Calibration and bias correction of climate projections for crop modelling: an idealised case study over Europe

Producing projections of future crop yields requires careful thought about the appropriate use of atmosphere-ocean global climate model (AOGCM) simulations. Here we describe and demonstrate multiple methods for ‘calibrating’ climate projections using an ensemble of AOGCM simulations in a ‘perfect sibling’ framework. Crucially, this type of analysis assesses the ability of each calibration methodology to produce reliable estimates of future climate, which is not possible just using historical observations. This type of approach could be more widely adopted for assessing calibration methodologies for crop modelling. The calibration methods assessed include the commonly used ‘delta’ (change factor) and ‘nudging’ (bias correction) approaches. We focus on daily maximum temperature in summer over Europe for this idealised case study, but the methods can be generalised to other variables and other regions. The calibration methods, which are relatively easy to implement given appropriate observations, produce more robust projections of future daily maximum temperatures and heat stress than using raw model output. The choice over which calibration method to use will likely depend on the situation, but change factor approaches tend to perform best in our examples. Finally, we demonstrate that the uncertainty due to the choice of calibration methodology is a significant contributor to the total uncertainty in future climate projections for impact studies. We conclude that utilising a variety of calibration methods on output from a wide range of AOGCMs is essential to produce climate data that will ensure robust and reliable crop yield projections.

Do variations in substitution rates and male mutation bias correlate with life-history traits? A study of 32 mammalian genomes

Life-history traits vary substantially across species, and have been demonstrated to affect substitution rates. We compute genomewide, branch-specific estimates of male mutation bias (the ratio of male-to-female mutation rates) across 32 mammalian genomes and study how these vary with life-history traits (generation time, metabolic rate, and sperm competition). We also investigate the influence of life-history traits on substitution rates at unconstrained sites across a wide phylogenetic range. We observe that increased generation time is the strongest predictor of variation in both substitution rates (for which it is a negative predictor) and male mutation bias (for which it is a positive predictor). Although less significant, we also observe that estimates of metabolic rate, reflecting replication-independent DNA damage and repair mechanisms, correlate negatively with autosomal substitution rates, and positively with male mutation bias. Finally, in contrast to expectations, we find no significant correlation between sperm competition and either autosomal substitution rates or male mutation bias. Our results support the important but frequently opposite effects of some, but not all, life history traits on substitution rates. KEY WORDS: Generation time, genome evolution, metabolic rate, sperm competition.

Domestic and foreign bias in global real estate mutual fund investment allocations

The purpose of the study is to seek a better understanding of the investment allocation behaviour of the real estate mutual funds by focusing on asset allocation at the country level. Analysing the country allocation of 553 real estate mutual funds domiciled in 20 countries, we attempt to trace how investment bias exists across countries and affects their country allocations. Our results evidence the existence of disproportionate country allocation to their domestic markets (domestic bias) and to each foreign market (foreign bias). We also find each bias is influenced by different sets of variables: real estate market influences for domestic bias and familiarity influences for foreign bias. This difference in factors influential for each bias in part explains the conflated relationship between the two biases.

Using seasonal hindcasts to understand the origin of the equatorial cold tongue bias in CGCMs and its impact on ENSO

The cold equatorial SST bias in the tropical Pacific that is persistent in many coupled OAGCMs severely impacts the fidelity of the simulated climate and variability in this key region, such as the ENSO phenomenon. The classical bias analysis in these models usually concentrates on multi-decadal to centennial time series needed to obtain statistically robust features. Yet, this strategy cannot fully explain how the models errors were generated in the first place. Here, we use seasonal re-forecasts (hindcasts) to track back the origin of this cold bias. As such hindcasts are initialized close to observations, the transient drift leading to the cold bias can be analyzed to distinguish pre-existing errors from errors responding to initial ones. A time sequence of processes involved in the advent of the final mean state errors can then be proposed. We apply this strategy to the ENSEMBLES-FP6 project multi-model hindcasts of the last decades. Four of the five AOGCMs develop a persistent equatorial cold tongue bias within a few months. The associated systematic errors are first assessed separately for the warm and cold ENSO phases. We find that the models are able to reproduce either El Niño or La Niña close to observations, but not both. ENSO composites then show that the spurious equatorial cooling is maximum for El Niño years for the February and August start dates. For these events and at this time of the year, zonal wind errors in the equatorial Pacific are present from the beginning of the simulation and are hypothesized to be at the origin of the equatorial cold bias, generating too strong upwelling conditions. The systematic underestimation of the mixed layer depth in several models can also amplify the growth of the SST bias. The seminal role of these zonal wind errors is further demonstrated by carrying out ocean-only experiments forced by the AOCGCMs daily 10-meter wind. In a case study, we show that for several models, this forcing is sufficient to reproduce the main SST error patterns seen after 1 month in the AOCGCM hindcasts.

