Identifying periods of forecast model confidence for improved subseasonal prediction of precipitation in southern Australia

2020 ◽  
Author(s):  
Doug Richardson ◽  
James Risbey ◽  
Didier Monselesan
Keyword(s):  
A Priori ◽  
Forecast Model ◽  
Reanalysis Data ◽  
Lead Times ◽  
Conditional Distributions ◽  
Atmospheric Variability ◽  
Data Set ◽  
Typical Application ◽  
Model Skill ◽  
Circulation Patterns

<p>Subseasonal prediction skill of precipitation is typically low. Sometimes, however, forecasts are accurate and it would be useful to end-users to assess <em>a priori</em> if this might be the case. We use a 20-year hindcast data set of the ECMWF S2S prediction system and identify periods of high forecast confidence, evaluating model skill of precipitation forecasts for these periods compared to lower confidence predictions.</p><p>From reanalysis data, we derive a set of circulation patterns, called archetypes, that represent the broad-scale atmospheric circulation over Australia. These archetypes are combinations of ridges and troughs, and yield different precipitation patterns depending on the location of these features. In the literature, a typical application of circulation patterns is assigning daily reanalysis fields to the closest-matching pattern, thus obtaining conditional distributions of precipitation corresponding to key modes of atmospheric variability. A problem common to such analyses is that the precipitation distributions associated with the circulation patterns can be too similar; distinct distributions are required in order for the patterns to be useful in estimating precipitation. We show that by subsampling the archetype occurrences only when they are particularly well-matched to the underlying field, the conditional precipitation distributions become more distinct.</p><p>We subsample hindcast fields in the same way, obtaining a sample of periods when the model is confident about its prediction of the upcoming archetype. We then calculate model skill in predicting precipitation for three regions in southern Australia during such periods compared to when the model is not confident about the predicted archetype. Our results suggest that during periods of forecast confidence, precipitation skill is greater than normal for shorter leads (up to ten days) in two of the three regions (the Murray Basin and Western Tasmania). Skill for the third region (Southwest Western Australia) is greater during confident periods for lead times greater than one week, although this is marginal.</p>

scholarly journals Multistep-ahead daily inflow forecasting using the ERA-Interim reanalysis data set based on gradient-boosting regression trees

2020 ◽  
Vol 24 (5) ◽  
pp. 2343-2363
Author(s):  
Shengli Liao ◽  
Zhanwei Liu ◽  
Benxi Liu ◽  
Chuntian Cheng ◽  
Xinfeng Jin ◽  
...  
Keyword(s):  
Extreme Values ◽  
Regression Trees ◽  
Reanalysis Data ◽  
Gradient Boosting ◽  
Support Vector ◽  
Lead Times ◽  
Feature Subset ◽  
Data Set ◽  
Inflow Forecasting ◽  
Maximal Information Coefficient

Abstract. Inflow forecasting plays an essential role in reservoir management and operation. The impacts of climate change and human activities have made accurate inflow prediction increasingly difficult, especially for longer lead times. In this study, a new hybrid inflow forecast framework – using the ERA-Interim reanalysis data set as input and adopting gradient-boosting regression trees (GBRT) and the maximal information coefficient (MIC) – is developed for multistep-ahead daily inflow forecasting. Firstly, the ERA-Interim reanalysis data set provides more information for the framework, allowing it to discover inflow for longer lead times. Secondly, MIC can identify an effective feature subset from massive features that significantly affects inflow; therefore, the framework can reduce computational burden, distinguish key attributes from unimportant ones and provide a concise understanding of inflow. Lastly, GBRT is a prediction model in the form of an ensemble of decision trees, and it has a strong ability to more fully capture nonlinear relationships between input and output at longer lead times. The Xiaowan hydropower station, located in Yunnan Province, China, was selected as the study area. Six evaluation criteria, namely the mean absolute error (MAE), the root-mean-squared error (RMSE), the Pearson correlation coefficient (CORR), Kling–Gupta efficiency (KGE) scores, the percent bias in the flow duration curve high-segment volume (BHV) and the index of agreement (IA) are used to evaluate the established models utilizing historical daily inflow data (1 January 2017–31 December 2018). The performance of the presented framework is compared to that of artificial neural network (ANN), support vector regression (SVR) and multiple linear regression (MLR) models. The results indicate that reanalysis data enhance the accuracy of inflow forecasting for all of the lead times studied (1–10 d), and the method developed generally performs better than other models, especially for extreme values and longer lead times (4–10 d).

