A regional neural network ensemble for predicting mean daily river water temperature

DeWeber, J. Tyrell; Wagner, Tyler

doi:10.1016/j.jhydrol.2014.05.035

Cited by 116 publications

(94 citation statements)

References 67 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…5), which is consistent with the higher runoff coefficient in HG ( Table 1). The 12% decrease in forest area and 60% increase in developed land area in catchment HG over the past 33 years (Table 1) implies a considerable loss in river riparian canopy cover and shading, which further contributes to increasing water temperature (Gu et al, 2014;DeWeber and Wagner, 2014;Simmons et al, 2014).…”

Section: Rivermentioning

confidence: 99%

“…Although the relationship between air and water temperature is generally strong, the strength of such a relationship varies regionally and temporally, and can be highly site specific due to additional influences from local hydrology and human activities, such as changes in land-use and population density (Arismendi et al, 2012;Orr et al, 2015;DeWeber and Wagner, 2014). It is commonly observed that water temperature is inversely related to river discharge, reflecting a reduced thermal buffering capacity due to decreasing flow volumes, increasing travel time, and diminished dilution capacity for inputs of thermal effluents (Gu and Li, 2002;Webb et al, 2003;Moatar and Gailhard, 2006;Albek and Albek, 2009).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Changes in river water temperature between 1980 and 2012 in Yongan watershed, eastern China: Magnitude, drivers and models

Chen

Guo

et al. 2016

Journal of Hydrology

View full text Add to dashboard Cite

s u m m a r yClimate warming is expected to have major impacts on river water quality, water column/hyporheic zone biogeochemistry and aquatic ecosystems. A quantitative understanding of spatio-temporal air (T a ) and water (T w ) temperature dynamics is required to guide river management and to facilitate adaptations to climate change. This study determined the magnitude, drivers and models for increasing T w in three river segments of the Yongan watershed in eastern China. Over the 1980-2012 period, T w in the watershed increased by 0.029-0.046°C yr À1 due to a $0.050°C yr À1 increase of T a and changes in local human activities (e.g., increasing developed land and population density and decreasing forest area). A standardized multiple regression model was developed for predicting annual T w (R 2 = 0.88-0.91) and identifying/ partitioning the impact of the principal drivers on increasing T w :T a (76 ± 1%), local human activities (14 ± 2%), and water discharge (10 ± 1%). After normalizing water discharge, climate warming and local human activities were estimated to contribute 81-95% and 5-19% of the observed rising T w , respectively. Models forecast a 0.32-1.76°C increase in T w by 2050 compared with the 2000-2012 baseline condition based on four future scenarios. Heterogeneity of warming rates existed across seasons and river segments, with the lower flow river and dry season demonstrating a more pronounced response to climate warming and human activities. Rising T w due to changes in climate, local human activities and hydrology has a considerable potential to aggravate river water quality degradation and coastal water eutrophication in summer. Thus it should be carefully considered in developing watershed management strategies in response to climate change.

show abstract

Section: Rivermentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Changes in river water temperature between 1980 and 2012 in Yongan watershed, eastern China: Magnitude, drivers and models

Chen

Guo

et al. 2016

Journal of Hydrology

View full text Add to dashboard Cite

show abstract

“…This fostered the development of many stochastic and deterministic models (e.g. Mohseni et al, 1998;Segura et al, 2014;Chang and Psaris, 2013;DeWeber and Wagner, 2014;Meier et al, 2003;Westhoff et al, 2007). The former type relies on a statistical analysis to empirically relate stream temperature to climatic and physiographic variables, such as air temperature, discharge, altitude or channel width (see Benyahya et al, 2007, for a complete review of this subject).…”

Section: A Gallice Et Al: Stream Temperature Prediction In Ungaugedmentioning

confidence: 99%

“…Similarly, values between 30 and 200 m are assumed for the width of the riparian buffer affecting stream temperature at a given point (e.g. Jones et al, 2006;Scott et al, 2002;DeWeber and Wagner, 2014).…”

Section: Space-averaging Of the Predictor Variablesmentioning

confidence: 99%

“…Arscott et al, 2001;Mayer, 2012;Imholt et al, 2013) or linear mixed modelling (Macedo et al, 2013), but also more advanced techniques such as geographically weighted regression (Pratt and Chang, 2012;Chang and Psaris, 2013), networked kriging models (Gardner and Sullivan, 2004;Isaak et al, 2010;Ruesch et al, 2012) or machine learning techniques (e.g. Westenbroek et al, 2010;Hill et al, 2013;DeWeber and Wagner, 2014).…”

Section: Investigation Of a New Modelling Approachmentioning

confidence: 99%

See 1 more Smart Citation

Stream temperature prediction in ungauged basins: review of recent approaches and description of a new physics-derived statistical model

Gallice

Schaefli

Lehning

et al. 2015

Hydrol. Earth Syst. Sci.

