Standardized Precipitation Evaporation Index Research Articles

An improved SPEI drought forecasting approach using the long short-term memory neural network

Droughts are slow-moving natural hazards that gradually spread over large areas and capable of extending to continental scales, leading to severe socio-economic damage. A key challenge is developing accurate drought forecast model and understanding a models’ capability to examine different drought characteristics. Traditionally, forecasting techniques have used various time-series approaches and machine learning models. However, the use of deep learning methods have not been tested extensively despite its potential to improve our understanding of drought characteristics. The present study uses a deep learning approach, specifically the Long Short-Term Memory (LSTM) to predict a commonly used drought measure, the Standard Precipitation Evaporation Index (SPEI) at two different time scales (SPEI 1, SPEI 3). The model was compared with other common machine learning method, Random Forests, Artificial Neural Networks and applied over the New South Wales (NSW) region of Australia, using hydro-meteorological variables as predictors. The drought index and predictor data were collected from the Climatic Research Unit (CRU) dataset spanning from 1901 to 2018. We analysed the LSTM forecasted results in terms of several drought characteristics (drought intensity, drought category, or spatial variation) to better understand how drought forecasting was improved. Evaluation of the drought intensity forecasting capabilities of the model were based on three different statistical metrics, Coefficient of Determination (R2), Root Mean Square Error (RMSE), and Mean Absolute Error (MAE). The model achieved R2 value of more than 0.99 for both SPEI 1 and SPEI 3 cases. The variation in drought category forecasted results were studied using a multi-class Receiver Operating Characteristic based Area under Curves (ROC-AUC) approach. The analysis revealed an AUC value of 0.83 and 0.82 for SPEI 1 and SPEI 3 respectively. The spatial variation between observed and forecasted values were analysed for the summer months of 2016–2018. The findings from the study show an improvement relative to machine learning models for a lead time of 1 month in terms of different drought characteristics. The results from this work can be used for drought mitigation purposes and different models need to be tested to further enhance our capabilities.

Drought Monitoring in Northern Nigeria Using Four (4) Indices.

Drought can generally be defined as the extreme persistence of precipitation deficit over a region for a specific period. Eight study locations were picked from the Sudano-Sahelian agro-ecological zones of Nigeria (Bauchi, Bida. Kaduna, Kano, Maiduguri. Sokoto, Nguru, and Katsina) from 1981 to 2015. The Standardized Precipitation Index (SPI), Standardized Precipitation Evaporation Index- Thornthwaite (SPEI.T), Standardized Precipitation Evaporation Index-Hargreaves (SPEI-H) and Standardized Precipitation Evaporation Index-penman (SPEI-P) were used as the primary indicators of meteorological and agricultural droughts. The correlation coefficient shows an increasing correlation among the indices with increasing time scale, with SPI and SPEI-H having the highest correlation. The regression analysis shows a monotonic increasing relationship between indices while SPI vs SPEI-H has the highest correlation coefficient. The number of drought occurrences captured by the indices also increases with increasing time scale with SPEI-P detecting the highest number of drought events. All the drought indices reflect the historical drought periods between 1982-1989, 1992-2002, and 2008-2011. SPI, SPEI-P, and SPEI-H detected similar duration and intensity for the historical drought between 1982 and 1989 while SPEI-P showed the highest intensity and duration for the historical droughts between 1992 and 2002 and between 2008 and 2011.Analytic Hierarchy Process (AHP) evaluated that SPEI-P was more robust and sophisticated, SPI and SPEI-P had the same score for tractability while SPEI-H being the least tractable, and SPI had the highest for transparency and extendibility.

