Advanced

Present and future precipitation variations in the source region of the Yangtze River, China

Du, Yiheng LU (2021)
Abstract
The source region of the Yangtze River (SRYR), the origin of the longest river in China, is an area with high-mountains and river ecotones that face considerable challenges under climate change. Precipitation is fundamental for sustainable ecosystems in this area and for the downstream water supplies. Thus, this study investigated the present and future patterns of precipitation variation in the SRYR.
To investigate historical climate characteristics of the SRYR, analysis of hydro-climatic components during 1957-2013 was performed. Temperature in the SRYR increased at a rate of 0.34°C/decade, precipitation and evaporation increased by 11.4 and 7.6 mm/decade, respectively. Runoff depth increased by 3.3 mm/decade. Considering the water... (More)
The source region of the Yangtze River (SRYR), the origin of the longest river in China, is an area with high-mountains and river ecotones that face considerable challenges under climate change. Precipitation is fundamental for sustainable ecosystems in this area and for the downstream water supplies. Thus, this study investigated the present and future patterns of precipitation variation in the SRYR.
To investigate historical climate characteristics of the SRYR, analysis of hydro-climatic components during 1957-2013 was performed. Temperature in the SRYR increased at a rate of 0.34°C/decade, precipitation and evaporation increased by 11.4 and 7.6 mm/decade, respectively. Runoff depth increased by 3.3 mm/decade. Considering the water balance, annual water storage was constant despite a continuous small negative trend. Increase in precipitation is mainly caused by increasing evapotranspiration, leading to the relatively stable water storage during the study period, which also suggests an accelerating water cycle in the SRYR. This knowledge is essential for the understanding of water resources conditions in the area.
Rainy season precipitation (June-August) in the SRYR accounts for approximately 70% of the annual total, and its anomalies are essential for ecosystem resilience. Hence, analysis of rainy season precipitation variability in relation to sea surface temperature (SST) anomalies as well as large-scale circulations including El Niño Southern Oscillation (ENSO) and Pacific Decadal Oscillation (PDO) was conducted. Results indicate that the leading modes of rainy season precipitation variability can be explained by the variability of SST together with an integrated effect of ENSO and PDO. The influence of ENSO and PDO will enhance/decrease when they are in-/out-of-phase, respectively. Negative PDO induces more precipitation in La Niña years than in El Niño years for the SRYR, especially over central and eastern parts of the basin. Positive PDO induces precipitation decrease, and El Niño enhances the decrease. The mechanism behind this pattern is atmospheric circulation affecting the strength of westerlies that transport moisture to the inland areas and as well local convergence conditions. Results have implications for predicting the rainy season precipitation for coming decades over the study area. If the current negative PDO phase continues together with more frequent extreme La Niña events, as suggested in other research, more precipitation during rainy season is expected over the SRYR.
To further quantify precipitation variability, a multi-space model for seasonal precipitation prediction was developed using principal component analysis (PCA) and artificial neural network (ANN). Correlation analysis shows that the most important climate indices for precipitation in the SRYR vary depending on the season and spatial location. The North Atlantic Oscillation (NAO), Polar/Eurasia Pattern (POL), Southern Oscillation Index (SOI), and Scandinavia Pattern (SCA) events have influence on precipitation in the SRYR during the cold season, while NAO, PDO, and SOI are more important for the warm season. A spatiotemporal model for predicting grid precipitation using significant correlated indices was established for each season, the PCA-ANN model. Results show that the PCA-ANN model can predict precipitation in the study area. By reconstructing principal components, the model provides a simulated dataset with the same size as the original dataset. The PCA-ANN model performs well in terms of both temporal variability and spatial distribution following the rank summer> winter> spring> autumn. A small basin with many variables/grids is recommended for the PCA-ANN model.
To access future precipitation pattern, historical performance, and future projections of monthly precipitation in the SRYR were investigated, using the National Aeronautics Space Administration (NASA) Earth Exchange Global Daily Downscaled Projections (NEX-GDDP) dataset. Performance of the 21 models were compared against in situ observations for the historical period 1961–2005, therefore rankings were listed according to their performance. Projected future changes in precipitation were assessed under the Representative Concentration Pathways (RCPs) 4.5 and 8.5 emission scenarios, for near-future (2041-2060) and far-future (2081–2100) time slices with respect to 1986–2005. The results show that models derived from NEX-GDDP data effectively produce observed precipitation magnitude in the study area and optimum models were selected based on comprehensive ranking index. The future climate projections indicate a consistent rise in mean precipitation, especially in summer. The average annual precipitation during the near-future and the far-future showed an increase of 18.6% and 24.4% under RCP 4.5, and a larger increase under RCP 8.5 of 22.5% and 49.7%. The summer precipitation shows similar increase as the annual precipitation but with a slightly larger amplitude.
The findings in this thesis provide insights to improve the understanding of water resources variations under the background of climate change, and to establish sustainable management of water resources. The precipitation variability in relation to large-scale circulation can help to improve weather forecasting at a low-cost level. Besides, identification of physical mechanisms of integrated impacts from two major circulation patterns can improve the understanding of drivers behind precipitation variability. Future projections provide guidance for future adaptive solutions, including both spatial and temporal changes. With such information, adaptive plans for the study area can be set up with higher accuracy, lower budget, and localized suitability. (Less)
Please use this url to cite or link to this publication:
author
supervisor
opponent
  • Prof. Chen, Deliang, University of Gothenburg, Sweden.
organization
publishing date
type
Thesis
publication status
published
subject
keywords
Precipitation, water balance, ENSO, PDO, Yangtze River, Tibet Plateau, future projections
pages
147 pages
publisher
Water Resources Engineering, Lund University
defense location
Lecture hall V:C, building V, John Ericssons väg 1, Faculty of Engineering LTH, Lund University, Lund.
