Nick J. Mount

Two decades of anarchy? Emerging themes and outstanding challenges for neural network river forecasting

This paper traces two decades of neural network rainfall-runoff and streamflow modelling, collectively termed ‘river forecasting’. The field is now firmly established and the research community involved has much to offer hydrological science. First, however, it will be necessary to converge on more objective and consistent protocols for: selecting and treating inputs prior to model development; extracting physically meaningful insights from each proposed solution; and improving transparency in the benchmarking and reporting of experimental case studies. It is also clear that neural network river forecasting solutions will have limited appeal for operational purposes until confidence intervals can be attached to forecasts. Modular design, ensemble experiments, and hybridization with conventional hydrological models are yielding new tools for decision-making. The full potential for modelling complex hydrological systems, and for characterizing uncertainty, has yet to be realized. Further gains could also emerge from the provision of an agreed set of benchmark data sets and associated development of superior diagnostics for more rigorous intermodel evaluation. To achieve these goals will require a paradigm shift, such that the mass of individual isolated activities, focused on incremental technical refinement, is replaced by a more coordinated, problem-solving international research body.

Estimation of error in bankfull width comparisons from temporally sequenced raw and corrected aerial photographs

This study investigates the propagation of error through image-to-image comparison of 285 river bankfull width measurements of the Afon Trannon, mid-Wales. Bankfull width is quantified from both aerial photographs analysed as rectified images in ERDAS Imagine OrthoMax and raw images in Paintshop Pro. A method for the robust estimation of bankfull width measurement error through temporal sequences of scanned aerial photographs is presented and the improvement in accuracy achieved using rectified imagery is quantified. Results from this study are placed in the context of previously published rates of bankfull width change, from a wide range of river scales, and the bankfull change rates for robust medium-term analysis using approximately 1:10,000 historical aerial photography are identified.

Plantation forestry impacts on sediment yields and downstream channel dynamics in the UK: a review:

The impact of coniferous plantation forest on erosion and sediment yields in the UK uplands over the past three decades is reviewed by examining background or natural suspended sediment yields (SSY), bed load yields (BLY) and bank erosion and comparing with studies of‘disturbed’catchments. This paper collates all the UK studies that have monitored changes in erosion and sediment yields at the forest establishment, mature forest and timber harvesting phases of the forest cycle. A simple model based on this comprehensive examination of studies to date suggests that mean sediment yields increase at the initial ground disturbance phase, recover as the forest matures and increase again more significantly at the timber harvesting phase. The likely downstream impacts of these changes in sediment yields on channel dynamics and management is discussed, introducing sediment wave theory with particular reference to the generation of bed load waves downstream of forested catchments. Modelling of downstream changes in unit stream power has implications for the accumulation of nonpoint source sediment in sediment waves. Published studies in which channel changes have resulted from the passage of sediment waves are collated and it is concluded that the translating wave model is best supported. The paper concludes with a discussion of the broader implications for river management and presents advice for river managers.

Data-driven modelling approaches for socio-hydrology: opportunities and challenges within the Panta Rhei Science Plan

ABSTRACT “Panta Rhei – Everything Flows” is the science plan for the International Association of Hydrological Sciences scientific decade 2013–2023. It is founded on the need for improved understanding of the mutual, two-way interactions occurring at the interface of hydrology and society, and their role in influencing future hydrologic system change. It calls for strategic research effort focused on the delivery of coupled, socio-hydrologic models. In this paper we explore and synthesize opportunities and challenges that socio-hydrology presents for data-driven modelling. We highlight the potential for a new era of collaboration between data-driven and more physically-based modellers that should improve our ability to model and manage socio-hydrologic systems. Crucially, we approach data-driven, conceptual and physical modelling paradigms as being complementary rather than competing, positioning them along a continuum of modelling approaches that reflects the relative extent to which hypotheses and/or data are available to inform the model development process. EDITOR D. Koutsoyiannis; ASSOCIATE EDITOR not assigned

Learner immersion engagement in the 3D virtual world: principles emerging from the DELVE project

