Context-free and context-sensitive dynamics in recurrent neural networks

Continuous-valued recurrent neural networks can learn mechanisms for processing context-free languages. The dynamics of such networks is usually based on damped oscillation around fixed points in state space and requires that the dynamical components are arranged in certain ways. It is shown that qualitatively similar dynamics with similar constraints hold for a(n)b(n)c(n), a context-sensitive language. The additional difficulty with a(n)b(n)c(n), compared with the context-free language a(n)b(n), consists of 'counting up' and 'counting down' letters simultaneously. The network solution is to oscillate in two principal dimensions, one for counting up and one for counting down. This study focuses on the dynamics employed by the sequential cascaded network, in contrast to the simple recurrent network, and the use of backpropagation through time. Found solutions generalize well beyond training data, however, learning is not reliable. The contribution of this study lies in demonstrating how the dynamics in recurrent neural networks that process context-free languages can also be employed in processing some context-sensitive languages (traditionally thought of as requiring additional computation resources). This continuity of mechanism between language classes contributes to our understanding of neural networks in modelling language learning and processing.

Stochastic structure and individual-tree growth models

The majority of past and current individual-tree growth modelling methodologies have failed to characterise and incorporate structured stochastic components. Rather, they have relied on deterministic predictions or have added an unstructured random component to predictions. In particular, spatial stochastic structure has been neglected, despite being present in most applications of individual-tree growth models. Spatial stochastic structure (also called spatial dependence or spatial autocorrelation) eventuates when spatial influences such as competition and micro-site effects are not fully captured in models. Temporal stochastic structure (also called temporal dependence or temporal autocorrelation) eventuates when a sequence of measurements is taken on an individual-tree over time, and variables explaining temporal variation in these measurements are not included in the model. Nested stochastic structure eventuates when measurements are combined across sampling units and differences among the sampling units are not fully captured in the model. This review examines spatial, temporal, and nested stochastic structure and instances where each has been characterised in the forest biometry and statistical literature. Methodologies for incorporating stochastic structure in growth model estimation and prediction are described. Benefits from incorporation of stochastic structure include valid statistical inference, improved estimation efficiency, and more realistic and theoretically sound predictions. It is proposed in this review that individual-tree modelling methodologies need to characterise and include structured stochasticity. Possibilities for future research are discussed. (C) 2001 Elsevier Science B.V. All rights reserved.

Weak-force detection with superposed coherent states

We investigate the utility of nonclassical states of simple harmonic oscillators, particularly a superposition of coherent states, for sensitive force detection. We find that like squeezed states, a superposition of coherent states allows displacement measurements at the Heisenberg limit. Entangling many superpositions of coherent states offers a significant advantage over a single-mode superposition state with the same mean photon number.

Effect of shear stress on expression of a recombinant protein by Chinese hamster ovary cells

A flow chamber was used to impart a steady laminar shear stress on a recombinant Chinese hamster ovary (CHO) cell line expressing human growth hormone (hGH). The cells were subjected to shear stress ranging from 0.005 to 0.80 N m(-2). The effect of shear stress on the cell specific glucose uptake, cell specific hGH, and lactate productivity rates were calculated. No morphological changes to the cells were observed over the range of shear stresses examined. When the cells were subjected to 0.10 N m(-2) shear in protein-free media without Pluronic F-68, recombinant protein production ceased with no change in cell morphology, whereas control cultures were expressing hGH at 0.35 mug/10(6) cells/h. Upon addition of the shear protectants, Pluronic F-68 (0.2% [w/v]) or fetal bovine serum (1.0% [v/v] FBS), the productivity of the cells was restored. The effect of increasing shear stress on the cells in protein-free medium containing Pluronic F-68 was also investigated. Cell specific metabolic rates were calculated for cells under shear stress and for no-shear control cultures performed in parallel, with shear stress rates expressed as a percentage of those obtained for control cultures. Upon increasing shear from 0.005 to 0.80 N m(-2), the cell specific hGH productivity decreased from 100% at 0.005 N m(-2) to 49% at 0.80 N m(-2) relative to the no-shear control. A concurrent increase in the glucose uptake rate from 115% at 0.01 N m(-2) to 142% at 0.80 N m(-2), and decreased lactate productivity from 92% to 50%, revealed a change in the yield of products from glucose compared with the static control. It was shown that shear stress, at sublytic levels in medium containing Pluronic F-68, could decrease hGH specific productivity. (C) 2002 Wiley Periodicals, Inc.

