Search

The increase in the concentration of geological gas emissions in the atmosphere and particularly the increase of methane is considered by the majority of the scientific community as the main cause of global climate change. The main reasons that place methane at the center of interest, lie in its high global warming potential (GWP) and its lifetime in the atmosphere. Anthropogenic processes, like engineering geology ones, highly affect the daily profile of gasses in the atmosphere. Should direct measures be taken to reduce emissions of methane, immediate global warming mitigation could be achieved. Due to its significance, methane has been monitored by many space missions over the years and as of 2017 by the Sentinel-5P mission. Considering the above, we conclude that monitoring and predicting future methane concentration based on past data is of vital importance for the course of climate change over the next decades. To that end, we introduce a method exploiting state-of-the-art recurrent neural networks (RNNs), which have been proven particularly effective in regression problems, such as time-series forecasting. Aligned with the green artificial intelligence (AI) initiative, the paper at hand investigates the ability of different RNN architectures to predict future methane concentration in the most active regions of Texas, Pennsylvania and West Virginia, by using Sentinel-5P methane data and focusing on computational and complexity efficiency. We conduct several empirical studies and utilize the obtained results to conclude the most effective architecture for the specific use case, establishing a competitive prediction performance that reaches up to a 0.7578 mean squared error on the evaluation set. Yet, taking into consideration the overall efficiency of the investigated models, we conclude that the exploitation of RNN architectures with less number of layers and a restricted number of units, i.e., one recurrent layer with 8 neurons, is able to better compensate for competitive prediction performance, meanwhile sustaining lower computational complexity and execution time. Finally, we compare RNN models against deep neural networks along with the well-established support vector regression, clearly highlighting the supremacy of the recurrent ones, as well as discuss future extensions of the introduced work. [ABSTRACT FROM AUTHOR]

Simple Summary: Steel production emits manganese and other heavy metals into the environment, where these byproducts can lead to health problems in humans and other animals. Birds have historically been used to detect harmful chemicals in the air, and in this paper, we show that this role continues into the present day. Feathers of a small terrestrial songbird, the Tufted Titmouse, were collected from live birds captured both near (0.3 km) and far (4.0 km) from steel production facilities in western Pennsylvania, USA. Tail feathers from birds living near the facilities had significantly higher amounts of manganese embedded in their structure compared to the feathers of birds living farther away. This shows that manganese is present in the environment, is localized at industrial sites, is encountered by birds, and is sequestered in their tissues. Animals could suffer ill health if exposures are high enough. Feather data cannot be used to assess the magnitude of this health risk since metals laid down within feathers are no longer circulating in the birds' bodies. However, our data do suggest that the environmental monitoring of manganese should continue, as inadvertent exposure is still occurring. Human industry and land use has led to the anthropogenic release of manganese (Mn) into the air and soil near manufacturing centers. Overexposure to Mn can cause considerable health problems in birds. We studied whether the concentration of Mn in bird feathers correlates with the distance to point sources of Mn air emissions. Feathers were collected from Tufted Titmice (Baeolophus bicolor) at two sites in western Pennsylvania, USA. One site was in proximity (0.3 km) to a steel plant with documented Mn releases, and the other site was in a different town about 4.0 km away from other steel plants with documented Mn releases. Using the microwave plasma–atomic emission spectrometer (MP–AES), we found that tail feathers collected from nearest to a steel plant had a significantly higher concentration of Mn compared to the samples from the site further from the emission source. A body mass index was calculated for each set of birds; however, the indices did not vary significantly. This is the first published study of Mn sequestration in Tufted Titmouse feathers. This study develops our general understanding of the potential use of bird feathers as non-invasive bioindicators of environmental metal exposure. [ABSTRACT FROM AUTHOR]

Work zone safety is a high priority for transportation agencies across the United States. Enforcing speed compliance in work zones is an important factor for reducing the frequency and severity of crashes. This paper uses connected vehicle trajectory data to evaluate the impact of automated work zone speed enforcement on three work zones in Pennsylvania and two work zones in Indiana. Analysis was conducted on more than 300 million datapoints from over 71 billion records between April and August 2021. Speed distribution and speed compliance studies with and without automated enforcement were conducted along every tenth of a mile, and the results found that overall speed compliance inside the work zones increased with the presence of enforcement. In the three Pennsylvania work zones analyzed, the proportions of vehicles travelling within the allowable 11 mph tolerance were 63%, 75% and 84%. In contrast, in Indiana, a state with no automated enforcement, the proportions of vehicles travelling within the same 11 mph tolerance were found to be 25% and 50%. Shorter work zones (less than 3 miles) were associated with better compliance than longer work zones. Spatial analysis also found that speeds rebounded within 1–2 miles after leaving the enforcement location. [ABSTRACT FROM AUTHOR]

This paper describes a new method for detecting individual tree stems that was designed to perform well in the challenging hardwood-dominated, mixed-species forests common to the northeastern U.S., where canopy height-based methods have proven unreliable. Most prior research in individual tree detection has been performed in homogenous coniferous or conifer-dominated forests with limited hardwood presence. The study area in central Pennsylvania, United States, includes 17+ tree species and contains over 90% hardwoods. Existing methods have shown reduced performance as the proportion of hardwood species increases, due in large part to the crown-focused approaches they have employed. Top-down approaches are not reliable in deciduous stands due to the inherent complexity of the canopy and tree crowns in such stands. This complexity makes it difficult to segment trees and accurately predict tree stem locations based on detected crown segments. The proposed voxel column-based approach has advantages over both traditional canopy height model-based methods and computationally demanding point-based solutions. The method was tested on 1125 reference trees, ≥10 cm diameter at breast height (DBH), and it detected 68% of all reference trees and 87% of medium and large (sawtimber-sized) trees ≥28 cm DBH. Significantly, the commission rate (false predictions) was negligible as most raw false positives were confirmed in follow-up field visits to be either small trees below the threshold for recording or trees that were otherwise missed during the initial ground survey. Minimizing false positives was a priority in tuning the method. Follow-up in-situ evaluation of individual omission and commission instances was facilitated by the high spatial accuracy of predicted tree locations generated by the method. The mean and maximum predicted-to-reference tree distances were 0.59 m and 2.99 m, respectively, with over 80% of matches within <1 m. A new tree-matching method utilizing linear integer programming is presented that enables rigorous, repeatable matching of predicted and reference trees and performance evaluation. Results indicate this new tree detection method has potential to be operationalized for both traditional forest management activities and in providing the more frequent and scalable inventories required by a growing forest carbon offsets industry. [ABSTRACT FROM AUTHOR]