淮河流域干旱时空演变特征及成因

PDF HTML阅读 XML下载 导出引用 引用提醒 淮河流域干旱时空演变特征及成因 DOI: 10.5846/stxb201911042314 作者: 作者单位: 作者简介: 通讯作者: 中图分类号: 基金项目: 国家重点研发计划项目（2019YFA0606900）；安徽高校协同创新项目（GXXT 2019047）；国家自然科学基金项目（41601023，41771536）；安徽省自然科学基金（1808085QD117）；北京师范大学环境演变与自然灾害教育部重点实验室开放课题和安徽师范大学研究生科研创新与实践项目（2019kycx051） Spatio-temporal distribution characteristics of meteorological drought and climate influence factors Author: Affiliation: Fund Project: 摘要 | 图/表 | 访问统计 | 参考文献 | 相似文献 | 引证文献 | 资源附件 | 文章评论 摘要:基于淮河流域149气象站点1962-2016年标准化降水蒸散发指数、16个气候因子和NCEP/NCAR再分析资料，通过小波分析和旋转经验正交函数等手段，分析淮河流域干旱重心的转移轨迹，研究气象干旱与气候因子的相关关系，并通过大气环流的异常特征揭示气象干旱的主要成因。结果表明：（1）干旱重心分布主要从淮河流域中心向四周扩散，淮河流域大范围区域呈干旱化态势。2013年干旱重心从西北部→中部→西南部→中部变化，随着干旱面积的增大，干旱重心由四周向中心移动。（2） PDO、ONI、Nino4、Nino3.4、MEI、BEST与SPEI均呈显著正相关关系，SOI、TNI与SPEI则呈显著负相关关系。（3）干旱周期主要集中在1970年代、1990年代和2000年代存在2-5年显著周期，气候因子在3.4-4.5年存在显著周期。（4）春季高纬度地区的气流南下，与印度洋、孟加拉湾北上气流导致南湿北干；夏季蒙古气旋偏弱与异常偏北风覆盖导致东干西湿；秋季大陆高压控制，偏北风和南风相互影响造成东干西湿；冬季盛行下沉气流与盛行东南风造成东湿西干的气候特征。 Abstract:In this study, the shift track of the gravity of drought area was analyzed based on 16 climate index' the NCEP/NCAR reanalysis data and the standardized precipitation evaporation (SPEI) from the 149 meteorological stations from 1962 to 2016 in the Huai River Basin. While the correlation coefficient between meteorological drought and climate index was calculated by the Wavelet analysis and Rotated Empirical Orthogonal Function method in Huai River Basin with the aim of revealing drought conditions across the Huai River Basin in the anomalous largescale atmospheric circulation. The results of this study indicated that: (1) the spatial distribution of the gravity of drought area mainly showed decreased from the center of Huai River Basin to the surrounding area. Moreover, SPEI showed a decreasing trend in in most parts of the Huai River Baisn. The gravity of drought area shifted from northwest to central, southwest and central in 2003 and the gravity of drought area shifted to the center with the increasing of drought area. (2) There is a significant positive correlation between SPEI and Pacific Decadal Oscillation, Oceanic Nino Index, Nino4, Nino3.4, Multivariable Enso Index, Bivariate Enso Timeseries, while There is a significant negative correlation between SPEI and Southern Oscillation Index, Traffic Nino Index. (3) The drought was mainly concentrated in the 1970s, 1990s and 2000s with a significant period of 2-5a, and the climate index had a significant period of 3.4-4.5a. (4) The water vapor from the Indian Ocean and the bay of Bengal is meet the high latitude cold air, leading to the moisture in the south and the dryness in the north in spring. The weakly Mongolian cyclone and the abnormal wind in the norther has led to the dry in the eastern and wet in the western in summer in the Huai River Basin. The continental high pressure dominated in autumn, the north wind and south wind influenced each other, which is caused the dry in the eastern and wet in the western. The eastern of basin show wet and western of basin show dry, which is caused by the prevailing downdraft and the prevailing southeast in winter. 参考文献 相似文献 引证文献

Characterizing the climatic niche of mast seeding in beech: Evidences of trade-offs between vegetation growth and seed production

Masting is a complex mechanism which is mainly driven by a combination of internal plant resources and climatic conditions. While the driving role of climate in masting is being intensively studied, the interplay among climate, seed production, vegetation growth and phenology still needs further investigation. The objectives of thisstudy were to identify the climatic determinants of different levels of seed production and of NDVI-based vegetation growth and phenology in European beech, and to evaluate if exists a trade-off between these two plant processes. To answer these questions, we used a 25-year-long dataset of beech seed production. We exploited the concept of ecological niche assuming that a mast year can be modeled like a species with variable preferences for different resources, which are the underlying annual climatic conditions; we performed an Ecological Niche Factor Analysis (ENFA), a presence-only modeling tool conventionally used in zoology and botany, and used seasonal (spring, summer, autumn) Standardized Precipitation-Evaporation Index (SPEI) observations, considering the current year (y−0), and up to one (y−1) and two (y−2) years before the masting event. For analyzing the role of vegetation growth and phenology, we used seasonal Normalized Difference Vegetation Index (NDVI) values and associated NDVI-based phenological metrics derived from Landsat imagery. Results indicated the driving role of climate for masting, especially in VHSP years. A moist summer and dry spring at y−2 and a dry summer at y−1 represented the main driving climatic conditions for masting; while a moist spring during the observation year represented the key condition for triggering higher intensities of seed production. Summer NDVI at y−0 and y−1 represented the variables discriminating best between masting and non-masting years and resulted as driven by opposite summer climatic conditions than seed production, thus indicating a trade-off between seed production and vegetation phenology. We concluded that reproduction and vegetation growth act as two different climate-dependent plant responses in beech, in a way that certain conditions through the years promote mast seeding and the opposite conditions favor vegetation growth. The understanding of climate-growth-masting relationships represents indispensable knowledge for providing a holistic view of masting mechanisms and developing adaptive forest management strategies in this species.