scholarly journals A global ozone climatology from ozone soundings via trajectory mapping: a stratospheric perspective

2013 ◽  
Vol 13 (22) ◽  
pp. 11441-11464 ◽  
Author(s):  
J. Liu ◽  
D. W. Tarasick ◽  
V. E. Fioletov ◽  
C. McLinden ◽  
T. Zhao ◽  
...  
Keyword(s):  
Satellite Data ◽  
Lower Stratosphere ◽  
Stratospheric Ozone ◽  
A Priori ◽  
Forecast Model ◽  
Reanalysis Data ◽  
Trajectory Mapping ◽  
Decadal Variations ◽  
Mean Differences ◽  
The 1960S

Abstract. This study explores a domain-filling trajectory approach to generate a global ozone climatology from relatively sparse ozonesonde data. Global ozone soundings comprising 51 898 profiles at 116 stations over 44 yr (1965–2008) are used, from which forward and backward trajectories are calculated from meteorological reanalysis data to map ozone measurements to other locations and so fill in the spatial domain. The resulting global ozone climatology is archived monthly for five decades from the 1960s to the 2000s on a grid of 5° × 5° × 1 km (latitude, longitude, and altitude), from the surface to 26 km altitude. It is also archived yearly for the same period. The climatology is validated at 20 selected ozonesonde stations by comparing the actual ozone sounding profile with that derived through trajectory mapping of ozone sounding data from all stations except the one being compared. The two sets of profiles are in good agreement, both overall with correlation coefficient r = 0.991 and root mean square (RMS) of 224 ppbv and individually with r from 0.975 to 0.998 and RMS from 87 to 482 ppbv. The ozone climatology is also compared with two sets of satellite data from the Satellite Aerosol and Gas Experiment (SAGE) and the Optical Spectrography and InfraRed Imager System (OSIRIS). The ozone climatology compares well with SAGE and OSIRIS data in both seasonal and zonal means. The mean differences are generally quite small, with maximum differences of 20% above 15 km. The agreement is better in the Northern Hemisphere, where there are more ozonesonde stations, than in the Southern Hemisphere; it is also better in the middle and high latitudes than in the tropics where reanalysis winds are less accurate. This ozone climatology captures known features in the stratosphere as well as seasonal and decadal variations of these features. The climatology clearly shows the depletion of ozone from the 1970s to the mid 1990s and ozone increases in the 2000s in the lower stratosphere. When this climatology is used as the upper boundary condition in an Environment Canada operational chemical forecast model, the forecast is improved in the vicinity of the upper troposphere-lower stratosphere (UTLS) region. This ozone climatology is latitudinally, longitudinally, and vertically resolved and it offers more complete high latitude coverage as well as a much longer record than current satellite data. As the climatology depends on neither a priori data nor photochemical modeling, it provides independent information and insight that can supplement satellite data and model simulations of stratospheric ozone.

scholarly journals Identifying periods of forecast model confidence for improved subseasonal prediction of precipitation

2020 ◽  
Author(s):  
Doug Richardson ◽  
Amanda S. Black ◽  
Didier P. Monselesan ◽  
Thomas S. Moore ◽  
James S. Risbey ◽  
...  
Keyword(s):  
Western Australia ◽  
Forecast Model ◽  
Seasonal Prediction ◽  
Forecast Skill ◽  
Lead Times ◽  
Archetypal Analysis ◽  
Weather Forecasts ◽  
Atmospheric Circulation Patterns ◽  
Circulation Patterns ◽  
Murray Basin

