Figure 3 - uploaded by Christos A. Karavitis
Content may be subject to copyright.
Estimated (forecasted) SPI time series and calculated (occurred) SPI for the meteorological stations of Action, Livadia, Chios and Lamia, from January 2009 to December 2010.
Source publication
The current work presents the application of the seasonal Auto Regressive Integrated Moving Average Model
(ARIMA) using the Standard Precipitation Index (SPI) as a drought indicator and then depicting the spatial
distribution through geo-statistical methods. Greece is very often facing the hazardous impacts of droughts, hence
presenting an almost i...
Context in source publication
Context 1
... in all the SARIMA estimations seem to be more reliable than the other models tested and they have been applied to develop two projected SPI (6 and 12) time series on a monthly scale. Figure 3 presents the produced time series for the meteorological stations of Action, Livadia, Chios and Lamia, from January 2009 to December 2010. Such a time period was chosen in order to evaluate the performance of the models, instead of using them to forecast a future period, e.g. ...
Similar publications
Time series models are promising tools for forecasting commodity prices. However, their applications to guide producers in agricultural investments and marketing decisions are still limited. This article compares the ARIMA and Holt-Winters Exponential Smoothing models in terms of forecasting the monthly wholesale rice price in Tanzania. Even with v...
One of the most basic needs of any human being to survive is air. Unfortunately, this basic need is being polluted by many natural factors like volcanic eruptions, forest fires, and man-induced factors like transportation emission. Unpolluted air is now an ideal environment that can never be achieved. So, the pollution levels should be monitored co...
Airplane is one of the public transportations options that many people choose when traveling long distance. Nowadays, it is notes that the number of passengers domestic flight has increased from the previous months. This increase, especially occurs on the holidays, such as year-end holidays, Eid, and others. The increase of airplane passengers is i...
**Background.**
This study aimed to determine whether crime patterns in Mexico City changed due to the COVID-19 pandemic, and to test whether any changes observed were associated with the disruption of routine activities.
**Method.**
The first objective was assessed by comparing the observed incidence of crime after the COVID-19 pandemic was detec...
This paper presents a framework for the practical modeling of inflation, as one of the key economic indicators. Empirical research of monthly inflation trends in the Republic of Serbia was done covering the period from January 2007 to December 2015. The seasonally adjusted ARIMA model and Holt-Winters smoothing were used for determining the future...
Citations
... Their models proved to be efficient in predicting drought up to two months ahead. Karavitis et al. [26] demonstrated the effectiveness of combining a stochastic model and a geospatial method for short-term drought prediction. Bazrafshan et al. [27] applied ARIMA for seasonal drought forecasting of the river basin in Iran. ...
Drought forecasting is essential for risk management and preparedness of drought mitigation measures. The present study aims to evaluate the effectiveness of the proposed hybrid technique for regional drought forecasting. Multiple Linear Regression (MLR), Artificial Neural Network (ANN), and two wavelet techniques, namely, Discrete Wavelet Transform (DWT) and Wavelet Packet Transform (WPT), were evaluated in drought forecasting up to a lead time of six months. Standard error metrics were used to select optimal model parameters, such as number of inputs, number of hidden neurons, level of decomposition, and number of mother wavelets. Additionally, the performance of various mother wavelets, including the Haar wavelet (db1) and 19 Daubechies wavelets (db1 to db20), were evaluated. The results indicated that the ANN model produced better forecasts than the MLR model, whereas the hybrid models outperformed both ANN and MLR models, which failed to predict the SPI values for a lead time greater than two months. The performance of all the models was found to improve as the timescale increased from 3 to 12 months. However, all the models’ performances deteriorated as the lead time increased. The hybrid WPT-MLR was the best model for the study area. The findings indicated that a hybrid WPT-MLR model could be used for drought early warning systems in the study area.
... To date, many research papers have been performed to model and predict drought indices using machine learning (ML) techniques (Bacanli et al. 2009;Beyaztas and Yaseen 2019;Danandeh Mehr et al. 2014;Deo et al. 2016;Durdu 2010;Dehghani et al. 2014;Karavitis et al. 2015;Keskin et al. 2009;Mishra and Desai 2005;Nourani and Molajoo 2017;Özger et al. 2011;Yaseen et al. 2021). For example, the Artificial Neural Networks (ANN) were used to forecast both SPI and Effective Drought Index (EDI) with the lead times of 1 to 12 months for the Tehran Province, Iran (Morid et al. 2007). ...