Assessment of surface winds over the Atlantic, Indian and Pacific Ocean sectors of the Southern Ocean in CMIP5 models: historical bias, forcing response, and state dependence

An assessment of the fifth Coupled Models Intercomparison Project (CMIP5) models’ simulation of the near-surface westerly wind jet position and strength over the Atlantic, Indian and Pacific sectors of the Southern Ocean is presented. Compared with reanalysis climatologies there is an equatorward bias of 3.7° (inter-model standard deviation of ± 2.2°) in the ensemble mean position of the zonal mean jet. The ensemble mean strength is biased slightly too weak, with the largest biases over the Pacific sector (-1.6±1.1 m/s, 27 -22%). An analysis of atmosphere-only (AMIP) experiments indicates that 41% of the zonal mean position bias comes from coupling of the ocean/ice models to the atmosphere. The response to future emissions scenarios (RCP4.5 and RCP8.5) is characterized by two phases: (i) the period of most rapid ozone recovery (2000-2049) during which there is insignificant change in summer; and (ii) the period 2050-2098 during which RCP4.5 simulations show no significant change but RCP8.5 simulations show poleward shifts (0.30, 0.19 and 0.28°/decade over the Atlantic, Indian and Pacific sectors respectively), and increases in strength (0.06, 0.08 and 0.15 m/s/decade respectively). The models with larger equatorward position biases generally show larger poleward shifts (i.e. state dependence). This inter-model relationship is strongest over the Pacific sector (r=-0.89) and insignificant over the Atlantic sector (r=-0.50). However, an assessment of jet structure shows that over the Atlantic sector jet shift is significantly correlated with jet width whereas over the Pacific sector the distance between the sub-polar and sub-tropical westerly jets appears to be more important.

Climate spectrum estimation in the presence of timescale errors.

We introduce an algorithm (called REDFITmc2) for spectrum estimation in the presence of timescale errors. It is based on the Lomb-Scargle periodogram for unevenly spaced time series, in combination with the Welch's Overlapped Segment Averaging procedure, bootstrap bias correction and persistence estimation. The timescale errors are modelled parametrically and included in the simulations for determining (1) the upper levels of the spectrum of the red-noise AR(1) alternative and (2) the uncertainty of the frequency of a spectral peak. Application of REDFITmc2 to ice core and stalagmite records of palaeoclimate allowed a more realistic evaluation of spectral peaks than when ignoring this source of uncertainty. The results support qualitatively the intuition that stronger effects on the spectrum estimate (decreased detectability and increased frequency uncertainty) occur for higher frequencies. The surplus information brought by algorithm REDFITmc2 is that those effects are quantified. Regarding timescale construction, not only the fixpoints, dating errors and the functional form of the age-depth model play a role. Also the joint distribution of all time points (serial correlation, stratigraphic order) determines spectrum estimation.

Conditionally unbiased estimation in phase II/III clinical trials with early stopping for futility

Seamless phase II/III clinical trials combine traditional phases II and III into a single trial that is conducted in two stages, with stage 1 used to answer phase II objectives such as treatment selection and stage 2 used for the confirmatory analysis, which is a phase III objective. Although seamless phase II/III clinical trials are efficient because the confirmatory analysis includes phase II data from stage 1, inference can pose statistical challenges. In this paper, we consider point estimation following seamless phase II/III clinical trials in which stage 1 is used to select the most effective experimental treatment and to decide if, compared with a control, the trial should stop at stage 1 for futility. If the trial is not stopped, then the phase III confirmatory part of the trial involves evaluation of the selected most effective experimental treatment and the control. We have developed two new estimators for the treatment difference between these two treatments with the aim of reducing bias conditional on the treatment selection made and on the fact that the trial continues to stage 2. We have demonstrated the properties of these estimators using simulations

