Torben Schmith

Statistical and dynamical downscaling of precipitation: An evaluation and comparison of scenarios for the European Alps

[1] This paper compares six statistical downscaling models (SDMs) and three regional climate models (RCMs) in their ability to downscale daily precipitation statistics in a region of complex topography. The six SDMs include regression methods, weather typing methods, a conditional weather generator, and a bias correction and spatial disaggregation approach. The comparison is carried out over the European Alps for current and future (2071–2100) climate. The evaluation of simulated precipitation for the current climate shows that the SDMs and RCMs tend to have similar biases but that they differ with respect to interannual variations. The SDMs strongly underestimate the magnitude of the year-to-year variations. Clear differences emerge also with respect to the year-to-year anomaly correlation skill: In winter, over complex terrain, the better RCMs achieve significantly higher skills than the SDMs. Over flat terrain and in summer, the differences are smaller. Scenario results using A2 emissions show that in winter mean precipitation tends to increase north of about 45N and insignificant or opposite changes are found to the south. There is good agreement between the downscaling models for most precipitation statistics. In summer, there is still good qualitative agreement between the RCMs but large differences between the SDMs and between the SDMs and the RCMs. According to the RCMs, there is a strong trend toward drier conditions including longer periods of drought. The SDMs, on the other hand, show mostly nonsignificant or even opposite changes. Overall, the present analysis suggests that downscaling does significantly contribute to the uncertainty in regional climate scenarios, especially for the summer precipitation climate.

A Surrogate Ensemble Study of Climate Reconstruction Methods: Stochasticity and Robustness

Abstract Reconstruction of the earth’s surface temperature from proxy data is an important task because of the need to compare recent changes with past variability. However, the statistical properties and robustness of climate reconstruction methods are not well known, which has led to a heated discussion about the quality of published reconstructions. In this paper a systematic study of the properties of reconstruction methods is presented. The methods include both direct hemispheric-mean reconstructions and field reconstructions, including reconstructions based on canonical regression and regularized expectation maximization algorithms. The study will be based on temperature fields where the target of the reconstructions is known. In particular, the focus will be on how well the reconstructions reproduce low-frequency variability, biases, and trends. A climate simulation from an ocean–atmosphere general circulation model of the period a.d. 1500–1999, including both natural and anthropogenic forcings, is...

Fram Strait Ice Export during the Nineteenth and Twentieth Centuries Reconstructed from a Multiyear Sea Ice Index from Southwestern Greenland

Abstract Historical observations of multiyear ice, called “storis,” in the southwest Greenland waters exist from the period 1820–2000, obtained from ship logbooks and ice charts. It is argued that this ice originates in the Arctic Ocean and has traveled via the Fram Strait, southward along the Greenland coast in the East Greenland Current, and around the southern tip of Greenland. Therefore, it is hypothesized that these observations can be used as “proxies” for reconstructing the Fram Strait ice export on an annual basis. An index describing the storis extent is extracted from the observations and a linear statistical model formulated relating this index to the Fram Strait ice export. The model is calibrated using ice export values from a hindcast study with a coupled ocean–ice model over the period 1949–98. Subsequently, the model is used to reconstruct the Fram Strait annual ice export in the period 1820–2000. The model has significant skill, calculated on independent data. Based on this reconstruction...

A look at the ocean in the EC-Earth climate model

EC-Earth is a newly developed global climate system model. Its core components are the Integrated Forecast System (IFS) of the European Centre for Medium Range Weather Forecasts (ECMWF) as the atmosphere component and the Nucleus for European Modelling of the Ocean (NEMO) developed by Institute Pierre Simon Laplace (IPSL) as the ocean component. Both components are used with a horizontal resolution of roughly one degree. In this paper we describe the performance of NEMO in the coupled system by comparing model output with ocean observations. We concentrate on the surface ocean and mass transports. It appears that in general the model has a cold and fresh bias, but a much too warm Southern Ocean. While sea ice concentration and extent have realistic values, the ice tends to be too thick along the Siberian coast. Transports through important straits have realistic values, but generally are at the lower end of the range of observational estimates. Exceptions are very narrow straits (Gibraltar, Bering) which are too wide due to the limited resolution. Consequently the modelled transports through them are too high. The strength of the Atlantic meridional overturning circulation is also at the lower end of observational estimates. The interannual variability of key variables and correlations between them are realistic in size and pattern. This is especially true for the variability of surface temperature in the tropical Pacific (El Niño). Overall the ocean component of EC-Earth performs well and helps making EC-Earth a reliable climate model.