defense date
2021-03-02 10:15:00
ISBN
978-91-7895-739-2
978-91-7895-740-8
project
Present and future precipitation variations in the source region of the Yangtze River, China
language
English
LU publication?
yes
id
1dec9d4e-e626-42f5-8c8a-ea6651fc54b4
date added to LUP
2021-01-26 14:10:10
date last changed
2021-02-03 10:57:32
@phdthesis{1dec9d4e-e626-42f5-8c8a-ea6651fc54b4,
  abstract     = {The source region of the Yangtze River (SRYR), the origin of the longest river in China, is an area with high-mountains and river ecotones that face considerable challenges under climate change. Precipitation is fundamental for sustainable ecosystems in this area and for the downstream water supplies. Thus, this study investigated the present and future patterns of precipitation variation in the SRYR.<br>
To investigate historical climate characteristics of the SRYR, analysis of hydro-climatic components during 1957-2013 was performed. Temperature in the SRYR increased at a rate of 0.34°C/decade, precipitation and evaporation increased by 11.4 and 7.6 mm/decade, respectively. Runoff depth increased by 3.3 mm/decade. Considering the water balance, annual water storage was constant despite a continuous small negative trend. Increase in precipitation is mainly caused by increasing evapotranspiration, leading to the relatively stable water storage during the study period, which also suggests an accelerating water cycle in the SRYR. This knowledge is essential for the understanding of water resources conditions in the area. <br>
Rainy season precipitation (June-August) in the SRYR accounts for approximately 70% of the annual total, and its anomalies are essential for ecosystem resilience. Hence, analysis of rainy season precipitation variability in relation to sea surface temperature (SST) anomalies as well as large-scale circulations including El Niño Southern Oscillation (ENSO) and Pacific Decadal Oscillation (PDO) was conducted. Results indicate that the leading modes of rainy season precipitation variability can be explained by the variability of SST together with an integrated effect of ENSO and PDO. The influence of ENSO and PDO will enhance/decrease when they are in-/out-of-phase, respectively. Negative PDO induces more precipitation in La Niña years than in El Niño years for the SRYR, especially over central and eastern parts of the basin. Positive PDO induces precipitation decrease, and El Niño enhances the decrease. The mechanism behind this pattern is atmospheric circulation affecting the strength of westerlies that transport moisture to the inland areas and as well local convergence conditions. Results have implications for predicting the rainy season precipitation for coming decades over the study area. If the current negative PDO phase continues together with more frequent extreme La Niña events, as suggested in other research, more precipitation during rainy season is expected over the SRYR. <br>
To further quantify precipitation variability, a multi-space model for seasonal precipitation prediction was developed using principal component analysis (PCA) and artificial neural network (ANN). Correlation analysis shows that the most important climate indices for precipitation in the SRYR vary depending on the season and spatial location. The North Atlantic Oscillation (NAO), Polar/Eurasia Pattern (POL), Southern Oscillation Index (SOI), and Scandinavia Pattern (SCA) events have influence on precipitation in the SRYR during the cold season, while NAO, PDO, and SOI are more important for the warm season. A spatiotemporal model for predicting grid precipitation using significant correlated indices was established for each season, the PCA-ANN model. Results show that the PCA-ANN model can predict precipitation in the study area. By reconstructing principal components, the model provides a simulated dataset with the same size as the original dataset. The PCA-ANN model performs well in terms of both temporal variability and spatial distribution following the rank summer&gt; winter&gt; spring&gt; autumn. A small basin with many variables/grids is recommended for the PCA-ANN model. <br>
To access future precipitation pattern, historical performance, and future projections of monthly precipitation in the SRYR were investigated, using the National Aeronautics Space Administration (NASA) Earth Exchange Global Daily Downscaled Projections (NEX-GDDP) dataset. Performance of the 21 models were compared against in situ observations for the historical period 1961–2005, therefore rankings were listed according to their performance. Projected future changes in precipitation were assessed under the Representative Concentration Pathways (RCPs) 4.5 and 8.5 emission scenarios, for near-future (2041-2060) and far-future (2081–2100) time slices with respect to 1986–2005. The results show that models derived from NEX-GDDP data effectively produce observed precipitation magnitude in the study area and optimum models were selected based on comprehensive ranking index. The future climate projections indicate a consistent rise in mean precipitation, especially in summer. The average annual precipitation during the near-future and the far-future showed an increase of 18.6% and 24.4% under RCP 4.5, and a larger increase under RCP 8.5 of 22.5% and 49.7%. The summer precipitation shows similar increase as the annual precipitation but with a slightly larger amplitude. <br>
The findings in this thesis provide insights to improve the understanding of water resources variations under the background of climate change, and to establish sustainable management of water resources. The precipitation variability in relation to large-scale circulation can help to improve weather forecasting at a low-cost level. Besides, identification of physical mechanisms of integrated impacts from two major circulation patterns can improve the understanding of drivers behind precipitation variability. Future projections provide guidance for future adaptive solutions, including both spatial and temporal changes. With such information, adaptive plans for the study area can be set up with higher accuracy, lower budget, and localized suitability.},
  author       = {Du, Yiheng},
  isbn         = {978-91-7895-739-2},
  language     = {eng},
  publisher    = {Water Resources Engineering, Lund University},
  school       = {Lund University},
  title        = {Present and future precipitation variations in the source region of the Yangtze River, China},
  url          = {https://lup.lub.lu.se/search/ws/files/90496380/Yiheng_Du_web.pdf},
  year         = {2021},
}