Deconvolving the delta O-18 seawater component from subseasonal coral delta O-18 and Sr/Ca at Rarotonga in the southwestern subtropical Pacific for the period 1726 to 1997

To reconstruct oceanographic variations in the subtropical South Pacific, 271-year long subseasonal time series of Sr/Ca and delta(18)O were generated from a coral growing at Rarotonga (21.5degreesS, 159.5degreesW). In this case, coral Sr/Ca appears to be an excellent proxy for sea surface temperature (SST) and coral delta(18)O is a function of both SST and seawater delta(18)O composition (delta(18)O(sw)). Here, we focus on extracting the delta(18)O(sw) signal from these proxy records. A method is presented assuming that coral Sr/Ca is solely a function of SST and that coral delta(18)O is a function of both SST and delta(18)O(sw). This method separates the effects of delta(18)O(sw) from SST by breaking the instantaneous changes of coral delta(18)O into separate contributions by instantaneous SST and delta(18)O(sw) changes, respectively. The results show that on average delta(18)O(sw) at Rarotonga explains similar to39% of the variance in delta(18)O and that variations in SST explains the remaining similar to61% of delta(18)O variance. Reconstructed delta(18)O(sw) shows systematic increases in summer months (December-February) consistent with the regional pattern of variations in precipitation and evaporation. The delta(18)O(sw) also shows a positive linear correlation with satellite-derived estimated salinity for the period 1980 to 1997 (r = 0.72). This linear correlation between reconstructed delta(18)O(sw) and salinity makes it possible to use the reconstructed delta(18)O(sw) to estimate the past interannual and decadal salinity changes in this region. Comparisons of coral delta(18)O and delta(18)O(sw) at Rarotonga with the Pacific decadal oscillation index suggest that the decadal and interdecadal salinity and SST variability at Rarotonga appears to be related to basin-scale decadal variability in the Pacific. Copyright (C) 2002 Elsevier Science Ltd.

Causes and consequences of long-term climatic variability on the Australian continent

1. Ice-volume forced glacial-interglacial cyclicity is the major cause of global climate variation within the late Quaternary period. Within the Australian region, this variation is expressed predominantly as oscillations in moisture availability. Glacial periods were substantially drier than today with restricted distribution of mesic plant communities, shallow or ephemeral water bodies and extensive aeolian dune activity. 2. Superimposed on this cyclicity in Australia is a trend towards drier and/or more variable climates within the last 350 000 years. This trend may have been initiated by changes in atmospheric and ocean circulation resulting from Australia's continued movement into the Southeast Asian region and involving the onset or intensification of the El Nino-Southern Oscillation system and a reduction in summer monsoon activity. 3. Increased biomass burning, stemming originally from increased climatic variability and later enhanced by activities of indigenous people, resulted in a more open and sclerophyllous vegetation, increased salinity and a further reduction in water availability. 4. Past records combined with recent observations suggest that the degree of environmental variability will increase and the drying trend will be enhanced in the foreseeable future, regardless of the extent or nature of human intervention.

Low symbiont diversity in southern Great Barrier Reef corals, relative to those of the Caribbean

The specific identity of endosymbiotic dinoflagellates (Symbiodinium spp.) from most zooxanthellate corals is unknown. In a survey of symbiotic cnidarians from the southern Great Barrier Reef (GBR), 23 symbiont types were identified from 86 host species representing 40 genera. A majority (>85%) of these symbionts belong to a single phylogenetic clade or subgenus (C) composed of closely related (as assessed by sequence data from the internal transcribed spacer region and the ribosomal large subunit gene), yet ecologically and physiologically distinct, types. A few prevalent symbiont types, or generalists, dominate the coral community of the southern GBR, whereas many rare and/or specific symbionts, or specialists, are found uniquely within certain host taxa. The comparison of symbiont diversity between southern GBR and Caribbean reefs shows an inverse relationship between coral diversity and symbiont diversity, perhaps as a consequence of more-rapid diversification of Caribbean symbionts. Among clade C types, generalists C1 and C3 are common to both Caribbean and southern GBR symbiont assemblages, whereas the rest are regionally endemic. Possibly because of environmental changes in the Caribbean after geographic isolation through the Quaternary period, a high proportion of Caribbean fauna associate with symbiont taxa from two other distantly related Symbiodinium clades (A and B) that rarely occur in Pacific hosts. The resilience of Porites spp. and the resistance of Montipora digitata to thermal stress and bleaching are partially explained by their association with a thermally tolerant symbiont type, whereas the indiscriminant widespread bleaching and death among certain Pacific corals, during El Nino Southern Oscillation events, are influenced by associations with symbionts possessing higher sensitivity to thermal stress.