View full text Add to dashboard Cite

Abstract. The development of stream temperature regression models at regional scales has regained some popularity over the past years. These models are used to predict stream temperature in ungauged catchments to assess the impact of human activities or climate change on riverine fauna over large spatial areas. A comprehensive literature review presented in this study shows that the temperature metrics predicted by the majority of models correspond to yearly aggregates, such as the popular annual maximum weekly mean temperature (MWMT). As a consequence, current models are often unable to predict the annual cycle of stream temperature, nor can the majority of them forecast the inter-annual variation of stream temperature. This study presents a new statistical model to estimate the monthly mean stream temperature of ungauged rivers over multiple years in an Alpine country (Switzerland). Contrary to similar models developed to date, which are mostly based on standard regression approaches, this one attempts to incorporate physical aspects into its structure. It is based on the analytical solution to a simplified version of the energy-balance equation over an entire stream network. Some terms of this solution cannot be readily evaluated at the regional scale due to the lack of appropriate data, and are therefore approximated using classical statistical techniques. This physics-inspired approach presents some advantages: (1) the main model structure is directly obtained from first principles, (2) the spatial extent over which the predictor variables are averaged naturally arises during model development, and (3) most of the regression coefficients can be interpreted from a physical point of view -their values can therefore be constrained to remain within plausible bounds. The evaluation of the model over a new freely available data set shows that the monthly mean stream temperature curve can be reproduced with a rootmean-square error (RMSE) of ±1.3 • C, which is similar in precision to the predictions obtained with a multi-linear regression model. We illustrate through a simple example how the physical aspects contained in the model structure can be used to gain more insight into the stream temperature dynamics at regional scales.

show abstract

Spatial statistical network models for stream and river temperature in New England, USA

Detenbeck

Morrison

Abele

et al. 2016

Water Resources Research

View full text Add to dashboard Cite

Watershed managers are challenged by the need for predictive temperature models with sufficient accuracy and geographic breadth for practical use. We described thermal regimes of New England rivers and streams based on a reduced set of metrics for the May-September growing season (July or August median temperature, diurnal rate of change, and magnitude and timing of growing season maximum) chosen through principal component analysis of 78 candidate metrics. We then developed and assessed spatial statistical models for each of these metrics, incorporating spatial autocorrelation based on both distance along the flow network and Euclidean distance between points. Calculation of spatial autocorrelation based on travel or retention time in place of network distance yielded tighter-fitting Torgegrams with less scatter but did not improve overall model prediction accuracy. We predicted monthly median July or August stream temperatures as a function of median air temperature, estimated urban heat island effect, shaded solar radiation, main channel slope, watershed storage (percent lake and wetland area), percent coarse-grained surficial deposits, and presence or maximum depth of a lake immediately upstream, with an overall root-mean-square prediction error of 1.4 and 1.58C, respectively. Growing season maximum water temperature varied as a function of air temperature, local channel slope, shaded August solar radiation, imperviousness, and watershed storage. Predictive models for July or August daily range, maximum daily rate of change, and timing of growing season maximum were statistically significant but explained a much lower proportion of variance than the above models (5-14% of total). PUBLICATIONSHistorically, when developing predictive temperature models for streams, there has been a trade-off between accuracy of model predictions and practical spatial extent of model coverage. Mechanistically based heat budget models such as SNTEMP [Krause et al., 2004] can predict stream temperature within a few tenths of a degree. However, the only mechanistic model that has been linked with a GIS interface to facilitate regional application is BASIN TEMP [Allen 2008], which is not commercially available. One intermediate solution could be the application of WET-Temp [Cox and Bolte, 2007], a spatially explicit networkbased model for continuous temperature simulation. However, combined preprocessing and run times for an entire region would be prohibitive. LeBlanc et al. [1997] identified a second intermediate solution using a simulation model of the effects of urbanization on water temperature in unregulated streams. They determined that model outputs were sensitive to only four of the model inputs: vegetation transmissivity, channel width, sun angle, and groundwater discharge, thus paving the way to development of a much simpler predictive model. This approach has only been applied at the reach scale, however, and needs to be incorporated into a network model to allow examination of cumulative effects on temperature throughout...

show abstract

A regional neural network ensemble for predicting mean daily river water temperature

Cited by 116 publications

References 67 publications

Changes in river water temperature between 1980 and 2012 in Yongan watershed, eastern China: Magnitude, drivers and models

Changes in river water temperature between 1980 and 2012 in Yongan watershed, eastern China: Magnitude, drivers and models

Stream temperature prediction in ungauged basins: review of recent approaches and description of a new physics-derived statistical model

Spatial statistical network models for stream and river temperature in New England, USA

Contact Info

Product

Resources

About