Stable estimations for extreme wind speeds. An application to Belgium

Generalized Pareto distributions (GPD) are frequently applied for the statistical analysis of extreme wind speeds. A central topic in extreme-value theory is the adaptive estimation of the extreme-value index ??. Several authors have demonstrated a high sensitivity of ?? against the threshold when analyzing extreme wind speeds. This undesirable effect introduces the difficulty to provide reliable quantile estimates. This paper aims to bring this problem to meteorologists and proposes a stable estimator (the Zipf estimator) for ??. This could allow a more objective prior identification of the sign and range of ??. The method is based on regression in the so-called generalized quantile plots. A comparative study with a classical estimator (the probability-weighted method) is made and it is shown that the Zipf estimator significantly decreases the variance in the calibration of the GPD to extreme wind gusts. Finally, the new methodology is applied to get improved prediction of extreme wind gusts in Belgium.     

Changes in extreme wind speeds in NW Europe simulated by generalized linear models

Summary We investigate the capability of generalized linear models (GLMs) to simulate sequences of daily maximum wind speed (DMWS), at a selection of locations in NW Europe. Models involving both the gamma and Weibull distributions have been fitted to the NCEP reanalysis data for the period 1958–1998. In simulations, these models successfully reproduce the observed increasing trends up to 0.3 m/s per decade in coastal or oceanic locations for the wintertime and the decreasing trends down to –0.2 m/s per decade in inland Europe for the summertime. Annually extreme winds exhibit an increasing tendency (with median estimates up to 0.6 m/s per decade) at the studied locations. The gamma model slightly overestimates the upper percentiles of the wind speed distribution, but reproduces trends better than the Weibull model. In both the NCEP data and GLM simulations, local extreme DMWS events (defined in terms of threshold exceedances) have increased dramatically in frequency during winter; decreasing trends are more common in summer. The NCEP data indicate similar trends in the frequencies of large-scale windy events (defined via simultaneous exceedances at 2 or more locations). Overall, these events have increased in number; at the scale of the North Sea basin, their number may have changed from 3–5 days per year during the earlier decades, to 5–7 days per year during later decades based on observational estimates. An increase in the frequency of large-scale extreme winter storms is implied. The GLMs underestimate these large-scale event frequencies, and provide imprecise estimates of the corresponding secular trends. These problems could be rectified by using a better representation of spatial dependence. The present results suggest that GLMs offer a useful tool to study local climate extremes in the context of changing climate distributions; they also provide some pointers towards improving the representation of extremes at a regional scale.     

Analyses of possible changes in intense and extreme wind speeds over northern Europe under climate change scenarios

Dynamical downscaling of ECHAM5 using HIRHAM5 and RCA3 for a northern European domain focused on Scandinavia indicates sustained extreme wind speeds with long recurrence intervals (50?years) and intense winds are not likely to evolve out of the historical envelope of variability until the end of C21st. Even then, significant changes are indicated only in the SW of the domain and across the central Baltic Sea where there is some evidence for relatively small magnitude increases in the 50?year return period wind speed (of up to 15%). There are marked differences in results based on the two Regional Climate Models. Additionally, internal (inherent) variability and initial conditions exert a strong impact on projected wind climates throughout the twenty-first century. Simulations of wind gusts by one of the RCMs (RCA3) indicate some evidence for increased magnitudes (of up to +10%) in the southwest of the domain and across the central Baltic Sea by the end of the current century. As in prior downscaling of ECHAM4, dynamical downscaling of ECHAM5 indicates a tendency towards increased energy density and thus wind power generation potential over the course of the C21st. However, caution should be used in interpreting this inference given the high degree of wind climate projection spread that derives from the specific AOGCM and RCM used in the downscaling.     

河北省太阳总辐射经验系数两种气候学计算方法的比较分析

利用河北省周边8个日射站逐月日照百分率资料和太阳总辐射资料,采用最小二乘法拟合经验系数a、b,然后分别通过纬度分区和反距离权重插值两种方法得到河北省142个测站的经验系数a、b,并据此求出河北省内四个具有辐射观测数据台站的总辐射值,对比分析了实测值与不同经验系数下总辐射估算值之间的差异。结果表明：纬度分区和反距离权重插值两种方法所建立的太阳总辐射量估算公式的模拟精度总体差别不大,但采用纬度分区法所得的经验系数,在计算太阳辐射年总量时与实际观测值更为接近,因此建议采用纬度分区法计算河北省各地太阳总辐射量。     

Extreme wind speeds in the European sector of the Arctic

Variations in extreme wind speed over the European part of the Arctic are studied from the data of meteorological observations, reanalysis, and modeling based on the INM CM4 climate model. It is demonstrated that the extremes determined from the observational data are a mixture of two datasets well simulated by the Weibull distribution. According to the special metaphoric terminology, they are called “black swans” and “dragons.” The analysis of extreme wind speeds based on the reanalysis and INM CM4 data demonstrated that they consist of “black swans” only. This important fact indicates that the models (at least those with medium horizontal resolution) are not able to simulate some essential circulation mechanisms causing the formation of significant anomalies of wind speed. Hence, the problem of direct identification of wind speed extremes based on the atmospheric modeling remains open.     