Abstract This paper investigates the issues surrounding the use of 3D virtual worlds to enhance learner immersion through improved learner engagement. It is based on findings from the JISC-funded DEsign of Learning spaces in 3D Virtual Environments (DELVE) project at the University of Nottingham and the Open University. Given continued confusion about the term immersion, what it means for a learner to be immersed, and the relationship between immersion, presence and engagement, notions of immersion and engagement in 3D virtual environments are explored in the context of previous published studies ranging from virtual reality to psychology. The resultant improved understanding of the terminology is then used as the basis for coding results from a qualitative, inductive analysis of 20 students that undertook a substantive learning task in the virtual environment Second Life. Emergent themes from the analysis identify key factors that act to both enhance and restrict learner engagement in 3D virtual worlds and a set of principles for practitioners who wish to use 3D virtual environments to enhance learner engagement is presented.

Participatory modelling for stakeholder involvement in the development of flood risk management intervention options

Advancing stakeholder participation beyond consultation offers a range of benefits for local flood risk management, particularly as responsibilities are increasingly devolved to local levels. This paper details the design and implementation of a participatory approach to identify intervention options for managing local flood risk. Within this approach, Bayesian networks were used to generate a conceptual model of the local flood risk system, with a particular focus on how different interventions might achieve each of nine participant objectives. The model was co-constructed by flood risk experts and local stakeholders. The study employs a novel evaluative framework, examining both the process and its outcomes (short-term substantive and longer-term social benefits). It concludes that participatory modelling techniques can facilitate the identification of intervention options by a wide range of stakeholders, and prioritise a subset for further investigation. They can help support a broader move towards active stakeholder participation in local flood risk management. Case study uses participatory modelling approach to involve local stakeholders.Application of Bayesian networks to co-construct conceptual model of flood risk.Testing of model with stakeholders to identify recommended interventions.Approach evaluated using five requirements specific to flood risk management.Approach effective at identifying interventions that merit further exploration.

Discussion of “Evapotranspiration modelling using support vector machines”*

The data-driven modelling paradigm is special in the extent to which input–output data sets, and the structures that exist within them, are the fundamental drivers of the entiremodelling process. Consequently, those engaged in data-driven modelling must recognise that an adequate description and assessment of any data sets used is an essential requirement for the proper application of the paradigm. This is equally as important as the description of the particular mathematical algorithm that is used to model the data. It is also essential if data-drivenmodelling is to complywith theprinciplesof the “scientificmethod”, i.e. experiments must be repeatable and results must be reproducible. However, in many published papers, the description of the data sets involved and the particular algorithms used to model the data are unbalanced, with brief descriptions of data sets and pre-processing operations contrasting with long and detailed descriptions of algorithmic and computational approaches. As a result, criticisms of data-drivenmodels are emergingwhose foci includea lackofclarity in, and justificationof,datasetpreprocessing steps, input variable selection, repeatability and independent verification of modelling applications, aswell as theneed toperformdata-drivenmodelling in the first place. Dealing with these criticisms has sometimes taken the form of extensive post-publication discussions, in which authors offer piecemeal criticisms of different aspectsofdifferentpapersandof theerrorsoromissions in their reporting of data sets, including preand postprocessing operations; examples are provided below. This is an onerous process that is unlikely to deliver a sound or comprehensive “code of behaviour”. As a case in point, two recent papers on modelling reference crop evapotranspiration have been heavily criticised in subsequent discussions for failing to provide sufficient information about the data sets involved. Aksoy et al. (2007) raised the issue of unreportedmissing records in data sets used by Kisi (2006) to model reference crop evapotranspiration in California, USA. It was argued that if a particular data set had some missing observations, that fact should be reported clearly (particularly in journal papers) even if the number of missing cases was small, so that the reported results are comparable with other studies. Kisi (2007), in response to this challenge, subsequently disclosed that a linear regression model had actually been used to provide estimated records for a 12-day period and this too should have been reported in the original paper, not only to ensure others could repeat the work, but also because, through this data-infilling process, the authors had introduced an untested assumption of linearity in the data. Abrahart et al. (2009) questioned the operational processes involved in the removal of incomplete entries by Aytek et al. (2008); also modelling reference crop evapotranspiration in California. In particular, the description of their “data cleansing” operation, a process that was specified in terms of involving “somemissing records”, was criticised as being insufficient. Aytek et al. (2009), in response to this challenge, subsequently stated that removal comprised not just records that had missing observations but also the removal of items flagged in Hydrological Sciences Journal – Journal des Sciences Hydrologiques, 55(8) 201

