Principal Component Regression Techniques Research Articles

Development and Validation of Analytical Methods for Simultaneous Estimation of Lercanidipine Hydrochloride and Enalapril Maleate

The fixed dose of lercanidipine hydrochloride(LH) and enalapril maleate (EM) is used to provide more effective control of hypertension. The research work was aimed at the development and validation of simple and sensitive analytical methods for their quantification from tablets. The simultaneous equation, first-order derivative, and multivariate spectrophotometric methods were developed for the simultaneous estimation of LH and EM in tablets. The simultaneous equation method involves a determination of LH and EM at 236 nm and 207 nm, respectively. First order derivative UV spectrophotometry method involves using a zero crossing point of LH at 330 nm and EM at 219 nm. Both methods were validated as per ICH Q2 (R2) guidelines and found to be accurate and precise as they exhibited <2% relative standard deviation. The limit of detection (LOD) and limit of quantitation (LOQ) for LH were found to be 0.26 μg/ml and 0.78 μg/ml respectively whereas LOD and LOQ for EM were found to be 0.41 μg/ml and 1.24 μg/ml respectively for simultaneous equation method. The LOD and LOQ for LH were found to be 0.39 μg/ml and 1.19 μg/ml respectively whereas LOD and LOQ for EM were found to be 0.31 μg/ml and 0.97μg/ml respectively for the first-order derivative method. Three Chemometric methods namely classical least square (CLS), principal component regression (PCR), and partial least square (PLS) were studied for the simultaneous determination of LH and EM in tablets using spectrophotometry. A set of 25 standard mixtures containing both drugs were prepared in the range of 5–25 μg/ml for both drugs. The analytical figure of merit (FOM) was determined for all three chemometric methods. The LOD values for LH were found to be 0.97, 0.93, and 0.94 μg/ml and 0.90,0.88, and 0.89 μg/ml for EM using CLS, PCR, and PLS modeling techniques. The proposed methods were user-friendly, rapid, and sensitive enough for the determination of LH and EM.

Evaluation of some numerical techniques for determining paleotemperatures from planktonic foraminiferal assemblages

This study evaluates the results of various numerical techniques intended to estimate environmental variables from the composition of associations of organisms in sediment samples. An analysis is made of various transfer functions (species regression technique, 13 variants; weighted average technique, 8 variants; principal component regression technique, 270 variants; discriminant analysis technique, one procedure) and of the modern analog technique (11 variants). The initial point is the calibration data set composed of the associations of planktonic foraminifera contained in 605 core-top samples from the North Atlantic and the Mediterranean, together with the corresponding sea surface winter temperatures. To compare the results of the various techniques, the calibration data set is randomly divided into two subsets of 450 and 155 samples, respectively. The former is used as the provisional calibration data set; this is used to estimate the winter temperatures of the subset of 155 samples (the analog set) and of another set, formed of 27 samples from the South Atlantic and the Indian Ocean (no-analog set). After evaluating the errors of the estimate of the different techniques, the best ones are selected.

Water quality prediction and classification based on principal component regression and gradient boosting classifier approach

Estimating water quality has been one of the significant challenges faced by the world in recent decades. This paper presents a water quality prediction model utilizing the principal component regression technique. Firstly, the water quality index (WQI) is calculated using the weighted arithmetic index method. Secondly, the principal component analysis (PCA) is applied to the dataset, and the most dominant WQI parameters have been extracted. Thirdly, to predict the WQI, different regression algorithms are used to the PCA output. Finally, the Gradient Boosting Classifier is utilized to classify the water quality status. The proposed system is experimentally evaluated on a Gulshan Lake-related dataset. The results demonstrate 95% prediction accuracy for the principal component regression method and 100% classification accuracy for the Gradient Boosting Classifier method, which show credible performance compared with the state-of-art models.