The advancement of the machine learning (ML) models has demonstrated notable progress in geosciences. They can identify the underlying process or causality of natural hazards. This article introduces the development and verification procedures of a new hybrid ML model, namely Bat-ELM for predictive drought modelling. The multi-temporal standardized precipitation evapotranspiration index (SPEI-3 and SPEI-6) is computed as the meteorological drought index at two study regions (Beypazari and Nallihan), located in Ankara province, Turkey. The proposed hybrid model is obtained by integrating the Bat optimization algorithm as the parameter optimizer with an extreme learning machine (ELM) as the regressor engine. The efficiency of the intended model was evaluated against the classic artificial neural network (ANN) and standalone ELM models. The evaluation and assessment are conducted using statistical metrics and graphical diagrams. The forecasting results showed that the accuracy of the proposed model outperformed the benchmark models. In a quantitative assessment, the Bat-ELM model attained minimal root mean square error for the SPEI-3 and SPEI-6 (RMSE = 0.58 and 0.43 at Beypazari station and RMSE = 0.53 and 0.37 at Nallihan station) over the testing phase. This indicates the new model approximately 20 and 15% improves the forecasting accuracy of traditional ANN and classic ELM techniques, respectively.
... To date, a large number of research articles have been conducted to forecast drought indices using classical regression or data-driven techniques [5][6][7][8][9][10][11][12][13][14]. For example, the Artificial Neural Network (ANN) based forecast models of 1 to 12 months lead time were developed taking the advantage of SPI and Effective Drought Index [15]. ...
This study presents developing procedures and verification of a new hybrid model, namely wavelet packet-genetic programming (WPGP) for short-term meteorological drought forecast. To this end, the multi-temporal standardized precipitation evapotranspiration index (SPEI) has been used as the drought quantifying parameter at two meteorological stations at Ankara province, Turkey. The new WPGP model comprises two main steps. In the first step, the wavelet packet, which is a generalization of the well-known wavelet transform, is used to decompose the SPEI series into deterministic and stochastic sub-signals. Then, classic genetic programming (GP) is applied to formulate the deterministic sub-signal considering its effective lags. To characterize the stochastic component, different theoretical probability distribution functions were assessed, and the best one was selected to integrate with the GP-evolved function. The efficiency of the new model was cross-validated with the first order autoregressive (AR1), GP, and random forest (RF) models developed as the benchmarks in the present study. The results showed that the WPGP is a robust model, superior to AR1 and RF, and significantly increases the predictive accuracy of the standalone GP model.
... This method deserves some explanations; Statistical forecasting has been widely used in a plethora of disciplines and especially in cases where management and decision making should take place. However, a variety of forecasting methods and trend extrapolations seems to be the one predominantly used (Karavitis, Vasilakou, & Tsesmelis, 2015). According to Shatanawi (Shatanawi, Rahbeh, & Shatanawi, 2013), various methods can be used for extrapolation and predictions based on historical records assuming no threshold phenomena. ...
The Anthropocene and islands: vulnerability, adaptation and resilience to natural hazards and climate change include 8 original research chapters, of authors from around the World, explaining how islands are affected by natural hazards and global change. The volume contributions range from small islands in Alaska to large ones such as Sicily in the Mediterranean and focus on facts such as water resources, sustainability and societal impacts of risk and climate change. The author’s reflections share a wide scientific approach that will enrich a subject, islands and its future, which will become more and more important in the next decades.
... Previously, several statistical and machine learningbased time series models have been used for forecasting of drought by means of SPI and other indices. Auto-Regressive Integrated Moving Average (ARIMA), Seasonal ARIMA, recursive multi-step, and direct multi-step models based on neural networks were tested to forecast droughts using SPI time series (Mishra and Desai 2005;Han et al. 2013;Bazrafshan et al. 2015;Karavitis et al. 2015). Artificial Neural Network (ANN)-based forecast models of 1-12 month lead time were developed taking the advantage of SPI and EDI indices by Morid et al. (2007). ...
Among all the natural disasters, drought has the most catastrophic encroachment on the surrounding and environment. Gulbarga, one of the semi-arid districts of Karnataka state, India receives about 700 mm of average annual rainfall and is drought inclined. In this study, the forecasting of drought for the district has been carried out for a lead time of 1 month and 6 months. The multi-temporal Standardized Precipitation Index (SPI) has been used as the drought quantifying parameter due to the fact that it is calculated on the basis of one simplest parameter, i.e., rainfall and additionally due to its ease of use. The fine resolution daily gridded precipitation data (0.25º × 0.25º) procured from Indian Meteorological Department (IMD) of 21 grid locations within the study area have been used for the analysis. Forecasting of drought plays a significant role in drought preparedness and mitigation plans. With the advent of machine learning (ML) techniques over the past few decades, forecasting of any hydrologic event has become easier and more accurate. However, the use of these techniques for drought forecasting is still obscure. In this study, Artificial Neural Network (ANN) and Support Vector Regression (SVR) techniques have been employed to examine their accuracy in drought forecasting over shorter and longer lead times. Furthermore, two hybrid approaches have been formulated by coupling a data transformation method with each of the aforementioned ML approaches. At the outset, pre-processing of input data (i.e., SPI) has been carried out using Wavelet Packet Transform (WPT) and then used as inputs to ANN and SVR models to induce hybrid WP-ANN and WP-SVR models. The performance of the hybrid models has been evaluated based on the statistical indices such as R 2 (co-efficient of determination), RMSE (Root Mean Square Error), and MAE (Mean Absolute Error). The results showed that the hybrid techniques have better forecast performance than the standalone machine learning approaches. Hybrid WP-ANN model performed relatively better than WP-SVR model for most of the grid locations. Also, the forecasting results deteriorated as the lead time increased from 1 to 6 months.