EVALUATION OF THE SOIL MOISTURE AGRICULTURAL DROUGHT INDEX (SMADI) AND PRECIPITATION-BASED DROUGHT INDICES IN ARGENTINA

Abstract. Agricultural drought is one of the most critical hazards with regard to intensity, severity, frequency, spatial extension and impact on livelihoods. This is especially true for Argentina, where agricultural exports can represent up to 10% of gross domestic product (GDP), and where drought events for 2018 led to a decrease of nearly 0.5% of GDP. In this work, we investigate the applicability of the Soil Moisture Agricultural Drought Index (SMADI) for detection of droughts in Argentina, and compare its performance with the use of two well-known precipitation-based indices: the Standardized Precipitation Index (SPI) and the Standardized Precipitation- Evaporation Index (SPEI). SMADI includes satellite-based information of soil moisture, surface temperature and vegetation greenness, and was designed to capture the hydric stress on the soil-vegetation ensemble. Results indicate that SMADI has greater capabilities for agricultural drought detection than SPI and SPEI: it was able to recognize more than 83% of the registered emergencies, correctly classifying 75% of them as extreme droughts, and outperforming SPI and SPEI in all the analyzed metrics.

Long lead time drought forecasting using lagged climate variables and a stacked long short-term memory model

Drought forecasting with a long lead time is essential for early warning systems and risk management strategies. The use of machine learning algorithms has been proven to be beneficial in forecasting droughts. However, forecasting at long lead times remains a challenge due to the effects of climate change and the complexities involved in drought assessment. The rise of deep learning techniques can solve this issue, and the present work aims to use a stacked long short-term memory (LSTM) architecture to forecast a commonly used drought measure, namely, the Standard Precipitation Evaporation Index. The model was then applied to the New South Wales region of Australia, with hydrometeorological and climatic variables as predictors. The multivariate interpolated grid of the Climatic Research Unit was used to compute the index at monthly scales, with meteorological variables as predictors. The architecture was trained using data from the period of 1901–2000 and tested on data from the period of 2001–2018. The results were then forecasted at lead times ranging from 1 month to 12 months. The forecasted results were analysed in terms of drought characteristics, such as drought intensity, drought onset, spatial extent and number of drought months, to elucidate how these characteristics improve the understanding of drought forecasting. The drought intensity forecasting capability of the model used two statistical metrics, namely, the coefficient of determination (R2) and root-mean-square error. The variation in the number of drought months was examined using the threat score technique. The results of this study showed that the stacked LSTM model can forecast effectively at short-term and long-term lead times. Such findings will be essential for government agencies and can be further tested to understand the forecasting capability of the presented architecture at shorter temporal scales, which can range from days to weeks.

Atmospheric Circulation as a Factor Contributing to Increasing Drought Severity in Central Europe