Studies on population dynamics and management of pomegranate aphid, Aphis punicae Passerini (Hemiptera: Aphididae) on pomegranate under semi-arid conditions of South-western Punjab

Abstract Pomegranate aphid (Aphis punicae) is an important pest of the pomegranate. Seasonality of A. punicae and its relationship with weather factors were observed during 2015 and 2016 at the Regional Research Station, Abohar of Punjab Agricultural University. We analyses temporal patterns in A. punicae abundance, and also highlights the factors influencing its abundance. Principal Component Regression Analysis (PCRA) was used to find out the relationship between abundance of A. punicae and the climatic variables. The population build-up of A. punicae peaked in 12th SMW (160.25 aphids / 10 cm twig). The optimum range of temperature (Tm 21.44 °C – 30.77 °C; Tmn 7.52 °C–15.72 °C) and humidity (RHm 97.20%–98.64%; RHmn 44.39%–39.17%) for the multiplication of A. punicae population was depicted. Among different cultivars of pomegranate, Bhagwa was most and Phule Arkata was least susceptible to A. punicae infestation. The prediction model formulated based on maximum, minimum temperature, maximum minimum relative humidity, radiation and wind speed using Principal component regression technique performed well with reasonably accuracy (R2 = 0.88) to forecast A. punicae population on the pomegranate plants. Regarding the management of A. punicae in field the different pesticides tested gave 42.33–96.60% reduction of A. punicae population. The results were persisted upto 15 days of the treatment. The greatest control of A. punicae was achieved by flonicamid > thiamethoxam > imidacloprid > Azadirachtin> HMO > Rogar. Both Azadiractin and HMO were found least harmful to predators of A. punicae with an average reduction of 14.27–16.67 % at different dosages after 14 days of treatment. Rogar, imidacloprid, flonicamid and thiamethoxam was found high to moderately toxic to predators with 38.46–64.28% reduction of coccinelid larvae/adults. These results suggested that A. punicae population dynamics on pomegranate was greatly influenced by climatic variables and host nutrition and both azadirachtin and HMOs can be considered as potential weapons for controlling A. punicae as they are significantly less toxic to A. punicae predators.

Validated chemometrics-assisted spectrophotometric methods for simultaneous determination of bisphenol-A-diglycidyl ether and some of its reaction products in canned foods in the Egyptian market

Two multivariate calibration methods, namely principal component regression (PCR) and partial least squares (PLS-2) have been developed, validated and compared for the simultaneous determination of bisphenol-A-diglycidyl ether (BADGE) and some of its reaction products, including BADGE·HCl·H2O, BADGE·H2O and BADGE·2HCl. Chemometrics allowed rapid, accurate and precise simultaneous quantification of the analytes of interest which was not possible by other spectrophotometric methods due to their severe spectral overlap. PCR and PLS-2 techniques successfully quantified BADGE, BADGE·HCl·H2O, BADGE·H2O and BADGE·2HCl in the ranges of 1.4–3.4, 1–5, 1–4.2 and 1–7 μg mL−1, respectively. The constructed models were validated according to the International Conference on Harmonization guidelines and successfully applied for the determination of these compounds in pure form, laboratory prepared mixtures and in various types of canned foods following the limits and regulations of the European Union (EU) where satisfactory recovery results were obtained.

Deriving Global Convection Maps From SuperDARN Measurements

AbstractA new statistical modeling technique for determining the global ionospheric convection is described. The principal component regression (PCR)‐based technique is based on Super Dual Auroral Radar Network (SuperDARN) observations and is an advanced version of the PCR technique that Waters et al. (, https//:doi.org.10.1002/2015JA021596) used for the SuperMAG data. While SuperMAG ground magnetic field perturbations are vector measurements, SuperDARN provides line‐of‐sight measurements of the ionospheric convection flow. Each line‐of‐sight flow has a known azimuth (or direction), which must be converted into the actual vector flow. However, the component perpendicular to the azimuth direction is unknown. Our method uses historical data from the SuperDARN database and PCR to determine a fill‐in model convection distribution for any given universal time. The fill‐in data process is driven by a list of state descriptors (magnetic indices and the solar zenith angle). The final solution is then derived from a spherical cap harmonic fit to the SuperDARN measurements and the fill‐in model. When compared with the standard SuperDARN fill‐in model, we find that our fill‐in model provides improved solutions, and the final solutions are in better agreement with the SuperDARN measurements. Our solutions are far less dynamic than the standard SuperDARN solutions, which we interpret as being due to a lack of magnetosphere‐ionosphere inertia and communication delays in the standard SuperDARN technique while it is inherently included in our approach. Rather, we argue that the magnetosphere‐ionosphere system has inertia that prevents the global convection from changing abruptly in response to an interplanetary magnetic field change.

