AN EARLY WARNING SYSTEM FOR DROUGHT MANAGEMENT USING THE PALMER DROUGHT INDEX1

ABSTRACT: The Palmer Drought Severity Index (PDSI) is used in a non-homogeneous Markov chain model to characterize the stochastic behavior of drought. Based on this characterization an early warning system in the form of a decision tree enumerating all possible sequences of drought progression is proposed for drought management. Besides yielding probabilities of occurrence of different drought severity classes, the method associates a secondary measure in terms of likely cumulative precipitation deficit to provide timely guidance in deciding drought mitigation actions. The proposed method is particularly useful for water availability task forces in various states for issuing drought warnings in advance. The applicability of the technique is illustrated for the Tidewater climatic division of Virginia.     

DROUGHT,TREE RINGS,AND RESERVOIR DESIGN1

Droughts constitute one of the most important factors affecting the design and operation of water resources infrastructure. Hydrologists ascertain their duration, severity, and pattern of recurrence from instrumental records of precipitation or stream‐flow. Under suitable conditions, and with proper analysis, tree rings obtained from long living, climate sensitive species of trees can extend instrumental records of streamflow and precipitation over periods spanning several centuries. Those tree‐ring “reconstructions” provide a valuable insight about climate variability and drought occurrence in the Holocene, and yield long term hydrological data useful in the design of water infrastructure. This work presents a derivation of drought risk based on a renewal model of drought recurrence, a brief review of the basic theory of tree‐ring reconstructions, and a stochastic model for optimizing the design of water supply reservoirs. Examples illustrate the methodology developed in this work and the supporting role that tree‐ring reconstructed streamflow can play in characterizing hydrologic variability.     

HYDROLOGIC SCENARIOS FOR SEVERE SUSTAINED DROUGHT IN THE SOUTHWESTERN UNITED STATES1

ABSTRACT: This paper considers the risk of drought and develops drought scenarios for use in the study of severe sustained drought in the Southwestern United States. The focus is on the Colorado River Basin and regions to which Colorado River water is exported, especially southern California, which depends on water from the Colorado River. Drought scenarios are developed using estimates of unimpaired historic streamflow as well as reconstructions of streamflow based on tree ring widths. Drought scenarios in the Colorado River Basin are defined on the basis of annual flow at Lees Ferry. The risk, in terms of return period, of the drought scenarios developed, is assessed using stochastic models.     

ACCEPTING THE STANDARDIZED PRECIPITATION INDEX: A CALCULATION ALGORITHM1

ABSTRACT: The Palmer Drought Severity Index (PDSI) has been calculated for about 30 years as a means of providing a single measure of meteorological drought severity. It was intended to retrospectively look at wet and dry conditions using water balance techniques. The Standardized Precipitation Index (SPI) is a probability index that was developed to give a better representation of abnormal wetness and dryness than the Palmer indices. Before the user community will accept the SPI as an alternative to the Palmer indices, a standard method must be developed for computing the index. Standardization is necessary so that all users of the index will have a common basis for both spatial and temporal comparison of index values. If different probability distributions and models are used to describe an observed series of precipitation, then different SPI values may be obtained. This article describes the effect on the SPI values computed from different probability models as well as the effects on dry event characteristics. It is concluded that the Pearson Type III distribution is the “best” universal model, and that the reliability of the SPI is sample size dependent. It is also concluded that because of data limitations, SPIs with time scales longer than 24 months may be unreliable. An internet link is provided that will allow users to access Fortran 77 source code for calculating the SPI.     

