scholarly journals Surface Water Temperature Predictions at a Mid-Latitude Reservoir under Long-Term Climate Change Impacts Using a Deep Neural Network Coupled with a Transfer Learning Approach

Water ◽  
2021 ◽  
Vol 13 (8) ◽  
pp. 1109
Author(s):  
Nobuaki Kimura ◽  
Kei Ishida ◽  
Daichi Baba

Long-term climate change may strongly affect the aquatic environment in mid-latitude water resources. In particular, it can be demonstrated that temporal variations in surface water temperature in a reservoir have strong responses to air temperature. We adopted deep neural networks (DNNs) to understand the long-term relationships between air temperature and surface water temperature, because DNNs can easily deal with nonlinear data, including uncertainties, that are obtained in complicated climate and aquatic systems. In general, DNNs cannot appropriately predict unexperienced data (i.e., out-of-range training data), such as future water temperature. To improve this limitation, our idea is to introduce a transfer learning (TL) approach. The observed data were used to train a DNN-based model. Continuous data (i.e., air temperature) ranging over 150 years to pre-training to climate change, which were obtained from climate models and include a downscaling model, were used to predict past and future surface water temperatures in the reservoir. The results showed that the DNN-based model with the TL approach was able to approximately predict based on the difference between past and future air temperatures. The model suggested that the occurrences in the highest water temperature increased, and the occurrences in the lowest water temperature decreased in the future predictions.

2017 ◽  
Vol 18 (2) ◽  
pp. 418-429 ◽  
Author(s):  
Yang Li ◽  
Ting-lin Huang ◽  
Zi-zhen Zhou ◽  
Sheng-hai Long ◽  
Hai-han Zhang

Abstract Thermal stratification has a significant impact on water quality and ecological characteristics. Reservoir operation and climate change have an effect on the thermal regime. The Jinpen Reservoir is a large canyon-shaped reservoir located in Shaanxi Province with a strong thermal stratification, which resulted in an anaerobic condition in the hypolimnion. We used a hydrodynamic module based on MIKE 3 to simulate the thermal structure of the Jinpen Reservoir and study the relationship between the thermal regime, reservoir operation and climate change. Based on the daily hydrological and climatic data from 2004 to 2013, we made 13 hypothetical simulated conditions that included extreme change of inflow volume, water level, air temperature, radiation, inflow water temperature and selective withdrawal to explore the effect of different factors on the thermal regime. The results showed that the period of thermal stratification, water column stability and surface water temperature were influenced by these factors. With the increase of air temperature, the simulation results indicated a stronger thermal stratification and a higher surface water temperature, which could cause water safety problems. Deep withdrawal could decrease water column stability and prompt water column mixing early, which could be used by reservoir managers to optimize the reservoir operation.


Author(s):  
Katherine Eddings ◽  
Durga D Poudel ◽  
Timothy W. Duex ◽  
Robert Miller ◽  
J. Calvin Berry

Climate change impacts on rising temperatures, changes on rainfall patterns, drought, flooding, sea level rise, glacier melts, and incidence of diseases and parasites are reported worldwide in recent decades. This study investigates the effects of changing climatic conditions – particularly air temperature and precipitation, on surface water temperatures and other water quality parameters, such as the conductivity, dissolved oxygen (DO), pH, and turbidity. A statistical analysis was performed on air temperature and precipitation data from 1980 to 2005 to determine the changing climatic conditions. The water quality data for four waterbodies in southwestern Louisiana was also analyzed to examine trends between the air temperature and surface water temperatures, precipitation and surface water temperatures, and precipitation and water quality parameters. There was an unexpected increase in surface water temperature with an increase in precipitation. As the precipitation and air temperature increased, so did the surface water temperature. This increase in surface water temperature was correlated with decrease in DO levels. The increase in precipitation also correlated with an increase in pH and turbidity in Bayou Plaquemine Brule. This study’s findings could be utilized in a dynamic climate modeling system to provide more accurate predictions of climate change in southwestern Louisiana.