A Comparison of Regression Techniques for Estimation of Above-Ground Winter Wheat Biomass Using Near-Surface Spectroscopy

Above-ground biomass (AGB) provides a vital link between solar energy consumption and yield, so its correct estimation is crucial to accurately monitor crop growth and predict yield. In this work, we estimate AGB by using 54 vegetation indexes (e.g., Normalized Difference Vegetation Index, Soil-Adjusted Vegetation Index) and eight statistical regression techniques: artificial neural network (ANN), multivariable linear regression (MLR), decision-tree regression (DT), boosted binary regression tree (BBRT), partial least squares regression (PLSR), random forest regression (RF), support vector machine regression (SVM), and principal component regression (PCR), which are used to analyze hyperspectral data acquired by using a field spectrophotometer. The vegetation indexes (VIs) determined from the spectra were first used to train regression techniques for modeling and validation to select the best VI input, and then summed with white Gaussian noise to study how remote sensing errors affect the regression techniques. Next, the VIs were divided into groups of different sizes by using various sampling methods for modeling and validation to test the stability of the techniques. Finally, the AGB was estimated by using a leave-one-out cross validation with these powerful techniques. The results of the study demonstrate that, of the eight techniques investigated, PLSR and MLR perform best in terms of stability and are most suitable when high-accuracy and stable estimates are required from relatively few samples. In addition, RF is extremely robust against noise and is best suited to deal with repeated observations involving remote-sensing data (i.e., data affected by atmosphere, clouds, observation times, and/or sensor noise). Finally, the leave-one-out cross-validation method indicates that PLSR provides the highest accuracy (R2 = 0.89, RMSE = 1.20 t/ha, MAE = 0.90 t/ha, NRMSE = 0.07, CV (RMSE) = 0.18); thus, PLSR is best suited for works requiring high-accuracy estimation models. The results indicate that all these techniques provide impressive accuracy. The comparison and analysis provided herein thus reveals the advantages and disadvantages of the ANN, MLR, DT, BBRT, PLSR, RF, SVM, and PCR techniques and can help researchers to build efficient AGB-estimation models.

Prediction of peroxide value in omega-3 rich microalgae oil by ATR-FTIR spectroscopy combined with chemometrics.

Our work explored, for the first time, monitoring peroxide value (PV) of omega-3 rich algae oil using ATR-FTIR spectroscopic technique. The PV of the developed method was compared by that obtained by standard method of Association of Official Analytical Chemists (AOAC). In this study, peak area integration (PAI), Partial Least Squares Regression (PLSR), and Principal Component Regression (PCR) were used as the calibration techniques. PV obtained by the AOAC method and by FTIR-ATR technique were well correlated considering the peak area related to trans double bonds and chemometrics techniques of PLSR and PCR. Calibration model was established using the band with a peak point at 966cm-1 (990-940cm-1) related to CH out of plane deformation vibration of trans double bond. Algae oil oxidation could be successfully quantified using PAI, PLSR and PCR techniques. Additionally, hierarchical cluster analysis was performed and significant discrimination was observed coherently with oxidation process.

