A threshold artificial neural network model for improving runoff prediction in a karst watershed

Meng, Xianmeng; Yin, Maosheng; Ning, Libo; Liu, Dengfeng; Xue, Xianwu

doi:10.1007/s12665-015-4562-9

Cited by 27 publications

(15 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In view of the estimates of ET, all the models achieved almost consistent, high modeling accuracy, suggesting that machine learning techniques can be expected as powerful tools to simulate and predict ET. What's more, it should be acknowledged that in recent years, these modeling techniques have been triumphantly applied in numerous branches in hydrology for nonlinear time series analysis, such as reference ET [77,78] and evaporation prediction [33], rainfall [79] and runoff forecasting [80,81].…”

Section: Discussionmentioning

confidence: 99%

Modeling and Predicting Carbon and Water Fluxes Using Data-Driven Techniques in a Forest Ecosystem

Dou

Yang

2017

Forests

View full text Add to dashboard Cite

Accurate estimation of carbon and water fluxes of forest ecosystems is of particular importance for addressing the problems originating from global environmental change, and providing helpful information about carbon and water content for analyzing and diagnosing past and future climate change. The main focus of the current work was to investigate the feasibility of four comparatively new methods, including generalized regression neural network, group method of data handling (GMDH), extreme learning machine and adaptive neuro-fuzzy inference system (ANFIS), for elucidating the carbon and water fluxes in a forest ecosystem. A comparison was made between these models and two widely used data-driven models, artificial neural network (ANN) and support vector machine (SVM). All the models were evaluated based on the following statistical indices: coefficient of determination, Nash-Sutcliffe efficiency, root mean square error and mean absolute error. Results indicated that the data-driven models are capable of accounting for most variance in each flux with the limited meteorological variables. The ANN model provided the best estimates for gross primary productivity (GPP) and net ecosystem exchange (NEE), while the ANFIS model achieved the best for ecosystem respiration (R), indicating that no single model was consistently superior to others for the carbon flux prediction. In addition, the GMDH model consistently produced somewhat worse results for all the carbon flux and evapotranspiration (ET) estimations. On the whole, among the carbon and water fluxes, all the models produced similar highly satisfactory accuracy for GPP, R and ET fluxes, and did a reasonable job of reproducing the eddy covariance NEE. Based on these findings, it was concluded that these advanced models are promising alternatives to ANN and SVM for estimating the terrestrial carbon and water fluxes.

show abstract

Section: Discussionmentioning

confidence: 99%

Modeling and Predicting Carbon and Water Fluxes Using Data-Driven Techniques in a Forest Ecosystem

Dou

Yang

2017

Forests

View full text Add to dashboard Cite

show abstract

“…Natural or artificial tracers can provide additional information on water flow characteristics and matter transport (Herczeg et al 1997;Gabrovšek et al 2010;Reed et al 2010;Ravbar et al 2012;Bonacci & Andrić 2015). Numerical modelling and computational approaches have made significant progress in usefulness and application regarding thresholds in karst hydrogeology (Gill et al 2013;Mayaud et al 2014;Schmidt et al 2014;Meng et al 2015). However, an initial conceptual understanding of the area of interest is a prerequisite for further numerical modelling (Davis & Putnam 2013).…”

Section: Introductionmentioning

confidence: 99%

Treshold-controlled three-stage hydraulic behaviour of a mantled shallow carbonate aquifer (Tuhala karst area, North Estonia)