AbstractSubseasonal forecast skill is not homogeneous in time, and prior assessment of the likely forecast skill would be valuable for end-users. We propose a method for identifying periods of high forecast confidence using atmospheric circulation patterns, with an application to southern Australia precipitation. In particular, we use archetypal analysis to derive six patterns, called archetypes, of daily 500 hPa geopotential height (Z500) fields over Australia. We assign Z500 reanalysis fields to the closest-matching archetype and subsequently link the archetypes to precipitation for three key regions in the Australian agriculture and energy sectors: the Murray Basin, Southwest Western Australia and Western Tasmania. Using a 20-year hindcast dataset from the European Centre for Medium-Range Weather Forecasts subseasonal-to-seasonal prediction system, we identify periods of high confidence as when hindcast Z500 fields closely match an archetype according to a distance criterion. We compare the precipitation hindcast accuracy during these confident periods compared to normal. Considering all archetypes, we show that there is greater skill during confident periods for lead times of less than 10 days in the Murray Basin and Western Tasmania, and for greater than six days in Southwest Western Australia, although these conclusions are subject to substantial uncertainty. By breaking down the skill results for each archetype individually, we highlight how skill tends to be greater than normal for those archetypes associated with drier-than-average conditions.

scholarly journals A global ozone climatology from ozone soundings via trajectory mapping: a stratospheric perspective

2013 ◽  
Vol 13 (6) ◽  
pp. 16831-16883
Author(s):  
J. Liu ◽  
D. W. Tarasick ◽  
V. E. Fioletov ◽  
C. McLinden ◽  
T. Zhao ◽  
...  
Keyword(s):  
Satellite Data ◽  
Lower Stratosphere ◽  
Stratospheric Ozone ◽  
A Priori ◽  
Correlation Coefficients ◽  
Forecast Model ◽  
Reanalysis Data ◽  
Trajectory Mapping ◽  
Decadal Variations ◽  
Mean Differences

Abstract. This study explores a domain-filling trajectory approach to generate a global ozone climatology from relatively sparse ozonesonde data. Global ozone soundings comprising 51 898 profiles at 116 stations over 44 yr (1965–2008) are used, from which forward and backward trajectories are calculated from meteorological reanalysis data, to map ozone measurements to other locations and so fill in the spatial domain. The resulting global ozone climatology is archived monthly for five decades from the 1960s to the 2000s on a~grid of 5° × 5° × 1 km (latitude, longitude, and altitude), from the surface to 26 km altitude. It is also archived yearly from 1965 to 2008. The climatology is validated at 20 selected ozonesonde stations by comparing the actual ozone sounding profile with that derived through trajectory mapping of ozone sounding data from all stations except the one being compared. The two sets of profiles are in good agreement, both individually with correlation coefficients (r) between 0.975 and 0.998 and root mean square (RMS) differences of 87 to 482 ppbv, and overall with r = 0.991 and an RMS of 224 ppbv. The ozone climatology is also compared with two sets of satellite data, from the Satellite Aerosol and Gas Experiment (SAGE) and the Optical Spectrography and InfraRed Imager System (OSIRIS). The ozone climatology compares well with SAGE and OSIRIS data in both seasonal and zonal means. The mean differences are generally quite small, with maximum differences of 20% above 15 km. The agreement is better in the Northern Hemisphere, where there are more ozonesonde stations, than in the Southern Hemisphere; it is also better in the middle and high latitudes than in the tropics where reanalysis winds are less accurate. This ozone climatology captures known features in the stratosphere, as well as seasonal and decadal variations of these features. Compared to current satellite data, it offers more complete high latitude coverage as well as a much longer record. The climatology shows clearly the depletion of ozone from the 1970s to the mid 1990s and ozone recovery in the lower stratosphere in the 2000s. When this climatology is used as the upper boundary condition in an Environment Canada operational chemical forecast model, the forecast is improved in the vicinity of the upper troposphere–lower stratosphere (UTLS) region. As this ozone climatology is neither dependent on a priori data nor photochemical modeling, it provides independent information and insight that can supplement satellite data and model simulations of stratospheric ozone.

