measuring the impact of the Millennium Challenge Corporation's IWRM scheme in the Senegal River Valley using remote sensing and machine learning
Assessing climate resilience in rice production: measuring the impact of the Millennium Challenge Corporation's IWRM scheme in the Senegal River Valley using remote sensing and machine learning

Satellite remote sensing and machine learning can be combined to develop methods for measuring the impacts of climate change on biomass and agricultural systems. From 2015 to 2023, we applied this approach in a critical earth observation-based evaluation of the Irrigation and Water Resources Management component of the Millennium Challenge Corporation’s Senegal Compact. This project, funded by the United States Agency for International Development (USAID), was implemented in the Senegal River Valley from 2010 to 2015. Utilising these techniques, we successfully mapped rice cultivation areas, deciphered cropping practices, and analysed irrigation systems responses to different climatic conditions. A marked increase in cultivated rice area was found particularly in regions targeted by the project intervention. This is despite prolonged drought conditions which underscores a significant climate adaptation benefit from these irrigation works. We observed a notable dip in rice cultivation area in 2020, possibly due to the COVID-19 pandemic, followed by a recovery to pre-pandemic levels in 2023, likely aided by previously funded USAID’s socio-economic resilience programmes in the region. Economic analysis of increased rice yields in the region translates to approximately US$ 61.2 million in market value since 2015, highlighting the economic returns from the project investment. Both the remote sensing data and ground audits identify issues regarding post-project deterioration of irrigation infrastructure, emphasising the need for long-term maintenance of irrigation infrastructure to support climate adaptation benefits arising from irrigation. With a focus on crop irrigation, our findings stress the critical role of climate adaptation interventions for maintaining agricultural productivity in the face of adverse climate shocks. It further highlights the necessity of continuous investment and maintenance for ensuring climate resilient agrifood systems.

Coronal Mass Ejections and Type II Radio Emission Variability during a Magnetic Cycle on the Solar-type Star ϵ Eridani

We simulate possible stellar coronal mass ejection (CME) scenarios over the magnetic cycle of ϵ Eridani (18 Eridani; HD 22049). We use three separate epochs from 2008, 2011, and 2013, and estimate the radio emission frequencies associated with these events. These stellar eruptions have proven to be elusive, although a promising approach to detect and characterize these phenomena are low-frequency radio observations of potential type II bursts as CME-induced shocks propagate through the stellar corona. Stellar type II radio bursts are expected to emit below 450 MHz, similarly to their solar counterparts. We show that the length of time these events remain above the ionospheric cutoff is not necessarily dependent on the stellar magnetic cycle, but more on the eruption location relative to the stellar magnetic field. We find that these type II bursts would remain within the frequency range of LOFAR for a maximum of 20-30 minutes post-eruption for the polar CMEs (50 minutes for second harmonics). We find evidence of slower equatorial CMEs, which result in slightly longer observable windows for the 2008 and 2013 simulations. Stellar magnetic geometry and strength have a significant effect on the detectability of these events. We place the CMEs in the context of the stellar mass-loss rate (27-48× solar mass-loss rate), showing that they can amount to 3%-50% of the stellar wind mass-loss rate for ϵ Eridani. Continuous monitoring of likely stellar CME candidates with low-frequency radio telescopes will be required to detect these transient events.

λ And: a post-main-sequence wind from a solar-mass star

We investigate the wind of λ And, a solar-mass star that has evolved off the main sequence becoming a subgiant. We present spectropolarimetric observations and use them to reconstruct the surface magnetic field of λ And. Although much older than our Sun, this star exhibits a stronger (reaching up to 83 G) large-scale magnetic field, which is dominated by the poloidal component. To investigate the wind of λ And, we use the derived magnetic map to simulate two stellar wind scenarios, namely a ‘polytropic wind’ (thermally driven) and an ‘Alfven-wave-driven wind’ with turbulent dissipation. From our 3D magnetohydrodynamics simulations, we calculate the wind thermal emission and compare it to previously published radio observations and more recent Very Large Array observations, which we present here. These observations show a basal sub-mJy quiescent flux level at ~5 GHz and, at epochs, a much larger flux density (>37 mJy), likely due to radio flares. By comparing our model results with the radio observations of λ And, we can constrain its mass-loss rate M˙ . There are two possible conclusions. (1) Assuming the quiescent radio emission originates from the stellar wind, we conclude that λ And has M˙≃3×10−9 M⊙ yr -1, which agrees with the evolving mass-loss rate trend for evolved solar-mass stars. (2) Alternatively, if the quiescent emission does not originate from the wind, our models can only place an upper limit on mass-loss rates, indicating that M˙≲3×10−9 M⊙ yr -1.

The solar wind in time–II. 3D stellar wind structure and radio emission

Simulations of the solar wind along its main-sequence lifetime and thermal radio emission computations. Can we detect solar-type winds from other stars?

MOVES – II. Tuning in to the radio environment of HD189733b

We present stellar wind modelling of the hot Jupiter host HD189733, and predict radio emission from the stellar wind and the planet, the latter arising from the interaction of the stellar wind with the planetary magnetosphere. Our stellar wind models incorporate surface stellar magnetic field maps at the epochs 2013 June/July, 2014 September, and 2015 July as boundary conditions. We find that the mass-loss rate, angular momentum loss rate, and open magnetic flux of HD189733 vary by 9 per cent, 40 per cent, and 19 per cent over these three epochs. Solving the equations of radiative transfer, we find that from 10 MHz–100 GHz the stellar wind emits fluxes in the range of 10−3–5 μJy, and becomes optically thin above 10 GHz. Our planetary radio emission model uses the radiometric Bode’s law, and neglects the presence of a planetary atmosphere. For assumed planetary magnetic fields of 1–10 G, we estimate that the planet emits at frequencies of 2–25 MHz, with peak flux densities of 102 mJy. We find that the planet orbits through regions of the stellar wind that are optically thick to the emitted frequency from the planet. As a result, unattenuated planetary radio emission can only propagate out of the system and reach the observer for 67 per cent of the orbit for a 10 G planetary field, corresponding to when the planet is approaching and leaving primary transit. We also find that the plasma frequency of the stellar wind is too high to allow propagation of the planetary radio emission below 21 MHz. This means a planetary field of at least 8 G is required to produce detectable radio emission.

The solar wind in time: a change in the behaviour of older winds?

Has the solar wind changed over its lifetime? We use X and Y to show Z.

The solar wind in time: a change in the behaviour of older winds?

Has the solar wind changed over its lifetime?