Comparison between response surface models and artificial neural networks in hydrologic forecasting

Developing an efficient and accurate hydrologic forecasting model is crucial to managing water resources and flooding issues. In this study, response surface (RS) models including multiple linear regression (MLR), quadratic response surface (QRS), and nonlinear response surface (NRS) were applied to daily runoff (e.g., discharge and water level) prediction. Two catchments, one in southeast China and the other in western Canada, were used to demonstrate the applicability of the proposed models. Their performances were compared with artificial neural network (ANN) models, trained with the learning algorithms of the gradient descent with adaptive learning rate (ANN-GDA) and Levenberg-Marquardt (ANN-LM). The performances of both RS and ANN in relation to the lags used in the input data, the length of the training samples, long-term (monthly and yearly) predictions, and peak value predictions were also analyzed. The results indicate that the QRS and NRS were able to obtain equally good performance in runoff prediction, as compared with ANN-GDA and ANN-LM, but require lower computational efforts. The RS models bring practical benefits in their application to hydrologic forecasting, particularly in the cases of short-term flood forecasting (e.g., hourly) due to fast training capability, and could be considered as an alternative to ANN

Broadband futures : content, connectivity and control

This introduction to the Broadband Futures: Content, Connectivity and Control special issue begins with the claim that we are at the start of a new period of debate and analysis about network infrastructure and use that declares government failure and reinvigorates the drive to a free-market solution in Australia and New Zealand. But questions remain about the effectiveness and social value of network developments, regardless of the degree of government investment and regulation. As canvassed in this introduction, and the articles that make up the special issue, we can only move forward, both in research and practical implementation, if we accept the intricate weaving together of past, present and future interactions between governments, service providers and their consumers in broadband development

Genetic connectivity between trans-oceanic populations of Capreolia implexa (Gelidiales, Rhodophyta) in cool temperate waters of Australasia and Chile.

Capreolia is a monospecific genus of gelidioid red algae and has been considered to be endemic to Australasia. This is the first report on the occurrence of Capreolia implexa outside of Australasian waters, based on investigations of fresh collections in southern Chile as well as Australia and New Zealand. Thalli are prostrate and form entangled turfs, growing on high intertidal rocks at three locations in Chile. Analyses of rbcL and cox1 revealed that C. implexa was of Australasian origin and also distinct from its relatives. Analyses of 1356. bp of cox1 revealed cryptic diversity, consisting of two genealogical groups within C. implexa; one present in Australia and New Zealand, and the other in Chile and Stewart Island, New Zealand. The extremely low genetic diversity found in C. implexa in Chile and the absence of shared haplotypes between Chile and Australasia suggest genetic bottleneck possibly as a result of colonization after dispersal by rafting from Stewart Island, New Zealand to Chile. © 2014 Elsevier B.V.

Hydrologic landscape regionalisation using deductive classification and random forests

Landscape classification and hydrological regionalisation studies are being increasingly used in ecohydrology to aid in the management and research of aquatic resources. We present a methodology for classifying hydrologic landscapes based on spatial environmental variables by employing non-parametric statistics and hybrid image classification. Our approach differed from previous classifications which have required the use of an a priori spatial unit (e.g. a catchment) which necessarily results in the loss of variability that is known to exist within those units. The use of a simple statistical approach to identify an appropriate number of classes eliminated the need for large amounts of post-hoc testing with different number of groups, or the selection and justification of an arbitrary number. Using statistical clustering, we identified 23 distinct groups within our training dataset. The use of a hybrid classification employing random forests extended this statistical clustering to an area of approximately 228,000 km2 of south-eastern Australia without the need to rely on catchments, landscape units or stream sections. This extension resulted in a highly accurate regionalisation at both 30-m and 2.5-km resolution, and a less-accurate 10-km classification that would be more appropriate for use at a continental scale. A smaller case study, of an area covering 27,000 km2, demonstrated that the method preserved the intra- and inter-catchment variability that is known to exist in local hydrology, based on previous research. Preliminary analysis linking the regionalisation to streamflow indices is promising suggesting that the method could be used to predict streamflow behaviour in ungauged catchments. Our work therefore simplifies current classification frameworks that are becoming more popular in ecohydrology, while better retaining small-scale variability in hydrology, thus enabling future attempts to explain and visualise broad-scale hydrologic trends at the scale of catchments and continents.

Hydrologic response of a tropical watershed to urbanization

 Urbanization has profound influence on the hydrologic response of landscapes. Urban transformation affects the storages and processes that determine the generation of hydrologic fluxes. It also changes the time-scales associated with hydrologic processes. Shifts in hydrologic response of the watershed unit due to urban transformation may be more complex than the simple linear mixing (weighted sum) of responses from the urbanized and non-urbanized fractions of the landscape. This may especially be the case for tropical watersheds where the precipitation forcing of the watershed is frequent and intense - interacting with the shifting time-scales and changing storages with increasing urbanization. In this study, a fully distributed hydrological model (MOBIDIC) that captures hydrologic dynamics during storms and interstorms is applied in order to characterize the potentially nonlinear response of a tropical watershed to urban transformation. Indices that quantify the departures from linear response are introduced and used to test the effects of urbanization on different hydrologic processes and fluxes in a mixed (urban and non-urban) watershed. The tropical Kranji watershed in Singapore is used in this study. Fortunately two sub-watersheds within Kranji that have streamflow gaging stations are well-suited for the calibration of the model. One sub-watershed is nearly fully urbanized and another is pristine (non-urban). As a result the contrasting components (urban and non-urban) can be calibrated in the model. The simulation system is then used to assess the hydrologic response due to changing levels of urbanization. For some fluxes and storages, the hydrologic response due to changing urban fraction cannot be simply predicted from a linear mixing model.

