A machine learning algorithm to explore the drivers of carbon emissions in Chinese cities

Latest Results

Abstract

As the world’s largest energy consumer and carbon emitter, the task of carbon emission reduction is imminent. In order to realize the dual-carbon goal at an early date, it is necessary to study the key factors affecting China’s carbon emissions and their non-linear relationships. This paper compares the performance of six machine learning algorithms to that of traditional econometric models in predicting carbon emissions in China from 2011 to 2020 using panel data from 254 cities in China. Specifically, it analyzes the comparative importance of domestic economic, external economic, and policy uncertainty factors as well as the nonparametric relationship between these factors and carbon emissions based on the Extra-trees model. Results show that energy consumption (ENC) remains the root cause of increased carbon emissions among domestic economic factors, although government intervention (GOV) and digital finance (DIG) can significantly reduce it. Next, among the external economic and policy uncertainty factors, foreign direct investment (FDI) and economic policy uncertainty (EPU) are important factors influencing carbon emissions, and the partial dependence plots (PDPs) confirm the pollution haven hypothesis and also reveal the role of EPU in reducing carbon emissions. The heterogeneity of factors affecting carbon emissions is also analyzed under different city sizes, and it is found that ENC is a common driving factor in cities of different sizes, but there are some differences. Finally, appropriate policy recommendations are proposed by us to help China move rapidly towards a green and sustainable development path.

Databases and Applications of the Soil and Water Assessment Tool (SWAT) Model in Brazilian River Basins: a Review

Latest Results

Abstract

Hydrological models are used to assess natural and man-made changes in watersheds worldwide. Proper input data collection and handling are essential to reduce simulation uncertainty. Thus, this study reviews the sources of physical and hydroclimatic data, used in the last 5 years, from 55 articles that applied the SWAT model in Brazil. Most studies took place in the Atlantic Forest biome (20), followed by Cerrado (14), Amazon (11), and Caatinga (10). Worth noting that there are no studies published in the Pantanal and Pampa biomes. National databases (INPE, INMET, EMPRAPA, and ANA) are the most used in data acquisition process, followed by regional databases, more applied in smaller basins. Global databases are more sought after in studies of large basins due to their low spatial resolution. National climate data have low spatial density and are only available in five states at the regional level, so satellite data and reanalysis are viable alternatives in regions with little climate monitoring. Future research directions include (1) evaluating and comparing available data, (2) using high-resolution imagery to map land use in small catchments, (3) expanding the model’s database of vegetation parameters to cover all classes identified in high-resolution images, (4) create a database at regional level in the states, (5) develop software to manage hydroclimatic information, and (6) continuously monitor the quality of water bodies.

The Global Dam Watch database of river barrier and reservoir information for large-scale applications

Latest Results

Abstract

There are millions of river barriers worldwide, ranging from wooden locks to concrete dams, many of which form associated impoundments to store water in small ponds or large reservoirs. Besides their benefits, there is growing recognition of important environmental and social trade-offs related to these artificial structures. However, global datasets describing their characteristics and geographical distribution are often biased towards particular regions or specific applications, such as hydropower dams affecting fish migration, and are thus not globally consistent. Here, we present a new river barrier and reservoir database developed by the Global Dam Watch (GDW) consortium that integrates, harmonizes, and augments existing global datasets to support large-scale analyses. Data curation involved extensive quality control processes to create a single, globally consistent data repository of instream barriers and reservoirs that are co-registered to a digital river network. Version 1.0 of the GDW database contains 41,145 barrier locations and 35,295 associated reservoir polygons representing a cumulative storage capacity of 7,420 km3 and an artificial terrestrial surface water area of 304,600 km2.

Three Recent Books on Social Media Manipulation, Misunderstanding Science, and Caring about Future People: Opportunities for Behavior Analysts

Latest Results

Abstract

This article reviews three recent books that introduce challenging insights to and suggest critical implications for how behavioral science can address the broad, vexing social issues that challenge human well-being. The Chaos Machine (Fisher, 2022), How the World Really Works (Smil, 2022), and What We Owe the Future (MacAskill, 2022) offer provocative discussions of increasingly important social issues—the impact of social media, the consequences of misunderstanding science, and overcoming the delay discounting that will affect future generations, respectively—that can inform behavior scientists’ thinking about improving the future while presenting a multitude of opportunities to advance our science.

