Cities are and will continue to be affected by a set of problems considered wicked by the literature (Rittel & Webber, 1973) and the perception of these problems generates pressure on managers to create solutions that meet social demands (Botton, Pinheiro, Oliveira & Jesus-Lopes, 2021), as well as shows the need for innovative solutions and transformations (Veeckman & Temmerman, 2021). The adoption of Living Labs (LL) or Urban Living Labs (ULL) are considered tools for achieving innovation (Compagnucci, Spigarelli, Coelho & Duarte, 2021; Hossain, Leminen & Westerlund, 2019), which can help achieve more sustainable development (Evans & Karvonen, 2012), at different scales and themes. It can also contribute to the change of citizens through co-creation (Sauer, 2012), and is also considered a quadruple helix platform, which involves private companies, government, higher education and society as actors (Compagnucci et al., 2021). However, studies on this topic are concentrated in Europe, presenting a gap in studies that consider the characteristics and approaches to these laboratories in developing countries, such as Brazil (Amorim, Menezes & Fernandes, 2022; McCrory, Schäpke, Holmén & Holmberg 2020). This study, in progress, aims to point out the approach to sustainability found in Brazilian Living Labs or Urban Living Labs. This is a descriptive and exploratory research, with a mixed approach, carried out in two methodological stages for data collection and analysis. The first stage aimed to identify the Brazilian Living Labs and Urban Living Labs, initially carrying out a systematic literature review to understand how these initiatives have been studied in the Brazilian context, followed by searches: in the Directory of Research Groups of the National Council for Scientific and Technological Development (DGP/CNPq) of Brazil; on websites open to the public and through the use of the SnowBall method. The results obtained were the identification of studies that relate LL or ULL with Brazil, their geographic scope in the country, thematic focuses and the creation of an agenda for future research, aiming at the advancement of the theme based on Brazilian studies. In addition, a list was drawn up that identified 44 Brazilian laboratories, which includes some preliminary results such as activity status and description of LL or ULL. The next step, which consists of confirming the activities of the listed laboratories, will be carried out through the application of a survey. The variables found in theory, that characterize, dimension, and describe LLs or ULLs (McCrory, Holmén, Schäpke & Holmberg, 2022; Chronéer, Stahlbröst & Habibipour, 2019; Stten & Bueren, 2017; Almirall, Lee & Warehan, 2012; Niitamo, Kulkiki, Eiksson & Hribernik, 2006), were taken into account for preparing the survey questionnaire. The aim is to gather information that portrays the characteristics of Brazilian Living Labs or Urban Living Labs, as well as to expand knowledge regarding the approach they adopt to sustainability. This work is expected to assist actors in decision-making for innovative and more sustainable solutions and transformations in developing countries.

In this study we examine how a wide yet plausible range of socioeconomic, policy, and climate futures affects agricultural crop yields in the continental United States (CONUS) over the 21st century. We use the Global Change Analysis Model - USA (GCAM-USA) to simulate 8 different scenarios consistent with the Shared Socioeconomic Pathways - Representative Concentration Pathways (SSP-RCP) scenario framework. We combine two different socioeconomic pathways with four high-resolution climate projections for the US: two instantiations (e.g., hotter and cooler climate) each for RCP8.5 and RCP4.5. These climate projections are the result of applying a Thermodynamic Global Warming (TGW) approach derived from models with greater or lesser climate sensitivity (i.e., sensitivity on ensemble mean of projected temperature) in the Coupled Model Intercomparison Project Phase 6 (CMIP6). GCAM-USA requires data inputs specifying basin-scale crop yield over time, for which we use the global crop yield model Osiris to simulate climate impacts on agricultural productivity under the four TGW climate scenarios. Osiris takes as inputs gridded precipitation and near-surface air temperature at 12 km resolution over the CONUS, atmospheric carbon dioxide concentration, and applied nitrogen (i.e., fertilizer) to generate gridded crop yields. In addition to agricultural yield, the GCAM-USA modeling also includes the dynamic impacts of these SSP-RCP combinations on energy demands and water availability.