Late Quaternary Milankovitch-scale climatic change and variability and its impact on monsoonal Australasia

Four pollen and charcoal records derived from marine cores around the northern perimeter of Australia are examined to provide a regional picture of patterns, causes and impacts of climate change over the last 100-300 ka. The availability of radiocarbon dates and oxygen isotope records for the cores provides primary chronological control. Spectral analysis of components of these records demonstrates an overall importance of Milankovitch frequencies with clear glacial-interglacial cyclicity dominated by variation in precipitation. In addition, a number of pollen taxa, as well as charcoal particles, exhibit a 30 ka frequency that is considered, from its relationship with biomass burning and with results of past modelling, to reflect changes in the intensity of El Nino-Southern Oscillation (ENSO) variability. Pollen components of all records show a decline, frequently stepwise, in more fire-sensitive vegetation and its replacement with more fire-tolerant vegetation. There is some evidence that this trend is linked to an onset or general increase in ENSO activity and perhaps also to variation in monsoon activity dating from about 300 ka BP that was caused by changes to oceanic circulation within the Indonesian region. The trend may have accelerated within the last 45 ka due to burning by indigenous people. (C) 2003 Elsevier B.V. All rights reserved.

History of vegetation and habitat change in the Austral-Asian region

Over 1000 marine and terrestrial pollen diagrams and Some hundreds of vertebrate faunal sequences have been studied in the Austral-Asian region bisected by the PEPII transect, from the Russian arctic extending south through east Asia, Indochina, southern Asia, insular Southeast Asia (Sunda), Melanesia, Australasia (Sahul) and the western south Pacific. The majority of these records are Holocene but sufficient data exist to allow the reconstruction of the changing biomes over at least the past 200,000 years. The PEPII transect is free of the effects of large northern ice caps yet exhibits vegetational change in glacial cycles of a similar scale to North America. Major processes that can be discerned are the response of tropical forests in both lowlands and uplands to glacial cycles, the expansion of humid vegetation at the Pleistocene-Holocene transition and the change in faunal and vegetational controls as humans occupy the region. There is evidence for major changes in the intensity of monsoon and El Nino-Southern oscillation variability both on glacial-interglacial and longer time scales with much of the region experiencing a long-term trend towards more variable and/or drier climatic conditions. Temperature variation is most marked in high latitudes and high altitudes with precipitation providing the major climate control in lower latitude, lowland areas. At least some boundary shifts may be the response of vegetation to changing CO2 levels in the atmosphere. Numerous questions of detail remain, however, and current resolution is too coarse to examine the degree of synchroneity of millennial scale change along the transect. (C) 2003 Elsevier Ltd and INQUA. All rights reserved.

Anharmonic effects on a phonon-number measurement of a quantum-mesoscopic-mechanical oscillator

We generalize a proposal for detecting single-phonon transitions in a single nanoelectromechanical system (NEMS) to include the intrinsic anharmonicity of each mechanical oscillator. In this scheme two NEMS oscillators are coupled via a term quadratic in the amplitude of oscillation for each oscillator. One NEMS oscillator is driven and strongly damped and becomes a transducer for phonon number in the other measured oscillator. We derive the conditions for this measurement scheme to be quantum limited and find a condition on the size of the anharmonicity. We also derive the relation between the phase diffusion back-action noise due to number measurement and the localization time for the measured system to enter a phonon-number eigenstate. We relate both these time scales to the strength of the measured signal, which is an induced current proportional to the position of the read-out oscillator.

Risk attitude, planting conditions and the value of seasonal forecasts to a dryland wheat grower

The value of a seasonal forecasting system based on phases of the Southern Oscillation was estimated for a representative dryland wheat grower in the vicinity of Goondiwindi. In particular the effects on this estimate of risk attitude and planting conditions were examined. A recursive stochastic programming approach was used to identify the grower's utility-maximising action set in the event of each of the climate patterns over the period 1894-1991 recurring In the imminent season. The approach was repeated with and without use of the forecasts. The choices examined were, at planting, nitrogen application rate and cultivar and, later in the season, choices of proceeding with or abandoning each wheat activity, The value of the forecasting system was estimated as the maximum amount the grower could afford to pay for its use without expected utility being lowered relative to its non use.