Assessing the vulnerability of wind energy to climate change and extreme events

This article presents a review of the status and basis of wind-generated electricity production, the state of knowledge regarding possible changes in the spatio-temporal characteristics of the wind resource and wind turbine operating conditions, the principal extreme events that are of relevance to the wind energy industry, and the major potential vulnerabilities of the wind energy industry to climate change, with a specific focus on extreme events. Generally, the magnitude of projected changes over Europe and the contiguous USA are within the ‘conservative’ estimates embedded within the Wind Turbine Design Standards. However, more research is needed to quantify (i) how global climate evolution may influence the operation of wind turbines outside these regions, (ii) events causing coincident extreme wind speeds, gusts, and vertical wind shear, and (iii) combined wind-wave loading on offshore turbines.     

Improved stove programs need robust methods to estimate carbon offsets

Current standard methods result in significant discrepancies in carbon offset accounting compared to approaches based on representative community based subsamples, which provide more realistic assessments at reasonable cost. Perhaps more critically, neither of the currently approved methods incorporates uncertainties inherent in estimates of emission factors or non-renewable fuel usage (fNRB). Since emission factors and fNRB contribute 25% and 47%, respectively, to the overall uncertainty in offset estimates for Purépecha communities in Mexico, exclusion of this uncertainty is a critical omission. When the recommended uncertainty for default emission factors and the uncertainty associated with regional estimates of fNRB are included the lower 95% confidence intervals of both Clean Development Mechanism and Gold Standard methods exceed the total amount of carbon saved, which would result in zero marketable carbon savings if approaches recommended by the IPCC Good Practice Guidance and Uncertainty Management in National Greenhouse Gas Inventories, or Land use, Land-Use Change and Forestry (LULUCF) are to be followed. In contrast, for the same communities, methods using representative subsamples of emission factors and fuel consumption, combined with community-level fNRB estimates, result in significant carbon offsets with a lower 95% confidence interval of 2.3 tCO₂e home^???1 year^???1. Given the misleading estimates, revision of the currently approved methodologies to provide robust estimates of carbon offsets is strongly recommended.     

风向的统计方法研究

在核安全分析和环境影响评价中,必须使用到的是小时气象数据。对自动气象站小时风向的计算方法和小时数据如何统计,地面气象规范中和核安全导则中还没有定论,因此本文对小时风向统计方法展开讨论。目前小时风向值的计算方法有算术平均法、滑动平均法、矢量平均法和频率最高法,针对算术平均法和滑动平均法对经过0°的风向统计容易出现误差,本文提出对此的修正方法并对4种计算方法进行了比较。结果表明：文中过零风向修正方法简便准确,小时数据统计方法为正点前10min数据时,修正算术平均法更可靠,但该方法对风速为零时的判断容易出现误差,因此在小风、静风频率高的地方推荐矢量平均法。关于小时数据的划分方法,美国核管会RG1.23与我国核安全导则及地面气象规范中的规定不同,因此文中利用实测资料对不同小时数据统计方法所得结果比较,分析表明,取整点前或其他时段的10min和15min的数据进行平均的风向相关矩阵一致性为97.87%;取4个15min平均值的平均或6个10min数据平均值的平均作为小时值的风向相关矩阵一致性为99.96%,这两种统计方法与取10min和15min的一致性为86.00%,相对较差;取60min时段的平均值作为小时值则与其余方法一致性最差。     

在空气污染研究中应用风廓线参数之探讨

本文使用对数律和幂指数律公式,计算了年平均和不同稳定度、不同风速等级下的丘陵山地的风指数(α)和综合粗糙度(Z_0)值,对公式及计算结果的适用性进行了探讨。结果表明,对于地面边界层(SBL),此两公式在丘陵山区条件下均有较好的适用性。可利用某地年平均地面风速(V_(10))和数次实测烟囱口高度的风速(V_烟),应用对数律或幂指数律推算该地各高度上的风速。对于丘陵山区,对数律可适用于地面以上300m层次内。     

A statistical approach to climatology

Summary After introductory remarks on the manyfold difficulties in climatological work the author attempts a more precise definition of Climatology, which is often called the science of the average weather, as the science of the multivariate distributions of meteorological elements with time and space. A climatological study has to be made in various stemps, each of which has to keep constant a minor number of variables than the preceding one. In the selection of the mathematical models one has to pay attention to two principles, which often oppose each other: simplicity and generality. The goodness of fit of the model has to be proved. In proceeding to higher levels of the climatological analysis. Studies should be undertaken by means of some stations with long series of data with regard to the necessary length of the period of observations. The author is convinced that all these costly investigations would be of utmos value for humanity.