THE TREE-RING RECORD OF SEVERE SUSTAINED DROUGHT1

ABSTRACT: Frequent and persistent droughts exacerbate the problems caused by the inherent scarcity of water in the semiarid to arid parts of the southwestern United States. The occurrence of drought is driven by climatic variability, which for years before about the beginning of the 20th century in the Southwest must be inferred from proxy records. As part of a multidisciplinary study of the potential hydrologic impact of severe sustained drought on the Colorado River, the physical basis and limitations of tree rings as indicators of severe sustained drought are reviewed, and tree-ring data are analyzed to delineate a “worst-case” drought scenario for the Upper Colorado River Basin (UCRB). Runs analysis of a 121-site tree-ring network, 1600–1962, identifies a four-year drought in the 1660s as the longest-duration large-scale drought in the Southwest in the recent tree-ring record. Longer tree-ring records suggest a much longer and more severe drought in 1579–1598. The regression estimate of the mean annual Colorado River flow for this period is 10.95 million acre-feet, or 81 percent of the long-term mean. The estimated flows for the 1500s should be used with caution in impact studies because sample size is small and some reconstructed values are extrapolations.     

Past climate, future perspective: an exploratory analysis using climate proxies and drought risk assessment to inform water resources management and policy in Maine, USA

In recent decades, significant progress has been made toward reconstructing the past climate record based on environmental proxies, such as tree rings and ice core records. However, limited examples of research that utilizes such data for water resources decision-making and policy exist. Here, we use the reconstructed record of Palmer Drought Severity Index (PDSI), dating back to 1138AD to understand the nature of drought occurrence (severity and duration) in the state of Maine. This work is motivated by the need to augment the scientific basis to support the water resources management and the emerging water allocation framework in Maine (Maine Department of Environmental Protection, Chapter 587). Through a joint analysis of the reconstructed PDSI and historical streamflow record for twelve streams in the state of Maine, we find that: (a) the uncertainties around the current definition of natural drought in the Chapter 587 (based on the 20th century instrumental record) can be better understood within the context of the nature and severity of past droughts in this region, and (b) a drought index provides limited information regarding at-site hydrologic variations. To fill this knowledge gap, a drought index-based risk assessment methodology for streams across the state is developed. Based on these results, the opportunities for learning and challenges facing water policies in a changing hydroclimate are discussed.     

HYDROLOGICAL ASPECTS OF THE 1988 DROUGHT IN ILLINOIS1

ABSTRACT: Drought is evaluated in terms of the magnitude and duration of the 1988 spring and summer precipitation shortfall, and according to various components of the hydrologic budget, both surface and sub-surface. The response time of some of these components is investigated, relative to the time of precipitation. Individual water users perceived a beginning and ending of the drought at different times relative to their activities. Some statistics better describe some components of a drought to some users, and better answer some questions, than do others.     

NEW ENGLAND DROUGHT AND RELATIONS WITH LARGE SCALE ATMOSPHERIC CIRCULATION PATTERNS1

ABSTRACT: To provide a basis for regional hydroclimatic forecasting, New England (NE) precipitation and streamflow are compared with indices for the El Niño/Southern Oscillation, the Pacific North American (PNA) pattern, and the North Atlantic Oscillation (NAO). Significant positive correlations are found between the NAO index and monthly streamflow at western inland locations, with the strongest seasonal correlations occurring in winter. Smoothed records for the winter NAO and winter streamflow are highly correlated at some sites, suggesting that interrelationships are most significant in the low frequency spectrum. However, correlations between the NAO and precipitation are not significant, so further examination of other factors is needed to explain the relationship between the NAO and streamflow. NAO related regional air temperature, sea surface temperature (SST), storm tracking, and snowfall variability are possible mechanisms for the observed teleconnection. Exceptionally cool regional air temperatures, and SSTs, and unique regional storm track patterns characterized NE's climate during the famous 1960s drought, suggesting that concurrent (persistent) negative NAO conditions may have contributed to the severity of that event. Monthly and winter averaged regional streamflow variability are also significantly correlated with the PNA index. This, along with results from previous studies, suggests that tropospheric wave character and associated North Pacific SST anomalies are also related to NE regional drought conditions.     