AbstractLong‐lasting and severe droughts seriously threaten agriculture, ecosystems, and society. Summer 2018 in central Europe was characterized by unusually persistent heat and drought, causing substantial economic losses, and became a part of a several years long dry period observed across this region. This study assesses the magnitude of the recent drought within a long‐term context and links the increased drought severity to changes in atmospheric circulation. Temporal variability of drought conditions since the late 19th century was analyzed at seven long‐term stations distributed across the Czech Republic using the Palmer Drought Severity Index and the Standardized Precipitation Evaporation Index. The Palmer Z Index and a variation of the Standardized Precipitation Evaporation Index were used to study rapidly emerging short‐term droughts and to link these episodes to atmospheric circulation. Changes in circulation were analyzed through circulation types calculated from flow strength, direction and vorticity in mean sea level pressure data from the National Centers for Environmental Prediction (NCEP)/National Center for Atmospheric Research (NCAR) reanalysis for 1948–2018. Increasing drought severity across the Czech Republic with record‐low values of the drought indices during 2015–2018 was found. The trend was distinctive in both vegetation (April–September) and cold (October–March) periods. The tendency toward more severe droughts in recent decades was linked to changes in frequency of dry and wet circulation types, highlighting the important role of atmospheric circulation in regional climate. It remains an open question whether the significantly increasing frequency of dry circulation types in the vegetation period is related to climate change, or rather represents multidecadal climate variability.

Linking drought indices to impacts in the Liaoning Province of China

Abstract. Drought is an inherent meteorological characteristic of any given region, but is particularly important in China due to its monsoon climate and the “three ladder” landform system. The Chinese government has constructed large-scale water conservation projects since 1949, and developed drought and water scarcity relief frameworks. However, drought still causes huge impacts on water supply, environment and agriculture. China has, therefore, created specialized agencies for drought management called Flood Control and Drought Relief Headquarters, which include four different levels: state, provincial, municipal and county. The impact datasets they collect provide an effective resource for drought vulnerability assessment, and provide validation options for hydro-meteorological indices used in risk assessment and drought monitoring. In this study, we use the statistical drought impact data collected by the Liaoning province Drought Relief Headquarter and meteorological drought indices (Standardized Precipitation Index, SPI and Standard Precipitation Evaporation Index, SPEI) to explore a potential relationship between drought impacts and these indices. The results show that SPI-24 and SPEI-24 are highly correlated to the populations that have difficulties in obtaining drinking water in four out of the six cities studied. Three impacts related to reservoirs and the availability of drinking water for humans and livestock exhibit strong correlations with SPI and SPEI of different accumulated periods. Results reveal that meteorological indices used for drought monitoring and early warning in China can be effectively linked to drought impacts. Further work is exploring how this information can be used to optimize drought monitoring and risk assessment in the whole Liaoning province and elsewhere in China.

Assessment of the potential impacts of climate change on the hydrology at catchment scale: modelling approach including prediction of future drought events using drought indices

In this study, the Distributed Catchment-Scale Model, DiCaSM, was used to study the impact of climate change on the hydrology of the Eden catchment, north east of Scotland. As a first step, the model was successfully calibrated and validated for a 42 years period. The DiCaSM model was then used to study the impact of climate change on the water availability. Data from the UKCP09 Climate change scenarios for the 2010–2039, 2040–2069 and 2070–2099 periods, considering three gas emission scenarios (low, medium and high), were applied. The results indicated that the greatest decrease in streamflow and groundwater recharge was projected to happen under the high emission scenarios towards the end of the century, i.e. between 2070 and 2099. This would mainly be due to the summers becoming drier. Meanwhile, the projected increase in winter precipitation did not contribute much towards groundwater recharge due the projected increases in evapotranspiration and soil moisture deficit.The following drought indices were calculated and were found to be effective in predicting different types of droughts: the Standardized Precipitation Index, SPI, and the Standardized Precipitation Evaporation Index, SPEI, the Reconnaissance Drought Index, RDI, the modified adjusted RDI, the Soil Moisture Deficit, SMD and the Wetness Index, WI. The findings of the study have broader implications in water resources management considering the future changes in climate.

Detection and attribution of lake water loss in the semi‐arid Mongolian Plateau—A case study in the Lake Dalinor

AbstractExtensive lake shrinkages have been observed in the semi‐arid Mongolian Plateau over the past several decades, causing a great challenge to the local socio‐economic sustainability and ecological security. However, issue on the dominant drivers of this water loss has not been well addressed. In this work, variations in water area and water level of the Lake Dalinor during the period 1984–2018 were tracked using the satellite altimetry and Landsat images. Dominant drivers for the derived variations were then investigated based on the Standard Precipitation Evaporation Index (SPEI) analysis and the Vegetation Interface Processes model. Results showed that both the water area and water level displayed a two‐stage variation with a slightly increasing trend before 2000 and a drastic decline since 2000 (p < 0.01). The SPEI analysis revealed a close relationship between water level variations and the cumulative water anomalies at seasonal and annual scales. The model‐based attribution analysis suggested that climate change and anthropogenic impacts contributed 76.5% and 23.5% of the observed declining trend, respectively. Climate warming and drying dominated the water level decline since 2000, while the large‐scale conversion from the natural grassland to the seeded pasture might have accelerated this water loss process. This study confirmed the prominent impacts of climate warming and drying on the lake desiccation in the semi‐arid Mongolian Plateau and highlighted the gradually intensifying impacts of human activities such as grassland conversion on the lake water budget.