Detection of Broken Kernels Content in Bulk Wheat Samples Using Near-Infrared Hyperspectral Imaging

Broken kernels in stored wheat are contaminants, and these reduce the quality of wheat and increase risk of spoilage. The present study was conducted to predict the amount of broken kernels in bulk Canada Western Red Spring wheat samples using a near-infrared hyperspectral imaging system. The hyperspectral images of bulk wheat samples with different levels of broken kernels (0, 3, 6, 9, 12, and 15 %) were acquired in the 1000–1600 nm wavelength range at a 10-nm interval. The reflectance spectra acquired from these samples were used to develop regression models using principal component regression (PCR) and partial least-squares regression (PLSR) techniques. A tenfold cross-validation technique was used for determining the optimal number of components required to develop these regression models. Among the two regression techniques used, the PLSR technique [with mean square errors of predictions (MSEP), standard error of cross-validation (SECV), and correlation coefficient of 0.483, 0.70, and 0.94, respectively] performed better than the PCR technique (with MSEP, SECV, and correlation coefficient of 0.74, 0.86, and 0.88, respectively). The classification models were developed using the linear discriminant analysis (LDA) and quadratic discriminant analysis (QDA) statistical techniques. Both classifiers accurately classified (100.0 ± 0.0 %) the uncontaminated wheat sample from the higher contamination level (>6 %) wheat samples. Both classifiers also gave similar classification results (QDA 89.8 ± 2.6 % and LDA 87.7 ± 1.6 %) when multi-way classification was performed to classify uncontaminated sample from the contaminated samples.

Quantitative structure–property relationships in propene polymerization by zirconocenes with a rac-SiMe2[Ind]2 based ligand framework

Quantitative structure–property relationships in group 4 metallocene olefin polymerization catalysts were constructed for a dataset of 8 zirconocenes in a combined experimental and quantum chemical study. A total of 45 descriptors measuring the steric and electronic influence of the rac-SiMe2[Ind]2 based ligand framework were correlated with the observed polymerization activities, together with the molecular weights, melting temperatures, stereoerrors and regioerrors of the produced isotactic polypropene. Direct correlations between the calculated descriptors and observed catalytic properties turned out to be relatively weak due to the combined steric and electronic effects of multiple descriptors. Multiple regression models based on ordinary least squares linear regression and principal component regression techniques typically led to significant improvements, particularly for the molecular weights, melting temperatures and regioerrors, providing R2 of up to 0.92. The results provide a solid starting point for an extension of the methodology for a greater variety of catalyst structures, and thereby for the development of new high-performance catalysts for this industrially important process.

Deficit irrigation scheduling and yield prediction of ‘Kinnow’ mandarin (Citrus reticulate Blanco) in a semiarid region

Scarcity of irrigation water in critical growth stages of the crop is one of the major causes of low productivity and decline of citrus orchards. Regulated deficit irrigation (RDI) is a recently proposed water saving technique in irrigated agriculture. The present study was planned with a hypothesis that the optimal RDI scheduling at early fruit growth period (EFGP), which coincides with summer months could save substantial amount of water, without significantly affecting the yield of ‘Kinnow’ mandarin (Citrus reticulata Blanco) plants. Two DI strategies: (a) withholding irrigation at EFGP (RDI0) and (b) irrigation at 50% crop evapotranspiration (ETc) at EFGP (RDI50) were compared with full irrigation (FI, 100% ETc) in relation to gas exchange, water relation and nutrient composition of leaves along with growth and yield of the plants. The greater plant growth with maximum fruit yield (61.9–63.2tha−1) was recorded with fully-irrigated plants. However, the yield under RDI50 was statistically (p>0.05) at par with that under FI. The reduction in water application of around 24% with RDI50 resulted in 30% improvement in irrigation water use efficiency with this treatment over that with FI. The maximum rate of net-photosynthesis, stomatal conductance and transpiration of leaves was recorded with fully-irrigated plants. However, the plants under RDI50 exhibited the highest leaf water use efficiency. The leaf nutrients (N, P, K, Fe, Mn, Cu and Zn) analysis revealed that RDI0 produced significantly (p<0.05) lower concentration of all the nutrients except P and Cu than that in other treatments. Relative leaf water content (RLWC), leaf water concentration (LWC) and mid-day stem water potential (Ψ) showed a decreasing trend, whereas water stress integral (SΨ) and plant canopy reflectance indices (water band index, WBI; normalised difference water index, NDWI; and moisture stress index, MSI) showed the reverse trend of RLWC with water stress. The prediction model formulated based on midday stem water stress integral, leaf N, leaf K, stomatal conductance and water stress index using Principal component regression technique during EFGP performed well with reasonably accuracy (R2=0.85) to forecast annual fruit yield of the citrus plants. Overall, these results reveal that irrigation at 50% ETc during EFGP could impose desirable water stress on ‘Kinnow’ mandarin plants, improving their water use efficiency, without significantly affecting the fruit yield under water scarce condition.