... It usually incorporates conflicting natural and anthropogenic interests and objectives, multiple water uses, sensitive social and ecological equilibria, recurring natural hazards. Frameworks like integrated water resources management and sustainable development are implement within the environmental decision making process in order to mitigate adverse effects and achieve consensus among different water user groups (Margerum and Born 1995;Grigg 1996;Loucks et al. 2005;Grigg 2008;Karavitis et al. 2015b). Thus, regarding the implementation of any comprehensive drought management scheme, the crucial obstacles to be surmounted are having as epicenters drought's onset, areal extent, and severity. ...
... Data gaps in all 24 stations were minimal and raw data seem more appropriate to be inserted so as to represent the natural drought conditions (extreme minimum values), hence, no attempt made to apply "corrective" homogeneity in the data sets (Karavitis et al. 2011a). The implicit stationary assumptions of the SPI were tested with all stations exhibiting stationary process (Karavitis et al. 2015b). The SPI then was calculated on monthly basis both with the 6 month (SPI6) and 12 month (SPI12) step. ...
Indices are used for representing complex phenomena; however, concerns usually arise regarding their objectivity and reliability, particularly dealing with their uncertainties during the development process. The current overarching objective is to reveal the significance of employing different weighting techniques in the application of the Standardized Drought Vulnerability Index (SDVI) and demarcate any pertinent implications that may emerge in drought decision making. Greece, as it is very often facing the catastrophic effects of droughts, presents an almost ideal case for the SDVI testing. SDVI outcomes were tested utilizing five weighting techniques deriving from four weighting methods. The analyses indicated that the use of complex weighting models may not be necessary in all cases and that the simple equal weighting method seems more effective to estimate drought vulnerability. It also seems more important to address the search for valid, reliable and relevant individual indicators forming the complex index as well as appropriate index development processes that would measure performance of water bodies, systems and schemes, monitor the process of equitable sharing, and provide mechanisms for monitoring the state and changes in interdependent water systems.
... The combining effects of increased water demands and climate change impacts will further stress our ability to cope with droughts in the future Guo et al., 2018;Wang et al., 2017). The drought nature, characteristics and impacts have continuously drawn the attention of the scientific community, the state and federal entities, as well as the general public, resulting in the production of a rich menu of drought literature that mostly provides crucial information on its parameters and management strategies (Cancelliere et al., 2005;Fontane and Frevert, 1995;Grigg, 2014;Grigg and Vlachos, 1993;Hagman et al., 1984;Karavitis, 1992Karavitis, , 1998Karavitis et al., 2012aKaravitis et al., , 2015aKaravitis et al., , 2015bLoukas and Vasiliades, 2004;Playán et al., 2013;Pulwarty and Maia, 2015;Rosenberg, 1978Rosenberg, , 1980Traore and Fontane, 2007;Vasiliades and Loukas, 2009;Vlachos, 1982;Wilhite, 2004;Zarch et al., 2015;Wang et al., 2016a). ...
Drought is a complex natural hazard with its adverse multifaceted impacts cascading in every physical and human system. The vulnerability magnitude of various areas to drought mostly depends on their exposure to water deficiency, the existing water management policy framework and its implementation. The Standardized Drought Vulnerability Index (SDVI) is an integrated attempt towards characterizing drought vulnerability based on a comparative classification system, incorporating precipitation patterns, the supply and demand trends, and the socioeconomic background as the most crucial contributors to drought vulnerability. This work attempts to evolve the SDVI by presenting a more rigorous method of index parameters estimation and argues that the combination of in-situ and satellite data improve the index results in an effort to further minimize the paucity of drought related information. At the same time, it helps to surpass previous limitations in temporal and spatial propagation of the vulnerability concept. The new framework is applied in the South Platte Basin, within Colorado, on the 2012 summer drought (July-September). The proposed index modification may convey drought information in a more holistic manner to decision makers. SDVI could aid in advancing the understanding of each component contribution through in situ and remote sensing data integration and in avoiding existing practices of broken linkages and fragmentation of the reported impacts. Thus, it is believed that the SDVI could serve as an additional tool to guide decisions and target mitigation and adaptation actions, allowing for a more integrated management approach.