Drought-related hot summers: A joint probability analysis in the Iberian Peninsula

Droughts and hot extremes are major sources of risk to several socio-economic activities and their impacts are expected to increase under future global warming. Moreover, the simultaneous or sequential occurrence (compound events) of different climate extremes may lead to the amplification of the associated impacts. Even though the latest efforts in assessing hot and dry extremes and their interactions, the development of models describing the joint behavior of climate extremes is still a challenge. To contribute to the understanding of these compound events, we propose to assess the probability of extremely hot summer days in the Iberian Peninsula (IP) being preceded by drought events in spring and early summer, based on their joint probability distribution through copula theory. The precursor drought hazard was characterized by the Standardized Precipitation Evaporation Index (SPEI) for the months of May, June and July for different timescales (3-, 6- and 9-months). The Number of Hot Days per month (NHD) summed over the months of July and August were considered for modelling. The dependence structure between SPEI and NHD was very well identified for the most of the IP's regions by asymmetrical copulas with upper tail dependence (except in northwestern regions), suggesting that compound hot and dry extremes are strongly associated. The results show that the transition from previous wet to dry regimes increases substantially the probability of exceeding summer NHD extreme values, depending on the region and the drought timescale and target month. The results suggest a spatial heterogeneity over the IP when characterizing the influence of water deficits on following summer extreme temperatures, whereas northeastern, western and central regions were found to be the regions more prone to summer hot extremes induced by dryness, in contrast to southwestern, northwestern and southeastern regions (depending on the month and the timescale). This study provides estimates of the probability of drought-related hot extremes in the summer of the IP for different regions, which could be an important tool for responsible authorities to mitigate the impacts magnified by the interactions between the different hazards.

Tree-ring lignin proxies in Larix gmelinii forest growing in a permafrost area of northeastern China: Temporal variation and potential for climate reconstructions

The hydrogen and carbon stable isotope ratios (δ2HLM and δ13CLM values) of tree-ring lignin methoxy groups have recently been recognized as valuable palaeoclimate indicators. However, the environmental factors but also sample preparation processes that might cause variations of δ2HLM and δ13CLM values have been not fully explored. Furthermore, the temporal dynamics of wood lignin content on both isotopes hasn’t been investigated. To investigate the effects of total lignin content, removal of lipids and differences between individual and pooled tree-ring series on isotopic values, we analyzed ring samples of dominant Larix gmelinii trees from a typical permafrost region of northeastern China. We also examined relationships between δ2HLM and δ13CLM values with annual-resolution lignin content and climatic parameters over the common period of overlap 1901 to 2013. We found that the inter-tree variability between the individual and pooled isotopic chronologies was consistent, and the removal of lipids did not change the results of stable isotope measurements. The total lignin content shifted between the heartwood and sapwood, whereas the content of guaiacyl monomers (G-lignin), the only donor of methoxy groups, remained constant. Correlations analysis between lignin proxies and climate variables indicated that the heartwood total lignin content was positively correlated with the intrinsic water-use efficiency, temperature and vapor-pressure deficit (VPD), but negatively with the early-growing-season standardized precipitation evaporation index. Whilst δ2HLM values were positively correlated with the growing season temperature (May to August) and VPD, particularly in August, δ13CLM series showed insignificant correlation. Therefore, we suggest that the total lignin content and δ2HLM values of tree rings (both individual and pooled series) are suitable for tracking environment dynamics, as they can reveal the climate responses of tree growth.