Ultra-Performance Liquid Chromatography for the Multicomponent Analysis of a Ternary Mixture Containing Thiamine, Pyridoxine, and Lidocaine in Ampules

New chemometric approaches based on the application of partial least squares (PLS) and principal component regression (PCR) algorithms with fractional wavelet transform (FWT) and continuous wavelet transform (CWT) are proposed for the spectrophotometric multicomponent determination of thiamine hydrochloride (B1), pyridoxine hydrochloride (B6), and lidocaine hydrochloride (LID) in ampules without any separation step. In this study PLS and PCR techniques were applied to the raw spectral data, FWT-coefficients, and FWT-CWT-coefficients. These calibration models were labeled as Raw-PLS and Raw-PCR, FWT-PLS and FWT-PCR, and FWT-CWT-PLS and FWT-CWT-PCR, respectively. A new ultra-performance liquid chromatographic (UPLC) method was developed for the comparison of the results obtained by applying the chemometric calibration methods. Chromatographic separation and determination of B1, B6, and LID in ampules were performed on an Acquity UPLC BEH C18 column (50x2.1 mm id, 1.7 pm particle size) using gradient elution with a mobile phase consisting of methanol and 0.01 M HCI at a constant flow rate of 0.6 mL/min. These combined chemometric calibrations and UPLC were validated by analyzing various ternary mixtures, B1, B6, and LID. The proposed chemometric approaches (signal processing-multivariate calibrations) and UPLC method were applied to the quantitative multicomponent analysis of marketed ampules containing the vitamins B1 and B6 with LID.

Forecasting of air quality in Delhi using principal component regression technique

Over the past decade, an increasing interest has evolved by the public in the day–to–day air quality conditions to which they are exposed. Driven by the increasing awareness of the health aspects of air pollution exposure, especially by most sensitive sub–populations such as children and the elderly, short–term air pollution forecasts are being provided more and more by local authorities. The Air Quality Index (AQI) is a number used by governmental agencies to characterize the quality of the air at a given location. AQI is used for local and regional air quality management in many metropolitan cities of the world. The main objective of the present study is to forecast short–term daily AQI through previous day’s AQI and meteorological variables using principal component regression (PCR) technique. This study has been made for four different seasons namely summer, monsoon, post monsoon and winter. AQI was estimated for the period of seven years from 2000–2006 at ITO (a busiest traffic intersection) for criteria pollutants such as respirable suspended particulate matter (RSPM), sulfur dioxide (SO2), nitrogen dioxide (NO2) and suspended particulate matter (SPM) using a method of US Environmental Protection Agency (USEPA), in which sub–index and breakpoint pollutant concentration depends on Indian National Ambient Air Quality Standard (NAAQS). The Principal components have been computed using covariance of input data matrix. Only those components, having eigenvalues ≥1, were used to predict the AQI using principal component regression technique. The performance of PCR model, used for forecasting of AQI, was better in winter than the other seasons as studied through statistical error analysis. The values of normalized mean square error (NMSE) were found as 0.0058, 0.0082, 0.0241 and 0.0418 for winter, summer, post monsoon and monsoon respectively. The other statistical parameters are also supporting the same result.