Effect of mood, social connectivity and age in online depression community via topic and linguistic analysis

Depression afflicts one in four people during their lives. Several studies have shown that for the isolated and mentally ill, the Web and social media provide effective platforms for supports and treatments as well as to acquire scientific, clinical understanding of this mental condition. More and more individuals affected by depression join online communities to seek for information, express themselves, share their concerns and look for supports [12]. For the first time, we collect and study a large online depression community of more than 12,000 active members from Live Journal. We examine the effect of mood, social connectivity and age on the online messages authored by members in an online depression community. The posts are considered in two aspects: what is written (topic) and how it is written (language style). We use statistical and machine learning methods to discriminate the posts made by bloggers in low versus high valence mood, in different age categories and in different degrees of social connectivity. Using statistical tests, language styles are found to be significantly different between low and high valence cohorts, whilst topics are significantly different between people whose different degrees of social connectivity. High performance is achieved for low versus high valence post classification using writing style as features. The finding suggests the potential of using social media in depression screening, especially in online setting.

Ecological connectivity or barrier fence? Critical choices on the agricultural margins of Western Australia

Western Australia's State Barrier Fence represents a continuation of colonial era attitudes that considered kangaroos, emus and dingoes as 'vermin'. Recent plans to upgrade and extend the Barrier Fence have shown little regard for ecological impacts or statutory environmental assessment processes.

Assessing spatio-temporal hydrologic variability: a case-study in western Victoria

This research explored spatial and temporal variability in the relationships between land cover, topography, and climate on streamflows and wetland extents in western Victoria.

Patterns of connectivity and population structure of the southern calamary Sepioteuthis australis in southern Australia

The southern calamary, Sepioteuthis australis, is a commercially and recreationally important inshore cephalopod endemic to southern Australia and New Zealand. Typical of other cephalopods, S. australis has a short life span, form nearshore spawning aggregations and undergo direct development. Such life history traits may restrict connectivity between spawning grounds creating highly structured and genetically differentiated populations that are susceptible to population crashes. Here we use seven polymorphic microsatellite markers to assess connectivity and population structure of S. australis across a large part of its geographic range in Australia. Little genetic differentiation was found between sampling locations. Overall, FST was low (0.005, 95% CI≤<0.001-0.011) and we detected no significant genetic differentiation between any of the locations sampled. There was no strong relationship between genetic and geographical distance, and our neighbour joining analysis did not show clustering of clades based on geographical locations. Similarly, network analysis showed strong connectivity amongst most locations, in particular, Tasmania appears to be well connected with several other locations and may act as an important source population. High levels of gene flow and connectivity between S. australis sampling sites across Australia are important for this short-lived species, ensuring resilience against spatial and temporal mortality fluctuations.

Increased corticolimbic connectivity in cocaine dependence versus pathological gambling is associated with drug severity and emotion-related impulsivity

Neural biomarkers for the active detrimental effects of cocaine dependence (CD) are lacking. Direct comparisons of brain connectivity in cocaine-targeted networks between CD and behavioural addictions (i.e. pathological gambling, PG) may be informative. This study therefore contrasted the resting-state functional connectivity networks of 20 individuals with CD, 19 individuals with PG and 21 healthy individuals (controls). Study groups were assessed to rule out psychiatric co-morbidities (except alcohol abuse and nicotine dependence) and current substance use or gambling (except PG). We first examined global connectivity differences in the corticolimbic reward network and then utilized seed-based analyses to characterize the connectivity of regions displaying between-group differences. We examined the relationships between seed-based connectivity and trait impulsivity and cocaine severity. CD compared with PG displayed increased global functional connectivity in a large-scale ventral corticostriatal network involving the orbitofrontal cortex, caudate, thalamus and amygdala. Seed-based analyses showed that CD compared with PG exhibited enhanced connectivity between the orbitofrontal and subgenual cingulate cortices and between caudate and lateral prefrontal cortex, which are involved in representing the value of decision-making feedback. CD and PG compared with controls showed overlapping connectivity changes between the orbitofrontal and dorsomedial prefrontal cortices and between amygdala and insula, which are involved in stimulus-outcome learning. Orbitofrontal-subgenual cingulate cortical connectivity correlated with impulsivity and caudate/amygdala connectivity correlated with cocaine severity. We conclude that CD is linked to enhanced connectivity in a large-scale ventral corticostriatal-amygdala network that is relevant to decision making and likely to reflect an active cocaine detrimental effect.