Drought Impacts on Vegetation in Southeastern Europe

We evaluated the response of vegetation’s photosynthetic activity to drought conditions from 1998 to 2014 over Romania and the Republic of Moldova. The connection between vegetation stress and drought events was assessed by means of a correlation analysis between the monthly Standardized Precipitation Evaporation Index (SPEI), at several time scales, and the Normalized Difference Vegetation Index (NDVI), as well as an assessment of the simultaneous occurrence of extremes in both indices. The analysis of the relationship between drought and vegetation was made for the growing season (from April to October of the entire period), and special attention was devoted to the severe drought event of 2000/2001, considered as the driest since 1961 for the study area. More than three quarters (77%) of the agricultural land exhibits a positive correlation between the two indices. The sensitivity of crop areas to drought is strong, as the impacts were detected from May to October, with a peak in July. On the other hand, forests were found to be less sensitive to drought, as the impacts were limited mostly to July and August. Moreover, vegetation of all land cover classes showed a dependence between the sign of the correlation and the elevation gradient. Roughly 60% (20%) of the study domain shows a concordance of anomalously low vegetation activity with dry conditions of at least 50% (80%) in August. By contrast, a lower value of concordance was observed over the Carpathian Mountains. During the severe drought event of 2000/2001, a decrease in vegetation activity was detected for most of the study area, showing a decrease lasting at least 4 months, between April and October, for more than two thirds (71%) of the study domain.

Short-Term Spatio-Temporal Drought Forecasting Using Random Forests Model at New South Wales, Australia

Droughts can cause significant damage to agriculture and water resources, leading to severe economic losses and loss of life. One of the most important aspect is to develop effective tools to forecast drought events that could be helpful in mitigation strategies. The understanding of droughts has become more challenging because of the effect of climate change, urbanization and water management; therefore, the present study aims to forecast droughts by determining an appropriate index and analyzing its changes, using climate variables. The work was conducted in three different phases, first being the determination of Standard Precipitation Evaporation Index (SPEI), using global climatic dataset of Climate Research Unit (CRU) from 1901–2018. The indices are calculated at different monthly intervals which could depict short-term or long-term changes, and the index value represents different drought classes, ranging from extremely dry to extremely wet. However, the present study was focused only on forecasting at short-term scales for New South Wales (NSW) region of Australia and was conducted at two different time scales, one month and three months. The second phase involved dividing the data into three sample sizes, training (1901–2010), testing (2011–2015) and validation (2016–2018). Finally, a machine learning approach, Random Forest (RF), was used to train and test the data, using various climatic variables, e.g., rainfall, potential evapotranspiration, cloud cover, vapor pressure and temperature (maximum, minimum and mean). The final phase was to analyze the performance of the model based on statistical metrics and drought classes. Regarding this, the performance of the testing period was conducted by using statistical metrics, Coefficient of Determination (R2) and Root-Mean-Square-Error (RMSE) method. The performance of the model showed a considerably higher value of R2 for both the time scales. However, statistical metrics analyzes the variation between the predicted and observed index values, and it does not consider the drought classes. Therefore, the variation in predicted and observed SPEI values were analyzed based on different drought classes, which were validated by using the Receiver Operating Characteristic (ROC)-based Area under the Curve (AUC) approach. The results reveal that the classification of drought classes during the validation period had an AUC of 0.82 for SPEI 1 case and 0.84 for SPEI 3 case. The study depicts that the Random Forest model can perform both regression and classification analysis for drought studies in NSW. The work also suggests that the performance of any model for drought forecasting should not be limited only through statistical metrics, but also by examining the variation in terms of drought characteristics.

Skill of large-scale seasonal drought impact forecasts

Abstract. Forecasting of drought impacts is still lacking in drought early-warning systems (DEWSs), which presently do not go beyond hazard forecasting. Therefore, we developed drought impact functions using machine learning approaches (logistic regression and random forest) to predict drought impacts with lead times up to 7 months ahead. The observed and forecasted hydrometeorological drought hazards – such as the standardized precipitation index (SPI), standardized precipitation evaporation index (SPEI), and standardized runoff index (SRI) – were obtained from the The EU-funded Enhancing Emergency Management and Response to Extreme Weather and Climate Events (ANYWHERE) DEWS. Reported drought impact data, taken from the European Drought Impact Report Inventory (EDII), were used to develop and validate drought impact functions. The skill of the drought impact functions in forecasting drought impacts was evaluated using the Brier skill score and relative operating characteristic metrics for five cases representing different spatial aggregation and lumping of impacted sectors. Results show that hydrological drought hazard represented by SRI has higher skill than meteorological drought represented by SPI and SPEI. For German regions, impact functions developed using random forests indicate a higher discriminative ability to forecast drought impacts than logistic regression. Moreover, skill is higher for cases with higher spatial resolution and less lumped impacted sectors (cases 4 and 5), with considerable skill up to 3–4 months ahead. The forecasting skill of drought impacts using machine learning greatly depends on the availability of impact data. This study demonstrates that the drought impact functions could not be developed for certain regions and impacted sectors, owing to the lack of reported impacts.