---

Zusammenfassung Es werden zunächst Beispiele für die mannigfachen Schwierigkeiten der klimatologischen Arbeit angeführt, wie sie z. B. durch unenheitliche Definitionen und Repräsentationsarten, heterogene Beobachtungsreihen, baobachtungslücken u. ä. m. verursacht werden. Dann wird die Definition der Klimatologie als Wissenschaft vom durchschnittlichen Wetter präzisiert als Wissenschaft, die die multivariablen statistischen Verteilungen der meteorologischen Elemente mit Orts- und Zeitkoordinaten zu studieren hat. Eine klimatologische Studie muß mit Teilstudien beginnen, die zunächst einige der Variablen konstant halten; das Resultat dieser Teilstudien muß Wegweiser sein beim Fortschreiten zu höheren Dimensionen. Bei der Auswahl der mathematischen Modelle in allen Stufen ist auf swei Prinzipien Rücksicht zu nehmen, die einander widersprechen können: das Modell soll einfach sein, soll aber doch zu verallgemeinern sein. Die Güte der Anpassung des Modells andie Daten muß geprüft werden, z. B. durch die Methoden der Konfidenzintervalle oder der Subperioden. Beim Fortschreiten zu höheren Modellen sind die Methoden der Analyse von einfachen oder multiplen Zeitreihen zu berücksichtigen. Für die Frage einer hinreichend langen Bearbeitungsperiode sollten Untersuchungen aller Stufen mit einer Anzahl von Schlüsselstationen durchgeführt werden. Der Autor ist der Ansicht, daß sich alle die angeführten kostspieligen Untersuchungen für die Allgemeinheit bezahlt machen werden.

---

Résumé L'auteur présente tout d'abord quelques exemples des nombreuses difficulté inhérentes à tout travail en climatologie. Ces difficulté proviennent soit dufait que les définitions et les formes de représentation des qu'il y ait des lacunes dans les séries d'observations, soit d'autres causes encore. Il définit ensuite la climatologie comme étant la science du temps moyen et plus particulièrement la science chargée d'étudier la répartition statistique multiple des éléments météorologiques et cela auusi bien dans le temps que dans l'espace. Une recherche en climatologie doit débuter par des études partielles c'est à dire ne laissant, au préalable, varier que quelques éléments du climat, les autres restant constants. Les résultats obtenus lors de ces études partielles indiqueront alors dans quelles directions poursuivre. Lors du choix des modèles mathématiques, il faut tenir compte, à tous les niveaux, de deux principes qui peuvent être contradictoires: Le modèle doit être simple, mais il doit permettre des généralisations. Il faut étudier les possibilités d'adaptation du modèle choisi aux séries de chiffres à étudier en utilisant par exemple des périodes secondaires ou l'intervalle situé entre les limites de confiance. En passant à des modèles de degré supérieur, il est nécessaire de tenir compte des méthodes d'analyse de séries chronologiques simples et multiples. Pour déterminer quelle longueur au moins doit avoir une période d'observations, il faudrait étudier à tous les degrés un contrôle avec un certain nombre de stations typiques. L'auteur pense que toutes ces recherches sont rentables pour la collectivité, même si leur coût est élevé.

---

---

With 2 Figures

---

Lecture given at the IV. Session of the Commission for Climatology of WMO, August 18, 1965 in Stockholm.     

Spatiotemporal long-term trends of extreme wind characteristics over the Black Sea

In this study, long-term change of wind characteristics on the Black Sea has been investigated using two widely used data sources, i.e., European Centre for Medium-Range Weather Forecasts (ECMWF) ERA-Interim and National Centers for Environmental Prediction/Climate Forecast System Reanalysis (NCEP/CFSR), spanning 40 years between 1979 and 2018. Spatial and seasonal variability of climatic features such as the wind speed, direction, number and duration of storms, and wind power density are discussed. Wind climate is characterized by strong, durable and stable winds in the northern and western Black Sea, and relatively weak, short-lived and highly-variable winds in the eastern Black Sea. These long-term wind patterns indicate that the eastern part of the basin is likely to be subjected to the impacts of climate change. Long-term stable and strong wind conditions in the southwest part indicate reliable, persistent and sustainable wind energy potential. Long-term and seasonal variation of wind power density (WPD) at 110 m altitude over the Black Sea is investigated. There is a significant difference in WPD values between winter and summer seasons, with around 2.8 times larger WPD in winter than that in summer. In the western Black Sea, narrow confidence intervals observed in each season indicate a low level of variation during a season and ensures stable wind power conditions.     