Nonparametric time series forecasting with dynamic updating

We present a nonparametric method to forecast a seasonal univariate time series, and propose four dynamic updating methods to improve point forecast accuracy. Our methods consider a seasonal univariate time series as a functional time series. We propose first to reduce the dimensionality by applying functional principal component analysis to the historical observations, and then to use univariate time series forecasting and functional principal component regression techniques. When data in the most recent year are partially observed, we improve point forecast accuracy by using dynamic updating methods. We also introduce a nonparametric approach to construct prediction intervals of updated forecasts, and compare the empirical coverage probability with an existing parametric method. Our approaches are data-driven and computationally fast, and hence they are feasible to be applied in real time high frequency dynamic updating. The methods are demonstrated using monthly sea surface temperatures from 1950 to 2008.

A multisite seasonal ensemble streamflow forecasting technique

We present a technique for providing seasonal ensemble streamflow forecasts at several locations simultaneously on a river network. The framework is an integration of two recent approaches: the nonparametric multimodel ensemble forecast technique and the nonparametric space‐time disaggregation technique. The four main components of the proposed framework are as follows: (1) an index gauge streamflow is constructed as the sum of flows at all the desired spatial locations; (2) potential predictors of the spring season (April–July) streamflow at this index gauge are identified from the large‐scale ocean‐atmosphere‐land system, including snow water equivalent; (3) the multimodel ensemble forecast approach is used to generate the ensemble flow forecast at the index gauge; and (4) the ensembles are disaggregated using a nonparametric space‐time disaggregation technique resulting in forecast ensembles at the desired locations and for all the months within the season. We demonstrate the utility of this technique in skillful forecast of spring seasonal streamflows at four locations in the Upper Colorado River Basin at different lead times. Where applicable, we compare the forecasts to the Colorado Basin River Forecast Center's Ensemble Streamflow Prediction (ESP) and the National Resource Conservation Service “coordinated” forecast, which is a combination of the ESP, Statistical Water Supply, a principal component regression technique, and modeler knowledge. We find that overall, the proposed method is equally skillful to existing operational models while tending to better predict wet years. The forecasts from this approach can be a valuable input for efficient planning and management of water resources in the basin.

Stability‐indicating chemometric methods for the determination of pyritinol dihydrochloride

Three multivariate calibration methods, including classical least square with nonzero intercept (CLS), principal component regression (PCR) and partial least square (PLS), have been used for the determination of pyritinol dihydrochloride in the presence of its degradation product. The CLS, PCR and PLS techniques are useful in spectral analysis because the simultaneous inclusion of many spectral wavelengths instead of the single wavelength used in derivative spectrophotometry has greatly improved the precision and predictive abilities of these multivariate calibrations. A training set was constructed for the mixture and the best model was used for the prediction of the concentration of the selected drug. The proposed procedures were applied successfully in the determination of pyritinol dihydrochloride in laboratory-prepared mixtures and in commercial preparations. Pyritinol dihydrochloride was analysed with mean accuracies 99.99 +/- 0.905, 99.91 +/- 0.966 and 99.92 +/- 0.962 using the CLS, PCR and PLS methods respectively. The validity of the proposed methods was assessed using the standard addition technique. The proposed procedures were found to be rapid and simple and required no preliminary separation. They can therefore be used for the routine analysis of pyritinol dihydrochloride in quality-control laboratories.

Vibrational spectra and antimicrobial activity of selected bivalent cation benzoates

Selected bands of FT-IR spectra of Mg(II), Ca(II), Cu(II) and Zn(II) benzoates of both solid state and water solution, were assigned to appropriate molecular vibrations. Next evaluation of electronic charge distribution in both carboxylic anion and aromatic ring of studied compounds was performed. Classical plate tests and turbidimetry measurements, monitoring growth of bacteria Escherichia coli, Bacillus subtilis and yeasts Pichia anomala and Saccharomyces cerevisiae during 24 h of incubation, in optimal growth conditions (control) and in medium with addition of studied benzoate (concentration of 0.01% expressed as the concentration of benzoic acid), proved antimicrobial activity of studied compounds against investigated micro-organisms. PLS (partially least square) and PCR (principal component regression) techniques were applied to build a model, correlating spectral data reflecting molecular structure of studied compounds, with degree of influence of those compounds on growth of studied micro-organisms. Statistically significant correlation within cross validation diagnostic of PLS-1 calibration was found, when log 1/T of selected spectral regions of water solution samples were used as input data. The correlation coefficients between predicted with PLS calibration based on created 1, 2 or 3 factor models, and actual values of antimicrobial activity were: 0.70; 0.76, 0.81 for P. anomala, B. subtilis, and E. coli, respectively. Log(PRESS) values of appropriate models were 2.10, 2,39 and 3.23 for P. anomala, B. subtilis, and E. coli, respectively.