... Mossad and Alazba [24] proved the potential ability of these modelling approaches to forecast drought. However, drought forecasts performed at monthly time-scale for early warning [25][26][27][28][29] do not account for long-term patterns of evolution, which are essential to study and monitor drought from a climate perspective [30]. Here, we target annual to decadal time scales. ...
Severity of drought in California (U.S.) varies from year-to-year and is highly influenced by precipitation in winter months, causing billion-dollar events in single drought years. Improved understanding of the variability of drought on decadal and longer timescales is essential to support regional water resources planning and management. This paper presents a soft-computing approach to forecast the Palmer Drought Severity Index (PDSI) in California. A time-series of yearly data covering more than two centuries (1801-2014) was used for the design of ensemble projections to understand and quantify the uncertainty associated with interannual-to-interdecadal predictability. With a predictable structure elaborated by exponential smoothing, the projections indicate for the horizon 2015-2054 a weak increase of drought, followed by almost the same pace as in previous decades, presenting remarkable wavelike variations with durations of more than one year. Results were compared with a linear transfer function model approach where Pacific Decadal Oscillation and El Niño Southern Oscillation indices were both used as input time series. The forecasted pattern shows that variations attributed to such internal climate modes may not provide more reliable predictions than the one provided by purely internal variability of drought persistence cycles, as present in the PDSI time series.
... Hence, even though the model provides the best unbiased estimate for a modeled cell, it cannot be used as a true measure of the system. Nevertheless, the Autoregressive Integrated Moving Average (ARIMA) ( Box and Pierce, 1970 ) family of models has been extensively and repeatedly used for the analysis and forecasting in a wide range of water related studies including floods, drought, water management, water quality ( Ahmad et al., 2001;Durdu, 2010;Irvine and Eberhardt, 1992;Karavitis et al., 2015;Kim et al., 2011;Papamichail and Georgiou, 2001;Salas et al., 1980;Sun and Koch, 2001;Toth et al., 1999;Wang et al., 2014Wang et al., ,2015. Modeling and forecasting of groundwater levels through ARIMA models with model variations, including seasonality and/or stationarity, is a very common practice ( Changnon et al., 1988;Mirzavand and Ghazavi, 2014;Von Asmuth et al., 2008;von Asmuth et al., 2012 ). ...
Groundwater level knowledge is one of the crucial information for sustainable use of groundwater resources and efficient and effective conjunctive management of surface and groundwater resources. Despite the fact that there is an increase of water related datasets collected, still groundwater information remains one of the Achilles heel in terms of analysis and modeling. The present paper is introducing a novel framework for filling data gaps in groundwater time series in an efficient scheme with low computational cost. It utilizes direct (known measurements) and auxiliary (regional groundwater model) available information. The framework employs an exogenous seasonal autoregressive integrated moving average (SARIMAX) stochastic model to describe the simulated groundwater level fluctuation process of a regional physical groundwater model and the Ensemble Smoother (ES) for predicting the water table level. The framework was evaluated by conducting three numerical experiments for five groundwater wells in the South Platte alluvial aquifer. The results indicate that this framework could serve as a valuable tool for the enhancement of groundwater time series, for both intermittent missing data and of continuous gaps with short span, which could then lift hindrances in their analysis and use in groundwater models, and thus aid water management decisions.
... Tsakiris and Vangelis (2004) found that SPI can support a drought watch system for an area of mesoscale dimensions. Karavitis et al. (2015) presented an application of a seasonal prediction model using the SPI as a drought indicator. Pai et al. (2011) have examined district-wise drought events over India using SPI and found that SPI was suitable for examining break and active events in the monsoon rainfall. ...
In the present study, skill of an extended range forecast system has been evaluated for identifying droughts over central India 20-days in advance. Rainfall forecasts from 44 ensemble members of the forecast system developed at Indian Institute of Tropical Meteorology (IITM), Pune have been used to prep are probabilistic rainfall forecasts. It i s seen that the uncertainties in the forecasts (in terms of ensemble spread) increases from day-5 to day 20. As the focus of the study is for drought predictions, forecasts in the bins 0-5 mm/5 day and 5-25 mm/5 day (no rain or less rain) were studied in detail. It is found that the modeling system has a tendency to over-forecast rainfall probabilities. With bias correction, the forecasts become more reliable. Various drought indices were computed using the mean of the forecast distribution up to 20-days in advance. Standardized precipitation index (SPI) computed using Gamma and Pearson type- III distributions are similar in the study region. It was found that these are in reasonable agreement with those from observations. Probabilistic forecasts of standardized precipitation index (SPI) were made and the relative operating characteristics (ROC) scores indicate that the forecasted SPI values are suitable for application.