Saharan Hot and Dry Sirocco Winds Drive Extreme Fire Events in Mediterranean Tunisia (North Africa)

With hot and dry summers, the Mediterranean basin is affected by recurrent fires. While drought is the major driver of the seasonal and inter-annual fire distribution in its northern and mildest climate conditions, some extreme fire events are also linked to extreme winds or heat waves. The southern part of the Mediterranean basin is located at the driest range of the Mediterranean bioclimate and is influenced by Saharan atmospheric circulations, leading to extreme hot and dry episodes, called Sirocco, and potentially acting as a major contributor to fire hazard. The recently created fire database for Tunisia was used to investigate the ±10-day pre- and post-fire timeframe of daily weather conditions associated with fire events over the 1985–2006 period. Positive anomalies in minimum and maximum temperatures, negative anomalies in air relative humidity, and a preferential south-eastern wind during fire events were identified, which were characteristic of Sirocco winds. +7 °C anomalies in air temperature and −30% in relative air humidity were the critical thresholds for the most extreme fire conditions. In addition, meteorological anomalies started two days before fire events and lasted for three days after for large fires >400 ha, which suggests that the duration of the Sirocco event is linked with fire duration and final fire size. Lastly, the yearly number of intense Sirocco events better explained the inter-annual variability of burned area over the 1950–2006 period than summer drought based on Standardized Precipitation Evaporation Index (SPEI) indices.

Temporal Hydrological Drought Index Forecasting for New South Wales, Australia Using Machine Learning Approaches

Droughts can cause significant damage to agriculture and water resources leading to severe economic losses. One of the most important aspects of drought management is to develop useful tools to forecast drought events, which could be helpful in mitigation strategies. The recent global trends in drought events reveal that climate change would be a dominant factor in influencing such events. The present study aims to understand this effect for the New South Wales (NSW) region of Australia, which has suffered from several droughts in recent decades. The understanding of the drought is usually carried out using a drought index, therefore the Standard Precipitation Evaporation Index (SPEI) was chosen as it uses both rainfall and temperature parameters in its calculation and has proven to better reflect drought. The drought index was calculated at various time scales (1, 3, 6, and 12 months) using a Climate Research Unit (CRU) dataset. The study focused on predicting the temporal aspect of the drought index using 13 different variables, of which eight were climatic drivers and sea surface temperature indices, and the remainder were various meteorological variables. The models used for forecasting were an artificial neural network (ANN) and support vector regression (SVR). The model was trained from 1901–2010 and tested for nine years (2011–2018), using three different performance metric scores (coefficient of determination (R2), root mean square error (RMSE), and mean absolute error (MAE). The results indicate that ANN was better than SVR in predicting temporal drought trends, with the highest R2 value of 0.86 for the former compared to 0.75 for the latter. The study also reveals that sea surface temperatures and the climatic index (Pacific Decadal Oscillation) do not have a significant effect on the temporal drought aspect. The present work can be considered as a first step, wherein we only study the temporal trends, towards the use of climatological variables and drought incidences for the NSW region.

Short-Term and Medium-Term Drought Forecasting Using Generalized Additive Models

Forecasting extreme hydrological events is critical for drought risk and efficient water resource management in semi-arid environments that are prone to natural hazards. This study aimed at forecasting drought conditions in a semi-arid region in north-eastern South Africa. The Standardized Precipitation Evaporation Index (SPEI) was used as a drought-quantifying parameter. Data for SPEI formulation for eight weather stations were obtained from South Africa Weather Services. Forecasting of the SPEI was achieved by using Generalized Additive Models (GAMs) at 1, 6, and 12 month timescales. Time series decomposition was done to reduce time series complexities, and variable selection was done using Lasso. Mild drought conditions were found to be more prevalent in the study area compared to other drought categories. Four models were developed to forecast drought in the Luvuvhu River Catchment (i.e., GAM, Ensemble Empirical Mode Decomposition (EEMD)-GAM, EEMD-Autoregressive Integrated Moving Average (ARIMA)-GAM, and Forecast Quantile Regression Averaging (fQRA)). At the first two timescales, fQRA forecasted the test data better than the other models, while GAMs were best at the 12 month timescale. Root Mean Square Error values of 0.0599, 0.2609, and 0.1809 were shown by fQRA and GAM at the 1, 6, and 12 month timescales, respectively. The study findings demonstrated the strength of GAMs in short- and medium-term drought forecasting.