MULTIVARIATE DETERMINATION OF PROPYL GALLATE IN RBD PALM OLEIN USING PARTIAL LEAST SQUARES AND PRINCIPAL COMPONENT REGRESSION BASED ON FTIR SPECTROSCOPY

ABSTRACT Fourier transform infrared (FTIR) spectroscopy with transmission cell was used to determine propyl gallate (PG) in refined, bleached and deoderized (RBD) palm olein. Partial least squares (PLS) and principal component regression (PCR) methods coupled with the “leave‐one‐out” cross‐validation procedure were employed to construct the calibration models for predicting PG in the FTIR spectral region of 3707–3262 cm−1. The prediction capabilities of these calibration models were evaluated by comparing their standard errors of calibration and standard errors of prediction. The precision of this method was shown to be comparable to the accuracy of the method used by the International Union for Pure and Applied Chemistry for the measurement of PG content in oil samples, with a coefficient of determination (R2) of 0.9921 for PLS and an R2of 0.9902 for PCR techniques. The results indicate that FTIR serves as a useful analytical tool for the determination of PG in RBD palm olein.

Five hundred years of gridded high-resolution precipitation reconstructions over Europe and the connection to large-scale circulation

We present seasonal precipitation reconstructions for European land areas (30°W to 40°E/30–71°N; given on a 0.5°×0.5° resolved grid) covering the period 1500–1900 together with gridded reanalysis from 1901 to 2000 (Mitchell and Jones 2005). Principal component regression techniques were applied to develop this dataset. A large variety of long instrumental precipitation series, precipitation indices based on documentary evidence and natural proxies (tree-ring chronologies, ice cores, corals and a speleothem) that are sensitive to precipitation signals were used as predictors. Transfer functions were derived over the 1901–1983 calibration period and applied to 1500–1900 in order to reconstruct the large-scale precipitation fields over Europe. The performance (quality estimation based on unresolved variance within the calibration period) of the reconstructions varies over centuries, seasons and space. Highest reconstructive skill was found for winter over central Europe and the Iberian Peninsula. Precipitation variability over the last half millennium reveals both large interannual and decadal fluctuations. Applying running correlations, we found major non-stationarities in the relation between large-scale circulation and regional precipitation. For several periods during the last 500 years, we identified key atmospheric modes for southern Spain/northern Morocco and central Europe as representations of two precipitation regimes. Using scaled composite analysis, we show that precipitation extremes over central Europe and southern Spain are linked to distinct pressure patterns. Due to its high spatial and temporal resolution, this dataset allows detailed studies of regional precipitation variability for all seasons, impact studies on different time and space scales, comparisons with high-resolution climate models as well as analysis of connections with regional temperature reconstructions.

Statistical–Dynamical Seasonal Prediction Based on Principal Component Regression of GCM Ensemble Integrations

Abstract A statistical approach to correct a dynamical ensemble forecast of future seasonal means based on the past performance of a general circulation model (GCM) is formulated. The approach combines principal component (PC) analysis with the regression technique to remove the systematic structural (distortion) error from the GCM ensemble. The performance of this statistical–dynamical method is assessed by comparing its cross-validated skill with the explicit skill achieved in the raw GCM ensembles. When the PC regression technique is applied to seasonal means from an ensemble of the Center for Ocean–Land–Atmosphere Studies (COLA) GCM, it not only recovers most of the explicit skill in the original ensemble mean, but also acts to correct significant errors in the ensemble. It is shown that some ensemble errors are due to noise that can be easily removed by applying a simple regression scheme. A novel aspect of the PC regression technique, however, is that it goes beyond the simple filtering of noise and...