A PROPOSED METHOD FOR DROUGHT MONITORING1

ABSTRACT: Existing definitions of drought have focused on limited hydrologic indicators and are less effective for the purpose of drought monitoring. This study uses historical records of streamflow, precipitation, ground water, temperature, and lake elevation to define drought. Based on the method of truncation, drought durations and conditional probabilities of each indicator were estimated to define the drought severity levels, namely, 70 percent, 80 percent, 90 percent, and 95 percent. A drought monitoring method was developed by a combination of truncation level, duration, and conditional probabilities of five indicators. A six-month period of the 1988 drought in the central Ohio region was used to test the monitoring method. It was found that the developed method could effectively detect an occurrence of drought.     

DROUGHT INDICATED IN CARBON-13/CARBON-12 RATIOS OF SOUTHWESTERN TREE RINGS1

ABSTRACT: Stomatal closure during periods of moisture deficiency should theoretically lead to elevated ¹³C/¹²C ratios as reduction of available CO₂ leads to diminished photosynthetic discrimination against ¹³C in favor of ¹²C. Stable-carbon isotope ratio chronologies developed from 5-yr tree-ring groups at 17 sites in six southwestern states were tested for a drought relationship by first fitting a spline curve to each chronology to remove the long-term trend and calculating indices as the ratio of actual to spline curve value. The time series of “Del Indices” so developed are significantly correlated with 5-yr mean Palmer Hydrological Drought Indices (post-1930 period) and reconstructed July Palmer Drought Severity Indices from respective areas. Overall, in the period since 1790, the driest pentads were 1900–04 and 1960–64, whereas the wettest were 1980–84 and 1915–19. Maps of drought represented for two pentads seem to be reasonable representations, although spatial correlations of Del Indices with PHDI were generally not significant. These Del Index drought reconstructions may provide a useful measure of past physiological response to drought (stomatal closure), although the present cost of analysis would prevent this from being a routine method.     

MEASURING DROUGHT IMPACTS: THE ILLINOIS CASE1

ABSTRACT: Drought is an interaction between physical processes and human activities. This study quantified the impacts of precipitation deficiencies on streamflow, reservoirs, and shallow ground water supplies. An in-depth analysis of newspaper accounts of droughts between paired cities, one in drought and one not in drought, were used to measure the differences in the types of drought impacts, and in the time of onset of impacts as related to developing precipitation deficiencies. Precipitation deficiencies related to the onset and the magnitude of surface water supply adjustments, and to shallow ground water problems, were established. Thus, monitoring and prediction of the onset and magnitude of drought problems can now be done from readily available data on precipitation deficiencies. Newspapers were found to be reliable indicators for the timing of drought impacts and adjustments as precipitation deficiency develops. A review of local and state adjustments during two recent droughts revealed most decision makers lacked information and experience in dealing with drought.     

Forecasting the Probability of Future Groundwater Levels Declining Below Specified Low Thresholds in the Conterminous U.S.

We present a logistic regression approach for forecasting the probability of future groundwater levels declining or maintaining below specific groundwater‐level thresholds. We tested our approach on 102 groundwater wells in different climatic regions and aquifers of the United States that are part of the U.S. Geological Survey Groundwater Climate Response Network. We evaluated the importance of current groundwater levels, precipitation, streamflow, seasonal variability, Palmer Drought Severity Index, and atmosphere/ocean indices for developing the logistic regression equations. Several diagnostics of model fit were used to evaluate the regression equations, including testing of autocorrelation of residuals, goodness‐of‐fit metrics, and bootstrap validation testing. The probabilistic predictions were most successful at wells with high persistence (low month‐to‐month variability) in their groundwater records and at wells where the groundwater level remained below the defined low threshold for sustained periods (generally three months or longer). The model fit was weakest at wells with strong seasonal variability in levels and with shorter duration low‐threshold events. We identified challenges in deriving probabilistic‐forecasting models and possible approaches for addressing those challenges.     