Assessing levels of uncertainty in recent temperature time series

We examine to what degree we can expect to obtain accurate temperature trends for the last two decades near the surface and in the lower troposphere. We compare temperatures obtained from surface observations and radiosondes as well as satellite-based measurements from the Microwave Soundings Units (MSU), which have been adjusted for orbital decay and non-linear instrument-body effects, and reanalyses from the European Centre for Medium-Range Weather Forecasts (ERA) and the National Centre for Environmental Prediction (NCEP). In regions with abundant conventional data coverage, where the MSU has no major influence on the reanalysis, temperature anomalies obtained from microwave sounders, radiosondes and from both reanalyses agree reasonably. Where coverage is insufficient, in particular over the tropical oceans, large differences are found between the MSU and either reanalysis. These differences apparently relate to changes in the satellite data availability and to differing satellite retrieval methodologies, to which both reanalyses are quite sensitive over the oceans. For NCEP, this results from the use of raw radiances directly incorporated into the analysis, which make the reanalysis sensitive to changes in the underlying algorithms, e.g. those introduced in August 1992. For ERA, the bias-correction of the one-dimensional variational analysis may introduce an error when the satellite relative to which the correction is calculated is biased itself or when radiances change on a time scale longer than a couple of months, e.g. due to orbit decay. ERA inhomogeneities are apparent in April 1985, October/November 1986 and April 1989. These dates can be identified with the replacements of satellites. It is possible that a negative bias in the sea surface temperatures (SSTs) used in the reanalyses may have been introduced over the period of the satellite record. This could have resulted from a decrease in the number of ship measurements, a concomitant increase in the importance of satellite-derived SSTs, and a likely cold bias in the latter. Alternately, a warm bias in SSTs could have been caused by an increase in the percentage of buoy measurements (relative to deeper ship intake measurements) in the tropical Pacific. No indications for uncorrected inhomogeneities of land surface temperatures could be found. Near-surface temperatures have biases in the boundary layer in both reanalyses, presumably due to the incorrect treatment of snow cover. The increase of near-surface compared to lower tropospheric temperatures in the last two decades may be due to a combination of several factors, including high-latitude near-surface winter warming due to an enhanced NAO and upper-tropospheric cooling due to stratospheric ozone decrease.

Bias or rationality? The case of UK commercial real estate investment

Purpose – The purpose of this paper is to test the hypothesis that investment decision making in the UK direct property market does not conform to the assumption of economic rationality underpinning portfolio theory. Design/methodology/approach – The developing behavioural real estate paradigm is used to challenge the idea that investor “man” is able to perform with economic rationality, specifically with reference to the analysis of the spatial dispersion of the entire UK “investible stock” and “investible locations” against observed spatial patterns of institutional investment. Location quotients are derived, combining different data sets. Findings – Considerably greater variation in institutional property holdings is found across the UK than would be expected given the economic and stock characteristics of local areas. This appears to provide evidence of irrationality (in the strict traditional economic sense) in the behaviour of institutional investors, with possible herding underpinning levels of investment that cannot be explained otherwise. Research limitations/implications – Over time a lack of distinction has developed between the cause and effect of comparatively low levels of development and institutional property investment across the regions. A critical examination of decision making and behaviour in practice could break this cycle, and could in turn promote regional economic growth. Originality/value – The entire “population” of observations is used to demonstrate the relationships between economic theory and investor performance exploring, for the first time, stock and local area characteristics.

Optimal estimation of sea surface temperature from split-window observations

Optimal estimation (OE) improves sea surface temperature (SST) estimated from satellite infrared imagery in the “split-window”, in comparison to SST retrieved using the usual multi-channel (MCSST) or non-linear (NLSST) estimators. This is demonstrated using three months of observations of the Advanced Very High Resolution Radiometer (AVHRR) on the first Meteorological Operational satellite (Metop-A), matched in time and space to drifter SSTs collected on the global telecommunications system. There are 32,175 matches. The prior for the OE is forecast atmospheric fields from the Météo-France global numerical weather prediction system (ARPEGE), the forward model is RTTOV8.7, and a reduced state vector comprising SST and total column water vapour (TCWV) is used. Operational NLSST coefficients give mean and standard deviation (SD) of the difference between satellite and drifter SSTs of 0.00 and 0.72 K. The “best possible” NLSST and MCSST coefficients, empirically regressed on the data themselves, give zero mean difference and SDs of 0.66 K and 0.73 K respectively. Significant contributions to the global SD arise from regional systematic errors (biases) of several tenths of kelvin in the NLSST. With no bias corrections to either prior fields or forward model, the SSTs retrieved by OE minus drifter SSTs have mean and SD of − 0.16 and 0.49 K respectively. The reduction in SD below the “best possible” regression results shows that OE deals with structural limitations of the NLSST and MCSST algorithms. Using simple empirical bias corrections to improve the OE, retrieved minus drifter SSTs are obtained with mean and SD of − 0.06 and 0.44 K respectively. Regional biases are greatly reduced, such that the absolute bias is less than 0.1 K in 61% of 10°-latitude by 30°-longitude cells. OE also allows a statistic of the agreement between modelled and measured brightness temperatures to be calculated. We show that this measure is more efficient than the current system of confidence levels at identifying reliable retrievals, and that the best 75% of satellite SSTs by this measure have negligible bias and retrieval error of order 0.25 K.