scholarly journals A Traveler’s Guide to the Multiverse: Promises, Pitfalls, and a Framework for the Evaluation of Analytic Decisions

2021 ◽  
Vol 4 (1) ◽  
pp. 251524592095492
Author(s):  
Marco Del Giudice ◽  
Steven W. Gangestad
Keyword(s):  
Degrees Of Freedom ◽  
A Priori ◽  
Simulated Data ◽  
Style Analysis ◽  
Data Set ◽  
Scant Attention ◽  
Equivalence Type ◽  
The Impact ◽  
Biased Estimates

Decisions made by researchers while analyzing data (e.g., how to measure variables, how to handle outliers) are sometimes arbitrary, without an objective justification for choosing one alternative over another. Multiverse-style methods (e.g., specification curve, vibration of effects) estimate an effect across an entire set of possible specifications to expose the impact of hidden degrees of freedom and/or obtain robust, less biased estimates of the effect of interest. However, if specifications are not truly arbitrary, multiverse-style analyses can produce misleading results, potentially hiding meaningful effects within a mass of poorly justified alternatives. So far, a key question has received scant attention: How does one decide whether alternatives are arbitrary? We offer a framework and conceptual tools for doing so. We discuss three kinds of a priori nonequivalence among alternatives—measurement nonequivalence, effect nonequivalence, and power/precision nonequivalence. The criteria we review lead to three decision scenarios: Type E decisions (principled equivalence), Type N decisions (principled nonequivalence), and Type U decisions (uncertainty). In uncertain scenarios, multiverse-style analysis should be conducted in a deliberately exploratory fashion. The framework is discussed with reference to published examples and illustrated with the help of a simulated data set. Our framework will help researchers reap the benefits of multiverse-style methods while avoiding their pitfalls.

scholarly journals MAX-DOAS observations of aerosols, formaldehyde and nitrogen dioxide in the Beijing area: comparison of two profile retrieval approaches

2015 ◽  
Vol 8 (2) ◽  
pp. 941-963 ◽  
Author(s):  
T. Vlemmix ◽  
F. Hendrick ◽  
G. Pinardi ◽  
I. De Smedt ◽  
C. Fayt ◽  
...  
Keyword(s):  
Standard Deviation ◽  
A Priori ◽  
Systematic Bias ◽  
Aerosol Extinction ◽  
Data Set ◽  
Near Surface ◽  
Beijing Area ◽  
Least Squares Fit ◽  
Relative Differences ◽  
Good Agreement