Five Hundred Years of Hydrological Drought in the Upper Colorado River Basin1

Abstract: This article evaluates drought scenarios of the Upper Colorado River basin (UCRB) considering multiple drought variables for the past 500 years and positions the current drought in terms of the magnitude and frequency. Drought characteristics were developed considering water‐year data of UCRB’s streamflow, and basin‐wide averages of the Palmer Hydrological Drought Index (PHDI) and the Palmer Z Index. Streamflow and drought indices were reconstructed for the last 500 years using a principal component regression model based on tree‐ring data. The reconstructed streamflow showed higher variability as compared with reconstructed PHDI and reconstructed Palmer Z Index. The magnitude and severity of all droughts were obtained for the last 500 years for historical and reconstructed drought variables and ranked accordingly. The frequency of the current drought was obtained by considering two different drought frequency statistical approaches and three different methods of determining the beginning and end of the drought period (annual, 5‐year moving, and ten year moving average). It was concluded that the current drought is the worst in the observed record period (1923‐2004), but 6th to 14th largest in terms of magnitude and 1st to 12th considering severity in the past 500 years. Similarly, the current drought has a return period ranging from 37 to 103 years based on how the drought period was determined. It was concluded that if the 10‐year moving average is used for defining the drought period, the current drought appears less severe in terms of magnitude and severity in the last 500 years compared with the results using 1‐ and 5‐year averages.     

Assessment of Future Drought Conditions in the Chesapeake Bay Watershed

Impacts of climate change on the severity and intensity of future droughts can be evaluated based on precipitation and temperature projections, multiple hydrological models, simulated hydrometeorological variables, and various drought indices. The objective of this study was to assess climate change impacts on future drought conditions and water resources in the Chesapeake Bay (CB) watershed. In this study, the Soil and Water Assessment Tool (SWAT) and the Variable Infiltration Capacity model were used to simulate a Modified Palmer Drought Severity Index (MPDSI), a Standardized Soil Moisture index (SSI), a Multivariate Standardized Drought Index (MSDI), along with Coupled Model Intercomparison Project Phase 5 (CMIP5) climate models for both historical and future periods (f1: 2020‐2049, f2: 2050‐2079). The results of the SSI suggested that there was a general increase in agricultural droughts in the entire CB watershed because of increases in surface and groundwater flow and evapotranspiration. However, MPDSI and MSDI showed an overall decrease in projected drought occurrences due to the increases in precipitation in the future. The results of this study suggest that it is crucial to use multiple modeling approaches with specific drought indices that combine the effects of both precipitation and temperature changes.     

COPING WITH A SEVERE SUSTAINED DROUGHT ON THE COLORADO RWER: INTRODUCTION AND OVERVIEW1

ABSTRACT: In arid regions of rapid economic and population growth, adverse effects of droughts are likely to be increasingly serious. This article presents an introduction and overview of the papers collected in this special issue of the Water Resources Bulletin. The papers report on the second phase of a study of the impacts of and responses to a potential severe sustained drought in the Colorado River Basin in the southwestern U.S. The analyses were performed by a consortium of researchers from universities and the private sector located throughout the Basin. Tree ring studies suggest that droughts of duration and magnitude much more serious than any found in the modern records probably occurred in the Basin during earlier centuries. Taking the present-day configuration of the storage and diversion structures and the economic conditions in the Basin as the base-point, the general objectives of the study are three: first, to define a representative Severe Sustained Drought (SSD) and assess its hydrologic impacts; second, to forecast the economic, social and environmental impacts on the southwestern U.S.; and finally, to assess alternative institutional arrangements for coping with an SSD. The evaluation of impacts and policies was conducted with two distinct modeling approaches. One involved hydrologic-economic optimization modeling where water allocation institutions are decision variables. The second was a simulation-gaming approach which allowed “players” representing each basin state to interact in a real-time decision making mode in response to the unfolding drought.     

