Significant gender differences are 1 observed on primary school students' perception of self-efficacy 2 and test anxiety in mathematics. Girls perceive themselves to be 3 significantly worse than boys in mathematics and report higher 4 test anxiety toward mathematics exams. Gender differences in 5 self-efficacy become more pronounced as students grow up, and 6 test anxiety increases for all students. However, the present study 7 shows that teachers' do not perceive differences in self-efficacy 8 in mathematics between boys and girls. 9 Background: The low presence of women in science, technol-10 ogy, engineering, and mathematics (STEM) might be explained 11 by the attitude of young students toward mathematics. Different 12 studies show that girls are less interested in STEM areas than 13 boys during secondary school. A study on the reasons for this 14 fact pointed out that the early years of education can provide 15 a relevant insight to reverse the situation. 16 Research Questions: Is there any age-dependent gender differ-17 ence in primary school students in aspects related to mathemat-18 ics? Are teachers aware of students' perceptions? 19 Methodology: This work presents a study of over 2000 pri-20 mary school students (6-12 years old) and 200 teachers in 21 Aragón (Spain). The study consists of a survey on aspects that 22 influence the experience of female and male students with math-23 ematics and Spanish language for comparison purposes and 24 teacher's awareness of students' perception. 25 Findings: The present study shows that during primary school, 26 girls are more likely to experiment a negative attitude toward 27 mathematics than boys as they grow up, and teachers may not 28 perceive girls' situation.

Abstract. Lagrangian particle dispersion models (LPDMs) have been used extensively to calculate source-receptor relationships (“footprints”) for use in applications such as greenhouse gas (GHG) flux inversions. Because a single model simulation is required for each data point, LPDMs do not scale well to applications with large data sets such as flux inversions using satellite observations. Here, we develop a proof-of-concept machine learning emulator for LPDM footprints over a ~350 km by 230 km region around an observation point, and test it for a range of in situ measurement sites from around the world. As opposed to previous approaches to footprint approximation, it does not require the interpolation or smoothing of footprints produced by the LPDM. Instead, the footprint is emulated entirely from meteorological inputs. This is achieved by independently emulating the footprint magnitude at each grid cell in the domain using gradient-boosted regression trees (GBRTs) with a selection of meteorological variables as inputs. The emulator is trained based on footprints from the UK Met Office Numerical Atmospheric dispersion Modelling Environment (NAME) for 2014 and 2015, and the emulated footprints are evaluated against hourly NAME output from 2016 and 2020. When compared to CH4 concentration time series generated by NAME, we show that our emulator achieves a mean R-squared score of 0.69 across all sites investigated between 2016 and 2020. The emulator can predict a footprint in around 10 ms, compared to around 10 minutes for the 3D simulator. This simple and interpretable proof-of-concept emulator demonstrates the potential of machine learning for LPDM emulation.

No abstract

No abstract

Abstract. Lagrangian particle dispersion models (LPDMs) have been used extensively to calculate source-receptor relationships (“footprints”) for use in applications such as greenhouse gas (GHG) flux inversions. Because a single model simulation is required for each data point, LPDMs do not scale well to applications with large data sets such as flux inversions using satellite observations. Here, we develop a proof-of-concept machine learning emulator for LPDM footprints over a ∼ 350 km × 230 km region around an observation point, and test it for a range of in situ measurement sites from around the world. As opposed to previous approaches to footprint approximation, it does not require the interpolation or smoothing of footprints produced by the LPDM. Instead, the footprint is emulated entirely from meteorological inputs. This is achieved by independently emulating the footprint magnitude at each grid cell in the domain using gradient-boosted regression trees with a selection of meteorological variables as inputs. The emulator is trained based on footprints from the UK Met Office's Numerical Atmospheric-dispersion Modelling Environment (NAME) for 2014 and 2015, and the emulated footprints are evaluated against hourly NAME output from 2016 and 2020. When compared to CH4 concentration time series generated by NAME, we show that our emulator achieves a mean R-squared score of 0.69 across all sites investigated between 2016 and 2020. The emulator can predict a footprint in around 10 ms, compared to around 10 min for the 3D simulator. This simple and interpretable proof-of-concept emulator demonstrates the potential of machine learning for LPDM emulation.