Neuroemulation: definition and key benefits for water resources research

Abstract Neuroemulation is the art and science of using a neural network model to replicate the external behaviour of some other model or component of a model. It is an independent activity that is distinct from neural network-based simulation. Neuroemulation has become a recognized and established sub-discipline in many spheres of study, but remains poorly defined in the field of water resources research. Its many potential benefits have not yet been adequately recognized or established. Lack of recognition can in part be attributed to difficulties involved in identifying, collating and synthesizing published studies on neuroemulation: query-based searching of a publications database fails to identify papers concerned with a field of study, for which no agreed conceptual and/or terminological framework as yet exists. Therefore, in this paper, we provide a first attempt at defining such a framework for use in water resources investigations. We identify eight key benefits offered by neuroemulation and exemplify current activities with relevant examples taken from published research in the field. The concluding section highlights a number of strategic research directions related to developing the identified potential of neuroemulator applications for water resources modelling. Editor D. Koutsoyiannis Citation Abrahart, R.J., Mount, N.J. and Shamseldin, A.Y., 2012. Neuroemulation: definition and key benefits for water resources research. Hydrological Sciences Journal, 57 (3), 407–423.

Self-organizing maps and boundary effects: quantifying the benefits of torus wrapping for mapping SOM trajectories

In this study the impact of a planar and toroidal self-organizing map (SOM) configuration are investigated with respect to their impact on SOM trajectories. Such trajectories are an encoding of processes within an n-dimensional input data set and offer an important means of visualizing and analyzing process complexity in large n-dimensional problem domains. However, discontinuity associated with boundaries in the standard, planar SOM results in error that limits their analytical use. Previous studies have recommended the use of a toroidal SOM to reduce these errors, but fall short of a fully quantified analysis of the benefits that result. In this study, the comparative analysis of fifteen pairs of identically initiated and trained SOMs, of planar and toroidal configuration, allows the error in trajectory magnitude to be quantified and visualized; both within the SOM and data space. This offers an important insight into the impact of planar SOM boundaries that goes beyond the general, statistical measures of clustering efficacy associated with previous work. The adoption of a toroidal SOM can be seen to improve the distribution of error in the trajectory sets, with the specific spatial configuration of SOM neurons associated with the largest errors changing from those at the corners of the planar SOM to a more complex and less predictable pattern in the toroidal SOM. However, this improvement is limited to the smallest 60% of errors, with torus and planar SOMs performing similarly for the largest 40%.

Improved validation framework and R-package for artificial neural network models

Validation is a critical component of any modelling process. In artificial neural network (ANN) modelling, validation generally consists of the assessment of model predictive performance on an independent validation set (predictive validity). However, this ignores other aspects of model validation considered to be good practice in other areas of environmental modelling, such as residual analysis (replicative validity) and checking the plausibility of the model in relation to a priori system understanding (structural validity). In order to address this shortcoming, a validation framework for ANNs is introduced in this paper that covers all of the above aspects of validation. In addition, the validann R-package is introduced that enables these validation methods to be implemented in a user-friendly and consistent fashion. The benefits of the framework and R-package are demonstrated for two environmental modelling case studies, highlighting the importance of considering replicative and structural validity in addition to predictive validity. A comprehensive validation framework for ANNs is proposed.The validann R-package for implementing the validation framework is introduced.Application of the framework and R-package is demonstrated on two real case studies.Results reveal that predictively valid ANN models may not be credible.Adoption of the framework leads to improvements in overall ANN validity.