Abstract. A 4-year data set of MAX-DOAS observations in the Beijing area (2008–2012) is analysed with a focus on NO2, HCHO and aerosols. Two very different retrieval methods are applied. Method A describes the tropospheric profile with 13 layers and makes use of the optimal estimation method. Method B uses 2–4 parameters to describe the tropospheric profile and an inversion based on a least-squares fit. For each constituent (NO2, HCHO and aerosols) the retrieval outcomes are compared in terms of tropospheric column densities, surface concentrations and "characteristic profile heights" (i.e. the height below which 75% of the vertically integrated tropospheric column density resides). We find best agreement between the two methods for tropospheric NO2 column densities, with a standard deviation of relative differences below 10%, a correlation of 0.99 and a linear regression with a slope of 1.03. For tropospheric HCHO column densities we find a similar slope, but also a systematic bias of almost 10% which is likely related to differences in profile height. Aerosol optical depths (AODs) retrieved with method B are 20% high compared to method A. They are more in agreement with AERONET measurements, which are on average only 5% lower, however with considerable relative differences (standard deviation ~ 25%). With respect to near-surface volume mixing ratios and aerosol extinction we find considerably larger relative differences: 10 ± 30, −23 ± 28 and −8 ± 33% for aerosols, HCHO and NO2 respectively. The frequency distributions of these near-surface concentrations show however a quite good agreement, and this indicates that near-surface concentrations derived from MAX-DOAS are certainly useful in a climatological sense. A major difference between the two methods is the dynamic range of retrieved characteristic profile heights which is larger for method B than for method A. This effect is most pronounced for HCHO, where retrieved profile shapes with method A are very close to the a priori, and moderate for NO2 and aerosol extinction which on average show quite good agreement for characteristic profile heights below 1.5 km. One of the main advantages of method A is the stability, even under suboptimal conditions (e.g. in the presence of clouds). Method B is generally more unstable and this explains probably a substantial part of the quite large relative differences between the two methods. However, despite a relatively low precision for individual profile retrievals it appears as if seasonally averaged profile heights retrieved with method B are less biased towards a priori assumptions than those retrieved with method A. This gives confidence in the result obtained with method B, namely that aerosol extinction profiles tend on average to be higher than NO2 profiles in spring and summer, whereas they seem on average to be of the same height in winter, a result which is especially relevant in relation to the validation of satellite retrievals.

Magnetotelluric static shift: Estimation and removal using the cokriging method

Geophysics ◽  
2007 ◽  
Vol 72 (1) ◽  
pp. F25-F34 ◽  
Author(s):  
Benoit Tournerie ◽  
Michel Chouteau ◽  
Denis Marcotte
Keyword(s):  
Apparent Resistivity ◽  
A Priori ◽  
Shift Factor ◽  
Static Shift ◽  
Geostatistical Model ◽  
Data Set ◽  
Resistivity Sounding ◽  
The Difference ◽  
Cross Variogram

We present and test a new method to correct for the static shift affecting magnetotelluric (MT) apparent resistivity sounding curves. We use geostatistical analysis of apparent resistivity and phase data for selected periods. For each period, we first estimate and model the experimental variograms and cross variogram between phase and apparent resistivity. We then use the geostatistical model to estimate, by cokriging, the corrected apparent resistivities using the measured phases and apparent resistivities. The static shift factor is obtained as the difference between the logarithm of the corrected and measured apparent resistivities. We retain as final static shift estimates the ones for the period displaying the best correlation with the estimates at all periods. We present a 3D synthetic case study showing that the static shift is retrieved quite precisely when the static shift factors are uniformly distributed around zero. If the static shift distribution has a nonzero mean, we obtained best results when an apparent resistivity data subset can be identified a priori as unaffected by static shift and cokriging is done using only this subset. The method has been successfully tested on the synthetic COPROD-2S2 2D MT data set and on a 3D-survey data set from Las Cañadas Caldera (Tenerife, Canary Islands) severely affected by static shift.

Character selection and the quantification of morphological disparity

Paleobiology ◽  
2016 ◽  
Vol 43 (1) ◽  
pp. 68-84 ◽  
Author(s):  
Bradley Deline ◽  
William I. Ausich
Keyword(s):  
A Priori ◽  
Morphological Diversity ◽  
Small Subset ◽  
Early Paleozoic ◽  
Data Set ◽  
Rarefaction Analysis ◽  
Morphological Studies ◽  
Body Regions ◽  
Scientific Hypotheses

AbstractA priori choices in the detail and breadth of a study are important in addressing scientific hypotheses. In particular, choices in the number and type of characters can greatly influence the results in studies of morphological diversity. A new character suite was constructed to examine trends in the disparity of early Paleozoic crinoids. Character-based rarefaction analysis indicated that a small subset of these characters (~20% of the complete data set) could be used to capture most of the properties of the entire data set in analyses of crinoids as a whole, noncamerate crinoids, and to a lesser extent camerate crinoids. This pattern may be the result of the covariance between characters and the characterization of rare morphologies that are not represented in the primary axes in morphospace. Shifting emphasis on different body regions (oral system, calyx, periproct system, and pelma) also influenced estimates of relative disparity between subclasses of crinoids. Given these results, morphological studies should include a pilot analysis to better examine the amount and type of data needed to address specific scientific hypotheses.