Natural disaster accounting bias and its equivalence across genetic resource stocks

Two sources of bias arise in conventional loss predictions in the wake of natural disasters. One source of bias stems from neglect of accounting for animal genetic resource loss. A second source of bias stems from failure to identify, in addition to the direct effects of such loss, the indirect effects arising from implications impacting animal-human interactions. We argue that, in some contexts, the magnitude of bias imputed by neglecting animal genetic resource stocks is substantial. We show, in addition, and contrary to popular belief, that the biases attributable to losses in distinct genetic resource stocks are very likely to be the same. We derive the formal equivalence across the distinct resource stocks by deriving an envelope result in a model that forms the mainstay of enquiry in subsistence farming and we validate the theory, empirically, in a World-Society-for-the-Protection-of-Animals application

Interpretation bias in preschool children at risk for anxiety: a prospective study

A story-stem paradigm was used to assess interpretation bias in preschool children. Data were available for 131 children. Interpretation bias, behavioural inhibition (BI) and anxiety were assessed when children were aged between 3 years 2 months and 4 years 5 months. Anxiety was subsequently assessed 12-months, 2-years and 5-years later. A significant difference in interpretation bias was found between participants who met criteria for an anxiety diagnosis at baseline, with clinically anxious participants more likely to complete the ambiguous story-stems in a threat-related way. Threat interpretations significantly predicted anxiety symptoms at 12-month follow-up, after controlling for baseline symptoms, but did not predict anxiety symptoms or diagnoses at either 2-year or 5- year follow-up. There was little evidence for a relationship between BI and interpretation bias. Overall, the pattern of results was not consistent with the hypothesis that interpretation bias plays a role in the development of anxiety. Instead, some evidence for a role in the maintenance of anxiety over relatively short periods of time was found. The use of a story-stem methodology to assess interpretation bias in young children is discussed along with the theoretical and clinical implications of the findings.

There’s that scary picture: attention bias to threatening scenes in Williams syndrome

There is increasing evidence that Williams syndrome (WS) is associated with elevated anxiety that is non-social in nature, including generalised anxiety and fears. To date very little research has examined the cognitive processes associated with this anxiety. In the present research, attentional bias for non-social threatening images in WS was examined using a dot-probe paradigm. Participants were 16 individuals with WS aged between 13 and 34 years and two groups of typically developing controls matched to the WS group on chronological age and attentional control ability respectively. The WS group exhibited a significant attention bias towards threatening images. In contrast, no bias was found for group matched on attentional control and a slight bias away from threat was found in the chronological age matched group. The results are contrasted with recent findings suggesting that individuals with WS do not show an attention bias for threatening faces and discussed in relation to neuroimaging research showing elevated amygdala activation in response to threatening non-social scenes in WS.

I see happy people: attention bias towards happy but not angry facial expressions in Williams syndrome

Introduction: Observations of behaviour and research using eye-tracking technology have shown that individuals with Williams syndrome (WS) pay an unusual amount of attention to other people’s faces. The present research examines whether this attention to faces is moderated by the valence of emotional expression. Method: Sixteen participants with WS aged between 13 and 29 years (Mean=19 years 9 months) completed a dot-probe task in which pairs of faces displaying happy, angry and neutral expressions were presented. The performance of the WS group was compared to two groups of typically developing control participants, individually matched to the participants in the WS group on either chronological age or mental age. General mental age was assessed in the WS group using the Woodcock Johnson Test of Cognitive Ability Revised (WJ-COG-R; Woodcock & Johnson, 1989; 1990). Results: Compared to both control groups, the WS group exhibited a greater attention bias for happy faces. In contrast, no between-group differences in bias for angry faces were obtained. Conclusions: The results are discussed in relation to recent neuroimaging findings and the hypersocial behaviour that is characteristic of the WS population.