Limnologica ◽  
2020 ◽  
Vol 82 ◽  
pp. 125777 ◽  
Author(s):  
Burak Öğlü ◽  
Tõnu Möls ◽  
Tanel Kaart ◽  
Fabien Cremona ◽  
Külli Kangur

2012 ◽  
Vol 25 (17) ◽  
pp. 5807-5816 ◽  
Author(s):  
Anning Huang ◽  
Yerubandi R. Rao ◽  
Weitao Zhang

Abstract The surface air and water temperatures increased at all seasonal and annual time scales during the last 40 yr in Lake Ontario. The annual mean air and surface water temperatures have increased by 1.43° ±0.39° and 1.26° ±0.32°C, respectively, over 1970–2009. The air temperature increased at a faster rate than the surface water temperature in winter and autumn, whereas in spring and summer the surface water temperature warmed faster than the air temperature. The length of summer stratified season has increased by 12 ± 2 days since the early 1970s due to the increase in water temperature. The decline of surface wind speed over Lake Ontario resulted in a shallower surface mixed layer and enhanced the summer thermal stratification, which increased the summer surface water temperature more rapidly than the air temperature.


Biology ◽  
2020 ◽  
Vol 9 (1) ◽  
pp. 19 ◽  
Author(s):  
Tatiana A. Rynearson ◽  
Sarah A. Flickinger ◽  
Diana N. Fontaine

Diatoms generate nearly half of marine primary production and are comprised of a diverse array of species that are often morphologically cryptic or difficult to identify using light microscopy. Here, species composition and realized thermal niches of species in the diatom genus Thalassiosira were examined at the site of the Narragansett Bay (NBay) Long-Term Plankton Time Series using a combination of light microscopy (LM), high-throughput sequencing (HTS) of the 18S rDNA V4 region and historical records. Thalassiosira species were identified over 6 years using a combination of LM and DNA sequences. Sixteen Thalassiosira taxa were identified using HTS: nine were newly identified in NBay. Several newly identified species have small cell diameters and are difficult to identify using LM. However, they appeared frequently and thus may play a significant ecological role in NBay, particularly since their realized niches suggest they are eurythermal and able to tolerate the >25 °C temperature range of NBay. Four distinct species assemblages that grouped by season were best explained by surface water temperature. When compared to historical records, we found that the cold-water species Thalassiosira nordenskioeldii has decreased in persistence over time, suggesting that increasing surface water temperature has influenced the ecology of phytoplankton in NBay.


2018 ◽  
Vol 77 (2) ◽  
Author(s):  
Bartosz Czernecki ◽  
Mariusz Ptak

The paper presents historical (1971-2015) and scenario-based (2006-2100) changes in surface water temperatures in 10 lakes of Poland. The analysis of historical measurement (1971-2015) showed that mean annual lake surface water temperature (LSWT) was characterised by an increasing tendency by 0.37°C∙dec-1 on average, and was higher by 0.01°C∙dec-1 than air temperature in the analogical period. The highest increase in LSWT was recorded in spring months (April, May) and in summer (July). The future changes in LSWT was based on simulations of 33 AOGCMs available in the scope of CMIP5 project for RCPs: 2.6, 4.5, 6.0, and 8.5. The developed empirical-statistical downscaling models (ESD) use the air temperature field as predictors, with consideration of autocorrelation for two preceding months. ESD models are characterised by high quality of reconstruction of water temperatures in the historical period, with correlation from 0.82 (December, February) to 0.93 (July). The future CMIP5 scenarios for the period 2006-2100 assume an increase in air temperature at the end of the 21st century from +1.8°C (RCP 2.6) to +5.1°C (RCP 8.5) in reference to the period 1971-2005. According to the downscaling models, this corresponds to an increase in water temperature in the analysed lakes ranging from +1.4°C (RCP 2.6) to +4.2°C (RCP 8.5) in the years 2081-2100, respectively, with evident variability between the adopted emission paths beginning from the period 2041-2060. At a monthly scale, water temperature will increase the slowest in February (2081-2100: RCP 2.6 = +0.5°C, RCP 8.5 = +1.8°C). The highest increase in temperature will occur from May to August (RCP 8.5 = +6°C in June).Substantial effects of transformations of the thermal regime are already observed today, e,g. in the reduction of the ice season length. According to developed scenarios, a further considerable increase in water temperature will be the primary factor determining the transformation of lake ecosystems. The obtained results provide a theoretical basis for further research conducted in the scope of many disciplines, among others hydrology, hydrobiology, ecology, water management, energy production, etc. In the case of Poland, issues related to low water resources per capita are particularly important. Contemporary studies concerning changes in water resources showed that the natural factor playing the key role in their reduction is temperature increase and therefore it should constitute for the possibly fast development of multidisciplinary concepts of mitigation policy to potential impact of climate change. 


Sign in / Sign up

Export Citation Format

Share Document