Statistical modelling of drought-related yield losses using soil moisture-vegetation remote sensing and multiscalar indices in the south-eastern Europe

Meteorological and agricultural information coupled with remote sensing observations has been used to assess the effectiveness of satellite-derived indices in yield estimations. The estimate yield models generated by both the regression (MLR) and Bayesian network (BBN) algorithms and their levels of predictive skill were assessed. The enhanced vegetation index (EVI2), soil water index (SWI), standardized precipitation evaporation index (SPEI) have been considered predictors for three rainfed crops (maize, sunflower and grapevine) grown in 37 districts in the Republic of Moldova (RM). We used the weekly EVI2, which was collected by MODIS instruments aboard the Terra satellite with a 250m × 250m spatial resolution and aggregated for each district during the 2000–2018 period. We also used the weekly SWI, which was collected from the ASCAT instruments with a 12 km x 12 km spatial resolution and aggregated for each district at the topsoil (0–40 cm; SWI-12) and the root-zone layer (0–100 cm; SWI-14) during 2000–2018. The multiscalar SPEI during 1951–2018 farming years proved to be a significant addition to the remote sensing indices and led to the development of a model that improved the yield assessment. The study also summarized (i) the optimal time window of satellite-derived SWIi and EVI2i for yield estimation, and (ii) the capability of remotely sensed indices for representing the spatio–temporal variations of agricultural droughts. We developed statistical soil-vegetation-atmosphere models to explore drought-related yield losses. The skill scores of the sunflower MLR and BBN models were higher than those for the maize and grape models and were able to estimate yields with reasonable accuracy and predictive power. The accurate estimation of maize, sunflower and grapevine yields was observed two months before the harvest (RMSE of ∼1.2 tha-1). Despite the fact that summer crops (maize, sunflower) are able to develop a root system that uses the entire root zone depth, however, the SWI-12 had the stronger correlation with crop yield, then SWI-14. This explains much better the fit between yields of the crops and SWI-12, which represents soil moisture anomaly in the key rooting layer of soil. In any case, all summer crops showed negative correlations with each of the remote sensing soil moisture indices in the early and middle of the growing season, with SWI-12 performing better than SWI-14. Based on the crop-specific soil moisture model, we found that topsoil moisture declines in the most drought-susceptible crop growth stages, which indicates that RM is a good candidate for studying drought persists as main driver of rainfed yield losses in the south-eastern Europe.

Tree mortality and biomass loss in drought-affected forests of East Texas, USA

Changes in tree mortality due to severe drought can alter forest structure, composition, dynamics, ecosystem services, carbon fluxes, and energy interactions between the atmosphere and land surfaces. We utilized long-term (2000‒2017, 3 full inventory cycles) Forest Inventory and Analysis (FIA) data to examine tree mortality and biomass loss in drought-affected forests for East Texas, USA. Plots that experienced six or more years of droughts during those censuses were selected based on 12-month moderate drought severity [Standardized Precipitation Evaporation Index (SPEI) -1.0]. Plots that experienced other disturbances and inconsistent records were excluded from the analysis. In total, 222 plots were retained from nearly 4000 plots. Generalized nonlinear mixed models (GNMMs) were used to examine the changes in tree mortality and recruitment rates for selected plots. The results showed that tree mortality rates and biomass loss to mortality increased overall, and across tree sizes, dominant genera, height classes, and ecoregions. An average mortality rate of 5.89% year−1 during the study period could be incited by water stress created by the regional prolonged and episodic drought events. The overall plot and species-group level recruitment rates decreased during the study period. Forest mortality showed mixed results regarding basal area and forest density using all plots together and when analyzed the plots by stand origin and ecoregion. Higher mortality rates of smaller trees were detected and were likely compounded by density-dependent factors. Comparative analysis of drought-induced tree mortality using hydro-meteorological data along with drought severity and length gradient is suggested to better understand the effects of drought on tree mortality and biomass loss around and beyond East Texas in the southeastern United States.