Classical and quantum signatures of competing chi((2)) nonlinearities

We report the observation of the quantum effects of competing chi((2)) nonlinearities. We also report classical signatures of competition, namely, clamping of the second-harmonic power and production of nondegenerate frequencies in the visible. Theory is presented that describes the observations as resulting from competition between various chi((2)) up-conversion and down-conversion processes. We show that competition imposes hitherto unsuspected limits to both power generation and squeezing. The observed signatures are expected to be significant effects in practical systems.

Local dispersion of the Eucalyptus leaf beetle Chrysophtharta bimaculata (Coleoptera: Chrysomelidae), and implications for forest protection

1. Chrysophtharta bimaculata is a native chrysomelid species that can cause chronic defoliation of plantation and regrowth Eucalyptus forests in Tasmania, Australia. Knowledge of the dispersion pattern of C. bimaculata was needed in order to assess the efficiency of an integrated pest management (IPM) programme currently used for its control. 2. Using data from yellow flight traps, local populations of C. bimaculata adults were monitored over a season at spatial scales relevant to commercial forestry: within a 50-ha operational management unit (a forestry 'coupe') and between coupes. In addition, oviposition was monitored over a season at a subset of the between-coupe sites. 3. Dispersion indices (Taylor's Power Law and Iwao's Mean Crowding regression method) demonstrated that C. bimaculata adults were spatially aggregated within and between coupes, although the number of egg-batches laid at the between-coupe scale was uniform. Spatial autocorrelation analysis showed that trap-catches at the within-coupe level were similar (positively autocorrelated) to a radius distance of approximately 110 m, and then dissimilar (negatively autocorrelated) at approximately 250 m. At the between-coupe scale, no repeatable spatial autocorrelation patterns were observed. 4. For any individual site, rapid changes in beetle density were observed to be associated with loosely aggregated flights of beetles into and out of that site. Peak adult catches (> the weekly mean plus standard deviation trap-catch) for a site occurred for a period of 2.0 +/- 0.22 weeks at a time (n = 37), with normally only one or two peaks per site per season. Peak oviposition events for a site occurred on average 1.4 +/- 0.11 times per season and lasted 1.5 +/- 0.12 weeks. 5. Analysis of an extensive data set (n = 417) demonstrated that adult abundance at a site was positively correlated with egg density, but negatively correlated with tree damage (caused by conspecifics) and the presence of conspecific larvae. There was no relationship between adult abundance and a visual estimate of the amount of young foliage on trees. 6. Adults of C. bimaculata are show n to occur in relatively small, mobile aggregations. This means that pest surveys must be both regular (less than 2 weeks apart) and intensive (with sampling points no more than 150 m apart) if beetle populations are to be monitored with confidence. Further refinement of the current IPM strategy must recognize the problems posed by this temporal and spatial patchiness, particularly with regard to the use of biological insecticides, such as Bacillus thuringiensis, for which only a very short operational window exists.

Forecasting regional crop production using SOI phases: an example for the Australian peanut industry

Using peanuts as an example, a generic methodology is presented to forward-estimate regional crop production and associated climatic risks based on phases of the Southern Oscillation Index (SOI). Yield fluctuations caused by a highly variable rainfall environment are of concern to peanut processing and marketing bodies. The industry could profitably use forecasts of likely production to adjust their operations strategically. Significant, physically based lag-relationships exist between an index of ocean/atmosphere El Nino/Southern Oscillation phenomenon and future rainfall in Australia and elsewhere. Combining knowledge of SOI phases in November and December with output from a dynamic simulation model allows the derivation of yield probability distributions based on historic rainfall data. This information is available shortly after planting a crop and at least 3-5 months prior to harvest. The study shows that in years when the November-December SOI phase is positive there is an 80% chance of exceeding average district yields. Conversely, in years when the November-December SOI phase is either negative or rapidly falling there is only a 5% chance of exceeding average district yields, but a 95% chance of below average yields. This information allows the industry to adjust strategically for the expected volume of production. The study shows that simulation models can enhance SOI signals contained in rainfall distributions by discriminating between useful and damaging rainfall events. The methodology can be applied to other industries and regions.