Analysis of future climate variability under CMIP6 scenarios based on a downscaling method considering wet days in the upper Yangtze River basin, China

Theoretical and Applied Climatology, Journal Year: 2025, Volume and Issue: 156(2)

Published: Jan. 13, 2025

Abstract According to recent studies, the past decade was hottest on record, and climate change is accelerating. As part of Yangtze River Basin, largest river basin in China, Upper Basin (UYRB) plays a crucial role as primary source hydropower. However, UYRB also one most climate-sensitive regions within basin, making impact this area particularly critical. We downscaled CMIP6 GCMs’ outputs precipitation (including wet/dry spells sequence correction), temperature projections (2024–2100), under four typical Shared Socioeconomic Pathways (SSPs), we pursued trend analysis upon these potential future series. found significant upward trends across all SSPs August, but no for same month. Additionally, SSP370 SSP585, there are December, while showed during that This may result drier winters than now, increased evapotranspiration, reduced surface (snow) water storage, impacting resources availability. Consecutive dry/wet days at station, scale show spatial-temporal heterogeneity, generally wet longer, dry shorten moving from South-East North-West.

Language: Английский

---

Building the resilient food waste supply chain for the megacity: Based on the Multi-scale Progressive Fusion framework

Resources Conservation and Recycling, Journal Year: 2025, Volume and Issue: 215, P. 108144 - 108144

Published: Jan. 24, 2025

Language: Английский

---

Future Soil Erosion Risk in China: Differences in Erosion Driven by General and Extreme Precipitation Under Climate Change

Earth s Future, Journal Year: 2025, Volume and Issue: 13(3)

Published: March 1, 2025

Abstract Soil erosion status is a comprehensive indicator reflecting the quality and stability of ecosystems. changes in China are becoming more unclear due to climate change intensified human activity. Within framework change, this study treats rainfall factor as dynamic examines three types contrasting precipitation—general, heavy, extreme—through integrates Revised Universal Loss Equation Geographic Information Systems reveal differences water driven by varying intensities precipitation. The results that over 63% China's land area has experienced soil during historical period (1980–2022), with slight being most common. Severe predominantly found Southwest Basin, Yangtze River Yellow basin. multi‐year average rate estimated at 2.46 t·ha −1 yr , R95P R99P contributing 26.50% 7.71%, respectively. Future projections (2023–2100) indicate PRCPTOT, R95P, could increase 22%–91% under SSP5‐RCP8.5 SSP2‐RCP4.5 scenarios. Overall, limited effect on spatial pattern China, mainly influencing intensity extent adversely impacting regions. Extreme precipitation sensitive making future risks associated it critical concern. These findings can guide decision‐makers resource managers regional planning enhance resilience secure food resources.

Language: Английский

---

Measuring walking convenience of Hong Kong resource recycling services: Utilizing a spatially refined model framework of multiple recyclable wastes

Resources Conservation and Recycling, Journal Year: 2025, Volume and Issue: 215, P. 108153 - 108153

Published: Feb. 5, 2025

Language: Английский

---

Combined effects of urban morphology on land surface temperature and PM2.5 concentration across fine-scale urban blocks in Hangzhou, China

Building and Environment, Journal Year: 2025, Volume and Issue: unknown, P. 112979 - 112979

Published: April 1, 2025

Language: Английский

---

Neural Network Based on Dynamic Collaboration of Flows for Temporal Downscaling

Remote Sensing, Journal Year: 2025, Volume and Issue: 17(8), P. 1434 - 1434

Published: April 17, 2025

Time downscaling is one of the most challenging topics in remote sensing and meteorological data processing. Traditional methods often face problems high computing cost poor generalization ability. The framework interpolation method based on deep learning provides a new idea for time data. A neural network multivariate designed this paper. It estimates kernel weight offset vector each target pixel independently among different variables generates output frames guided by feature space. Compared with other methods, model can deal large range complex movements. 2 h interval experiments m temperature, surface pressure, 1000 hPa specific humidity show that MAE proposed reduced about 14%, 25%, 18%, respectively, compared advanced such as AdaCof Zooming Slow-Mo. Performance fluctuates very little over time, an average performance fluctuation only 1% across all metrics. Even experiment 6 interval, still maintains leading advantage, which indicates has not good robustness but also excellent scalability transferability task field.

Language: Английский

---

Absolute Environmental Sustainability Assessment of Emerging Working Fluids in Organic Rankine Cycles

ACS Sustainable Chemistry & Engineering, Journal Year: 2025, Volume and Issue: unknown

Published: April 29, 2025

Language: Английский

---

Circular bioeconomy: a review of empirical practices across implementation scales

Journal of Cleaner Production, Journal Year: 2024, Volume and Issue: unknown, P. 143816 - 143816

Published: Oct. 1, 2024

Language: Английский

---

The carbon footprints of consumption of goods and services in Sweden at municipal and postcode level and policy interventions

Sustainable Production and Consumption, Journal Year: 2024, Volume and Issue: 52, P. 63 - 79

Published: Oct. 21, 2024

Language: Английский

---

Extreme Gradient Boosting pada Peramalan Pola Curah Hujan Bulanan Kabupaten Banyuwangi

JURNAL KRIDATAMA SAINS DAN TEKNOLOGI, Journal Year: 2024, Volume and Issue: 6(02), P. 430 - 440

Published: July 30, 2024

Data meteorologi jangka panjang sangat berguna untuk mengidentifikasi tanda-tanda fenomena perubahan iklim. Fenomena tersebut mengacu pada kondisi fisik atmosfer bumi seperti suhu dan pola cuaca. Hal berdampat besar, terutama di Banyuwangi yang merupakan salah satu wilayah produksi beras terbesar Jawa Timur. Memprediksi tren curah hujan bulanan penting mengantisipasi kegagalan panen akibat cuaca ekstrem bencana alam banjir tanah longsor. Penelitian ini menggunakan parameter skala global suhu, hujan, penguapan, kelembaban permukaan tekanan laut, sedangkan informasi lokal data tahun 2011 hingga 2023. Metode Extreme Gradient Boosting (XGBoost) akan digunakan memprediksi dalam model ensemble learning berbasis pendekatan boosting. Secara khusus, studi menekankan kemampuannya membangun prediktif deret waktu terbatas dampak pemisahan terhadap performa model. Hasil terbaik ditunjukkan oleh dengan rasio 1:12 atau mencakup 80% sebagai pelatihan. Akurasi mencapai MAE sebesar 72.579 mm pelatihan 80.777 pengujian. Selain itu RMSE 95.940 95.775 penelitian diharapkan dapat menjadi acuan peramalan lebih optimal.

---