Simulating runoff changes and evaluating under climate change using CMIP6 data and the optimal SWAT model: a case study

Latest Results

Abstract

This study examines the influence of climate change on hydrological processes, particularly runoff, and how it affects managing water resources and ecosystem sustainability. It uses CMIP6 data to analyze changes in runoff patterns under different Shared Socioeconomic Pathways (SSP). This study also uses a Deep belief network (DBN) and a Modified Sparrow Search Optimizer (MSSO) to enhance the runoff forecasting capabilities of the SWAT model. DBN can learn complex patterns in the data and improve the accuracy of runoff forecasting. The meta-heuristic algorithm optimizes the models through iterative search processes and finds the optimal parameter configuration in the SWAT model. The Optimal SWAT Model accurately predicts runoff patterns, with high precision in capturing variability, a strong connection between projected and actual data, and minimal inaccuracy in its predictions, as indicated by an ENS score of 0.7152 and an R2 coefficient of determination of 0.8012. The outcomes of the forecasts illustrated that the runoff will decrease in the coming years, which could threaten the water source. Therefore, managers should manage water resources with awareness of these conditions.

Advancing irrigation management: integrating technology and sustainability to address global food security

Latest Results

Abstract

Irrigation management is essential for addressing global food security challenges under changing climate. This review discusses the integration of advanced irrigation technologies and their roles in enhancing water use efficiency and managing energy demands within agricultural systems. High-efficiency irrigation systems, such as drip and sprinkler systems, have significant potential to reduce water use and increase crop yields. However, their adoption varies worldwide, and the efficiency of existing irrigation practices often remains inadequate, resulting in substantial water losses due to outdated management practices. Emerging technologies and innovative irrigation strategies, including precision agriculture and advanced crop models, provide promising pathways for improving irrigation efficiency. Nonetheless, the widespread integration of these technologies is hindered by high costs, the need for technical expertise, and challenges in adapting existing agricultural systems to new methodologies. Irrigation systems can have substantial energy requirements, particularly those dependent on groundwater. The exploration of the water-environment-energy-food (WEEF) nexus illustrates the importance of a balanced approach to resource management, which is crucial for achieving sustainable agricultural outcomes. Future research should include lowering barriers to technology adoption, enhancing data utilization for precision irrigation, promoting integrated management strategies within the WEEF framework, and strengthening policy support for sustainable practices. This review proposes a multidisciplinary approach to irrigation management that includes technological innovation, strategic policy development, and global cooperation to secure sustainable agricultural practices and ensure global food supply resilience in the face of climate change.

“Deepfakes and Dishonesty”

Latest Results

Abstract

Deepfakes raise various concerns: risks of political destabilization, depictions of persons without consent and causing them harms, erosion of trust in video and audio as reliable sources of evidence, and more. These concerns have been the focus of recent work in the philosophical literature on deepfakes. However, there has been almost no sustained philosophical analysis of deepfakes from the perspective of concerns about honesty and dishonesty. That deepfakes are potentially deceptive is unsurprising and has been noted. But under what conditions does the use of deepfakes fail to be honest? And which human agents, involved in one way or another in a deepfake, fail to be honest, and in what ways? If we are to understand better the morality of deepfakes, these questions need answering. Our first goal in this paper, therefore, is to offer an analysis of paradigmatic cases of deepfakes in light of the philosophy of honesty. While it is clear that many deepfakes are morally problematic, there has been a rising counter-chorus claiming that deepfakes are not essentially morally bad, since there might be uses of deepfakes that are not morally wrong, or even that are morally salutary, for instance, in education, entertainment, activism, and other areas. However, while there are reasons to think that deepfakes can supply or support moral goods, it is nevertheless possible that even these uses of deepfakes are dishonest. Our second goal in this paper, therefore, is to apply our analysis of deepfakes and honesty to the sorts of deepfakes hoped to be morally good or at least neutral. We conclude that, perhaps surprisingly, in many of these cases the use of deepfakes will be dishonest in some respects. Of course, there will be cases of deepfakes for which verdicts about honesty and moral permissibility do not line up. While we will sometimes suggest reasons why moral permissibility verdicts might diverge from honesty verdicts, we will not aim to settle matters of moral permissibility.