Koit

Ravbar²,

Marandi³

et al. 2017

View full text Add to dashboard Cite

UDC 556.34:551.44(474.2) Oliver Koit, Nataša Ravbar, Andres Marandi & Jaanus Terasmaa: Treshold-controlled three-stage hydraulic behaviour of a mantled shallow carbonate aquifer (Tuhala karst area, North Estonia) As karst aquifers are often characterised by non-linear behaviour, ascertaining the turning points in their hydraulic regime may provide essential information on the functioning of the aquifer. These characteristics also apply to the Silurian-Ordovician aquifer system, composed of diverse carbonate rocks, in Estonia. The aquifer system is an important source of drinking water in northern Estonia. It also comprises the NabalaRakvere aquifer, a locally important groundwater resource that underlies the Tuhala karst area famous for an intermittently overflowing karst spring known as the Witch's Well. The water rich in humic substances of the Tuhala River recharges the Tuhala karst system, which is drained by two spring groups. In order to develop measures for the sustainable management of the aquifer in the future, the aim of this study was to enhance understanding of the hydraulic behaviour and hydrodynamic properties of the Tuhala karst system. From October 2014 to December 2016 an extensive field campaign was carried out comprising the observation of 22 surface-and groundwater monitoring points for water level and physico-chemical parameters and the performance of four quantitative tracer tests. The data obtained were evaluated in accordance with the inputoutput water level relation curves and a conceptual model of the system was set up. The results show that a strong hydraulic link exists, primarily through well-developed conduits between the recharge and discharge area of the karst system. Depending on the hydrological conditions, groundwater flow direc- IzvlečekUDK 556.34:551.44(474.2) Oliver Koit, Nataša Ravbar, Andres Marandi & Jaanus Terasmaa: Določanje mejnih vrednosti tristopenjskega hidravljičnega obnašanja pokritega plitvega karbonatnega vodonosnika (Tuhalski kras, severna Estonija) Kraški vodonosniki se pogosto obnašajo nelinearno, zato lahko z ugotavljanjem mejnih vrednosti v hidravličnem režimu zagotovimo bistvene informacije o delovanju vodonosnika. To velja tudi za silurijsko-ordovicijski vodonosni sistem v Estoniji, ki ga sestavljajo različne karbonatne kamnine in je pomemben vir pitne vode na severu države. Vključuje tudi vodonosnik Nabala-Rakvere, ki je lokalno pomemben vodni vir. Na tem območju se nahaja Tuhalski kras, ki je znan po občasno bruhajočem kraškem izviru Witch's Well (čarovničin vodnjak). Tuhalski vodonosnik napaja reka Tuhala, bogata z huminskimi snovmi, prazni pa se skozi dve skupini izvirov. Da bi v prihodnje razvili ukrepe za trajnostno upravljanje vodonosnika, je bil namen te raziskave izboljšati razumevanje hidravličnega obnašanja in hidrodinamičnih lastnosti kraškega sistema. Od oktobra 2014 do decembra 2016 smo opravili obsežno terensko delo, ki je obsegalo spremljanje vodostajev in fizikalno-kemijskih parametrov v 23 površinskih in podzemnih točkah ter izve...

show abstract

“…By the action of lithology and tectonics, soluble carbonate rocks form a dual structure by corrosion and corrasion in the surface and subsurface. This structure with severe heterogeneity causes complex hydraulic conditions and spatiotemporal variability of parameters (Meng et al, 2015). Rain falls into shafts and sinks, thus causing the subsurface to crack rapidly, particularly in several karst mountain areas, the water infiltration coefficient is up to 80 % (Liu and Li, 2007;Meng and Wang, 2010) and the soil loss is also strong (Febles et al, 2014).…”

Section: Introductionmentioning

confidence: 99%

“…Moreover, karst also provides diverse subterranean habitats, including epikarst, cave streams, drip pools, springs, and interstices (Bonacci et al, 2009). In karst regions a large number of studies have focused on hydrology, soil erosion, water resources, and ecosystems based on the watershed unit (Rimmer and Salingar, 2006;Navas et al, 2013;McCormack et al, 2014). However, many studies do not assess the accuracy of the scope of the watershed, or several only assess the catchment scope for a single spring in the watershed (key papers are summarized in Table 1 in relation).…”

Section: Introductionmentioning

confidence: 99%

Delineating small karst watersheds based on digital elevation model and eco-hydrogeological principles

et al. 2016

View full text Add to dashboard Cite

Abstract. Dominated by specific eco-hydrogeological backgrounds, a small watershed delineated by using the traditional method is always inauthentic in karst regions because it cannot accurately reflect the eco-hydrological process of the dual structure of the surface and subsurface. This study proposes a new method for the delineation of small watersheds based on digital elevation models (DEMs) and ecohydrogeological principles in karst regions. This method is applied to one section of the tributary area (Sancha River) of the Yangtze River in China. By comparing the quantity, shape, superimposition, and characteristics of the internal hydrological process of a small watershed extracted by using the digital elevation model with that extracted by using the proposed method of this study, we conclude that the small karst watersheds extracted by the new method accurately reflect the hydrological process of the river basin. Furthermore, we propose that the minimum unit of the river basin in karst regions should be the watershed, whose exit is the corrosion and corrasion baselevel and a further division of watershed may cause a significant inconsistency with the true eco-hydrological process.

show abstract

A threshold artificial neural network model for improving runoff prediction in a karst watershed

Cited by 27 publications

References 38 publications

Modeling and Predicting Carbon and Water Fluxes Using Data-Driven Techniques in a Forest Ecosystem

Modeling and Predicting Carbon and Water Fluxes Using Data-Driven Techniques in a Forest Ecosystem

Treshold-controlled three-stage hydraulic behaviour of a mantled shallow carbonate aquifer (Tuhala karst area, North Estonia)

Delineating small karst watersheds based on digital elevation model and eco-hydrogeological principles

Contact Info

Product

Resources

About