In this study we focus on a subset of the results to highlight key insights pertaining to agriculture. Overall, we find that production of corn, fodder herb, fodder grass and oil crop are particularly sensitive to socioeconomics and climate uncertainty. We find that despite the high population growth in the US under the SSP5 scenario, total crop production in the US decreases due to a drop in international demand for fodder herb, driven by lower global population, compared to SSP3. Under the SSP3 scenario, fodder herb production in the US increases due to higher global demand (particularly in India, China, and Pakistan), and corn expands in the US (mainly northeastern river basins) due to increased national corn demands. The RCP8.5 climate has a positive impact on most crop types, especially fodder herb and fodder grass production in most of the US basins, compared to RCP4.5. However, the southeastern basins experience decreased corn production under a high-emission warmer climate. We will report on the results available up to June 2024. This research provides valuable insights into sustainable agricultural planning across diverse climate and socioeconomic scenarios.

Due to the great demand of plastics and deficient waste management systems around the world, a considerable amount of plastic polymers end up in marine ecosystems. The environmental impacts of marine plastic pollution are vast, specially to marine fauna. A considerable part of this waste enters the ocean through river channels, both in the form of macroplastics and microplastics. Consequently, scientific community has focused on estimating the presence of plastic in river, warning the associated hazards. There is a need to explore new methodologies to refine these estimates and reduce potential errors. At the end of the last decade, a novel approach for detecting macroplastics has been explored: the use of artificial intelligence. In this context, and based on the state-of-the-art of the field, the present research focuses on the adaptation of the YOLOv8 pre-trained model through the application of transfer learning technique for the detection of macroplastics in aquatic environments. Thus, the main objective is to estimate the amount of macroplastic types present in the study section and determine what percentage of these are transported to the ocean. Consequently, we analyze the fluctuations in the detected amount of macroplastics in each record over the course of the hydrological year. A 1.1 km stretch from the Rímac River has been selected as the case study, whose waters flow through the city of Lima and discharge into the Pacific Ocean, representing one of the main rivers of the capital of Peru. Visual recording campaigns have been established, which include images and videos from the use of a drone during the period of a hydrological year (Set/23-Aug/24). Therefore, the temporality and bimodal behavior of the river are considered for detecting macroplastic. The images collected to date (~ 2000) were used to generate a 3D model (orthophoto model) of the study section. This model facilitates the identification and monitoring areas where there is a greater density of plastic waste. Additionally, the model enables the identification of 8 types of polymer classes present along the slopes and within the course of the river. Being Yolov8 a supervised model, the previously filtered images have been processed to create labels, that will be utilized as input to the model. Initial findings demonstrate that variations in each time window (imaging campaign) are caused by the natural washing of the watercourse, especially during flood stages. Certainly, YOLOv8, which relies on convolutional neural networks, differs from other architectures in its processing speed and ability to perform detections in a single pass through an image and it is currently used in a diverse range of applications in the field of object detection. Therefore, once trained for the detection and classification of macroplastics, it has the potential to expand its coverage to a broader area and extrapolate this methodology to other watercourses, providing practical utility in real-time monitoring of plastic debris.

Using a large (36 M) database of new and used vehicles listed online at 60,000 dealerships in the U.S. between 2016 and 2022, we provide high resolution, nation-wide estimates of resale value and mileage in the United States for battery electric vehicles (BEVs), plug-in hybrid electric vehicles (PHEVs), hybrid electric vehicles (HEVs), and conventional gasoline vehicles (CVs). Previous estimates of both of these important metrics are based on limited or outdated data, or indirect measurements such as self-reported data from surveys. For resale value, we find that BEVs and PHEVs depreciate at faster rates than CVs and HEVs, though Tesla BEVs are a notable exception, depreciating slower than many CVs. Newer model year BEVs and those with larger ranges have significantly higher retention rates than older models with smaller ranges. Subsidies for new BEVs have a modest effect on resale market prices, with the $7,500 federal subsidy lowering resale prices by 3% on average. Disruptions from the COVID19 pandemic have affected affordability across all vehicles, with mean listing prices rising 37% and 39% for CVs and BEVs, respectively, from January 2020 to March 2022 in inflation-adjusted 2019 dollars. For mileage, we find that while HEVs and PHEVs are driven comparably to CVs, BEVs are driven significantly less, with BEV cars traveling on average 3,400 fewer miles per year and BEV SUVs traveling 2,900 fewer miles per year. Higher-range BEVs are driven further: 1,265 more annual miles for every additional 100 miles of driving range. We also find that the annual mileage gap between CV and BEV cars shrinks to just 660 miles in states with a $1.00 per gallon higher mean gasoline price. Our results indicate that while current BEVs are not being driven as much as conventional vehicles, they approach parity under conditions that already exist for some BEVs in some locations.