A Surrogate Ensemble Study of Sea Level Reconstructions

Abstract This study investigates the possibility of reconstructing past global mean sea levels. Reconstruction methods rely on historical measurements from tide gauges combined with knowledge about the spatial covariance structure of the sea level field obtained from a shorter period with spatially well-resolved satellite measurements. A surrogate ensemble method is applied based on sea levels from a 500-yr climate model simulation. Tide gauges are simulated by selecting time series from grid points along continental coastlines and on ocean islands. Reconstructions of global mean sea levels can then be compared to the known target, and the ensemble method allows an estimation of the statistical properties originating from the stochastic nature of the reconstructions. Different reconstruction methods previously used in the literature are studied, including projection and optimal interpolation methods based on EOF analysis of the calibration period. This study also includes methods where these EOFs are augm...

Comment on "A Semi-Empirical Approach to Projecting Future Sea-Level Rise"

Rahmstorf (Reports, 19 January 2007, p. 368) used the observed relation between rates of change of global surface temperature and sea level to predict future sea-level rise. We revisit the application of the statistical methods used and show that estimation of the regression coefficient is not robust. Methods commonly used within econometrics may be more appropriate for the problem of projected sea-level rise.

Stationarity of Regression Relationships: Application to Empirical Downscaling

Abstract The performance of a statistical downscaling model is usually evaluated for its ability to explain a large fraction of predictand variance. In this note, it is shown that although this fraction may be high, the longest time scales, including trends, may not be explained by the model. This implies that the model is nonstationary over the training period of the model, and it questions the basic stationarity assumption of statistical downscaling. This is exemplified by using a simple regression model for downscaling European precipitation and surface temperature where appropriate Monte Carlo–based field significance tests are developed, taking into account the intercorrelation between predictand series. Based on this test, it is concluded that care is needed in selecting predictors to avoid this form of nonstationarity. Even though this is illustrated for a simple regression-type statistical downscaling model, the main conclusions may also be valid for more complicated models.

Limitations on regression analysis due to serially correlated residuals: Application to climate reconstruction from proxies

[2] Studies of past climate can be based on reconstructions that rely on the capture of climate information in various properties of biological remnants and geological formations that remain accessible long after the record was made. Jones and Mann [2004] provide an overview of the field. The captured climate information can sometimes be recovered, but the quality of the extracted information depends on the record itself, the proxy, and the type of methods used to extract the information. This paper will look at one specific issue of this problem, related to the use of regression methods in the extraction process. [3] Typically an instrumental climate record exists for a location or region alongside proxy records, and the analysis method consists of finding the relationship between the two by regression: yt = a + bxt + ut, where ut is an error term.

How Much Have Variations in the Meridional Overturning Circulation Contributed to Sea Surface Temperature Trends since 1850? A Study with the EC-Earth Global Climate Model

The surface of the world’s oceans has been warming since the beginning of industrialization. In addition to this, multidecadal sea surface temperature (SST) variations of internal origin exist. Evidence suggests that the North Atlantic Ocean exhibits the strongest multidecadal SST variations and that these variations are connected to the overturning circulation. This work investigates the extent to which these internal multidecadal variations have contributed to enhancing or diminishing the trend induced by the external radiative forcing, globally and in the North Atlantic. A model study is carried out wherein the analyses of a long control simulation with constant radiative forcing at preindustrial level and of an ensemble of simulations with historical forcing from 1850 until 2005 are combined. First, it is noted that global SST trends calculated from the different historical simulations are similar, while there is a large disagreement between the North Atlantic SST trends. Then the control simulation is analyzed, where a relationship between SST anomalies and anomalies in the Atlantic meridional overturning circulation (AMOC) for multidecadal and longer time scales is identified. This relationship enables the extraction of the AMOC-related SST variability from each individual member of the ensemble of historical simulations and then the calculation of the SST trends with the AMOC-related variability excluded. For the global SST trends this causes only a little difference while SST trends with AMOC-related variability excluded for the North Atlantic show closer agreement than with the AMOC-related variability included. From this it is concluded that AMOC variability has contributed significantly to North Atlantic SST trends since the mid nineteenth century.

Comparing past and present climate – a tool to distinguish between natural and human-induced climate change

Variations in climate over the last approx. 2000 years and a contrast in the winter climate regime of NW Europe and West Greenland may be ascribed to variations in the teleconnection pattern expressed by the North Atlantic Oscillation (NAO). This well-known climate phenomenon is, among others, reflected in an increased inflow of warm Atlantic Water into the Labrador Sea under conditions of a colder European winter regime (negative NAO indices). Cooling of the Labrador Sea appears to have been a feature typical for a milder European winter climate (positive NAO indices). This pattern is seen both on shorter, (sub)decadal timescales of the last approx. 60 years and on a longer, centennial timescale through the last millennia. During a significant part of the 20th century warming, episodes of milder winters in Europe may be related to Atlantic patterns of natural climate variability similarly as recorded in the past millennia. However, records for recent decades (since the mid 1980s) show a different pattern indicating a significant change in North Atlantic ocean- and atmospheric circulations. This seemingly changed behaviour of the climate system corresponds in timing to the episode of the most extensive global temperature increase, suggesting an increasing anthropogenic effect on the climate development of the past ca. 20 years.