COLUMBIA RIVER FLOW AND DROUGHT SINCE 17501

ABSTRACT: A network of 32 drought sensitive tree‐ring chronologies is used to reconstruct mean water year flow on the Columbia River at The Dalles, Oregon, since 1750. The reconstruction explains 30 percent of the variability in mean water year (October to September) flow, with a large portion of unexplained variance caused by underestimates of the most severe low flow events. Residual statistics from the tree‐ring reconstruction, as well as an identically specified instrumental reconstruction, exhibit positive trends over time. This finding suggests that the relationship between drought and streamflow has changed over time, supporting results from hydrologic models, which suggest that changes in land cover over the 20th Century have had measurable impacts on runoff production. Low pass filtering the flow record suggests that persistent low flows during the 1840s were probably the most severe of the past 250 years, but that flows during the 1930s were nearly as extreme. The period from 1950 to 1987 is anomalous in the context of this record for having no notable multiyear drought events. A comparison of the flow reconstruction to paleorecords of the Pacific Decadal Oscillation (PDO) and El Nino/Southern Oscillation (ENSO) support a strong 20th Century link between large scale circulation and streamflow, but suggests that this link is very weak prior to 1900.     

USE AND EFFECTIVENESS OF MUNICIPAL WATER RESTRICTIONS DURING DROUGHT IN COLORADO1

ABSTRACT: Drought conditions in the summer of 2002 prompted several cities along Colorado's Front Range to enact restrictions on outdoor water use, focusing primarily on limiting the frequency of lawn watering. The different approaches utilized by eight water providers were tracked to determine the level of water savings achieved, measured as a comparison of 2002 usage to 2000 to 2001 average usage, and also based on a statistical estimate of 2002 “expected use” that accounts for the impact of drought conditions on demand. Mandatory restrictions were shown to be an effective tool for drought coping. During periods of mandatory restrictions, savings measured in expected use per capita ranged from 18 to 56 percent, compared to just 4 to 12 percent savings during periods of voluntary restrictions. As anticipated, providers with the most stringent restrictions achieved the greatest savings.     

SPATIAL COMPARABILITY OF THE PALMER DROUGHT SEVERITY INDEX1

ABSTRACT: The Palmer Drought Severity Index, which is intended to be of reasonable comparable local significance both in space and time, has been extensively used as a measure of drought for both agricultural and water resource management. This study examines the spatial comparability of Palmer's (1965) definition of severe and extreme drought. Index values have been computed for 1035 sites with at least 60 years of record that are scattered across the contiguous United States, and quantile values corresponding to a specified index value were calculated for given months and then mapped. The analyses show that severe or extreme droughts, as defined by Palmer (1965), are not spatially comparable in terms of identifying rare events. The wide variation across the country in the frequency of occurrence of Palmer's (1965) extreme droughts reflects the differences in the variability of precipitation, as well as the average amount of precipitation. It is recommended first, that a drought index be developed which considers both variability and averages; and second, that water resource managers and planners define a drought in terms of an index value that corresponds to the expected quantile (return period) of the event.     

COMPARING THE PALMER DROUGHT INDEX AND THE STANDARDIZED PRECIPITATION INDEX1

ABSTRACT: The Palmer Drought Index (PDI) is used as an indicator of drought severity, and a particular index value is often the signal to begin or discontinue elements of a drought contingency plan. The Standardized Precipitation Index (SPI) was recently developed to quantify a precipitation deficit for different time scales. It was designed to be an indicator of drought that recognizes the importance of time scales in the analysis of water availability and water use. This study compares historical time series of the PDI with time series of the corresponding SPI through spectral analysis. Results show that the spectral characteristics of the PDI vary from site to site throughout the U.S., while those of the SPI do not vary from site to site. They also show that the PDI has a complex structure with an exceptionally long memory, while the SPI is an easily interpreted, simple moving average process.     

A TREE-RING RECONSTRUCTION OF DROUGHT IN SOUTHERN CALIFORNIA1

ABSTRACT: Indices of annual diameter growth of trees were used to reconstruct drought in southern California back to A.D. 1700. A regional Palmer Drought Index served as predictand and tree-ring indices from eight sites as predictors in multiple linear regression analyses that yielded the prediction (reconstruction) equations. The regression explained 69 percent of the variance in Palmer Index in the period of calibration. The long-term reconstruction indicated that drought was rare in the first half of the current century relative to other discrete 50-year periods, and that based on evidence to date the last half of the 20th century may well turn out to be the most drought prone since A.D. 1700 in southern California.