应用泊松耿贝尔分布估算海上极端风速

采用泊松耿贝尔分布,基于中国气象局发布的《CMA-STI热带气旋最佳路径数据集》资料来估算海上极端风速,并以上川岛气象站多年实测资料通过概率评估来加以验证。结果表明,采用泊松耿贝尔分布可以得到较为保守的海上极值风速取值。对于海上的小面积区域的重现期风速估算,采用50 km半径区域进行评估,可以得到较为合理的估算结果,当评估区域较大时,则需考虑适当扩大评估半径。     

把脉极端天气气候事件,构建预测方法

<正>极端天气气候事件是造成我国重大自然灾害的主要根源,并且在近年来有加重的趋势。暴雨、台风、持续性强降温、暴雪、高温热浪、重霾污染等都是全社会高度关注的,科学家们需要剖析形成这些极端天气气候事件的原因和科学过程、机理,并研究构建有效的科学预测方法,以达到防灾减灾的目的。     

Regional effects of large-scale extreme wind events over orographically structured terrain

Summary ?Above orographically structured terrain considerable differences of the regional wind field may be identified during large-scale extreme wind events. So far, these regional differences could not be resolved by climate models. To determine the relationships between large-scale atmospheric conditions, the influence of orography, and the regional wind field, data measured in the upper Rhine valley within the framework of the REKLIP Regional Climate Project were analyzed and calculations were made using the KAMM mesoscale model. In the area of the upper Rhine valley, ratios of the wind velocity in the Rhine valley at 10 m above ground level, ν_val, and the large-scale flow velocity, ν_lar, are between ν_val/ν_lar ≈ 0.1 and ν_val/ν_lar ≈ 1. The ν_val/ν_lar ratio exhibits a strong dependence on thermal stratification, δ, and decreases from ν_val/ν_lar ≈ 1 at δ = 0 K m⁻¹ to ν_val/ν_lar ≈ 0.2 at δ = 0.0075 K m⁻¹. In areas, where the lateral mountainous border of the Rhine valley is interrupted, the ν_val/ν_lar ratio increases again with increasing stability or decreasing Froude number. This is obviously due to flow around the Black Forest under stable stratification. It is demonstrated by model calculations that a complex wind field develops in the Rhine valley at small Froude numbers (Fr < 1) irrespective of the direction of large-scale flow. The ν_val/ν_lar ratio is characterized by small values in the direct lee side (ν_val/ν_lar ≈ 0.2) and high values on the windward side of the lateral mountainous border of the Rhine valley (ν_val/ν_lar ≈ 0.8). Received October 22, 2001; revised June 18, 2002; accepted June 23, 2002     

日最高温度统计降尺度方法的比较研究

针对日最高温度的降尺度问题,发展了一种统计降尺度的新方法——优选格点回归法(OPR)。利用该方法与双线性插值法(BI)对平原(山东)和高原(云南)的日最高温度进行降尺度对比分析,结果表明:无论对于平原(山东)还是高原(云南)地区以及夏季(7月)还是冬季(1月),OPR方法都明显优于BI方法,特别是从高原地区的均方根误差来看,降尺度效果优势更加明显。进一步对OPR方法降尺度过程中所做的方差放大对比分析显示,方差放大后对日最高温度的降尺度效果不但没有改进,在某些方面如均方根误差和极端误差等还有变差的表现。     

A cautionary note on automated statistical downscaling methods for climate change

The urge for higher resolution climate change scenarios has been widely recognized, particularly for conducting impact assessment studies. Statistical downscaling methods have shown to be very convenient for this task, mainly because of their lower computational requirements in comparison with nested limited-area regional models or very high resolution Atmosphere–ocean General Circulation Models. Nevertheless, although some of the limitations of statistical downscaling methods are widely known and have been discussed in the literature, in this paper it is argued that the current approach for statistical downscaling does not guard against misspecified statistical models and that the occurrence of spurious results is likely if the assumptions of the underlying probabilistic model are not satisfied. In this case, the physics included in climate change scenarios obtained by general circulation models, could be replaced by spatial patterns and magnitudes produced by statistically inadequate models. Illustrative examples are provided for monthly temperature for a region encompassing Mexico and part of the United States. It is found that the assumptions of the probabilistic models do not hold for about 70 % of the gridpoints, parameter instability and temporal dependence being the most common problems. As our examples reveal, automated statistical downscaling “black-box” models are to be considered as highly prone to produce misleading results. It is shown that the Probabilistic Reduction approach can be incorporated as a complete and internally consistent framework for securing the statistical adequacy of the downscaling models and for guiding the respecification process, in a way that prevents the lack of empirical validity that affects current methods.