scholarly journals Advantages of using a fast urban canopy model as compared to a full mesoscale model to simulate the urban heat island of Barcelona

2016 ◽  
Author(s):  
M. García-Díez ◽  
D. Lauwaet ◽  
H. Hooyberghs ◽  
J. Ballester ◽  
K. De Ridder ◽  
...  
Keyword(s):  
Simple Model ◽  
Land Surface ◽  
Climate Model ◽  
Mesoscale Model ◽  
Urban Climate ◽  
Urban Canopy ◽  
Forecast Model ◽  
Reanalysis Data ◽  
Urban Canopy Model ◽  
Canopy Model

Abstract. As most of the population lives in urban environments, the simulation of the urban climate has become a key problem in the framework of the climate change impact assessment. However, the high computational power required by these simulations is a severe limitation. Here we present a study on the performance of a Urban Climate Model (UrbClim), designed to be several orders of magnitude faster than a full-fledge mesoscale model. The simulations are validated with station data and with land surface temperature observations retrieved by satellites. To explore the advantages of using a simple model like UrbClim, the results are compared with a simulation carried out with a state-of-the-art mesoscale model, the Weather Research and Forecasting model, using an Urban Canopy model. The effect of using different driving data is explored too, by using both relatively low resolution reanalysis data (70 km) and a higher resolution forecast model (15 km). The results show that, generally, the performance of the simple model is comparable to or better than the mesoscale model. The exception are the winds and the day-to-day correlation in the reanalysis driven run, but these problems disappear when taking the boundary conditions from the higher resolution forecast model.

scholarly journals Skill of Rain–Snow Level Forecasts for Landfalling Atmospheric Rivers: A Multimodel Assessment Using California’s Network of Vertically Profiling Radars

2020 ◽  
Vol 21 (4) ◽  
pp. 751-771 ◽  
Author(s):  
Brian Henn ◽  
Rachel Weihs ◽  
Andrew C. Martin ◽  
F. Martin Ralph ◽  
Tashiana Osborne
Keyword(s):  
Sierra Nevada ◽  
Lead Time ◽  
Flood Hazard ◽  
Critical Factor ◽  
Forecast Model ◽  
Lead Times ◽  
Grid Spacing ◽  
Atmospheric Rivers ◽  
Operational Forecasts ◽  
Resources Planning

AbstractThe partitioning of rain and snow during atmospheric river (AR) storms is a critical factor in flood forecasting, water resources planning, and reservoir operations. Forecasts of atmospheric rain–snow levels from December 2016 to March 2017, a period of active AR landfalls, are evaluated using 19 profiling radars in California. Three forecast model products are assessed: a global forecast model downscaled to 3-km grid spacing, 4-km river forecast center operational forecasts, and 50-km global ensemble reforecasts. Model forecasts of the rain–snow level are compared with observations of rain–snow melting-level brightband heights. Models produce mean bias magnitudes of less than 200 m across a range of forecast lead times. Error magnitudes increase with lead time and are similar between models, averaging 342 m for lead times of 24 h or less and growing to 700–800 m for lead times of greater than 144 h. Observed extremes in the rain–snow level are underestimated, particularly for warmer events, and the magnitude of errors increases with rain–snow level. Storms with high rain–snow levels are correlated with larger observed precipitation rates in Sierra Nevada watersheds. Flood risk increases with rain–snow levels, not only because a greater fraction of the watershed receives rain, but also because warmer storms carry greater water vapor and thus can produce heavier precipitation. The uncertainty of flood forecasts grows nonlinearly with the rain–snow level for these reasons as well. High rain–snow level ARs are a major flood hazard in California and are projected to be more prevalent with climate warming.

Sign in / Sign up

Export Citation Format

Share Document