1. Introduction
Since almost one-half of the earth’s terrestrial surface is susceptible to drought, it is a widespread phenomenon that has significant social, economic, and environmental impacts (Korgan 1997). As a result, it is necessary to be able to model and predict the duration, extent, and severity of drought in order to mitigate its negative consequences. To accomplish these tasks, however, it is first necessary to have a thorough understanding of the effects of drought on the processes linking the land surface to the atmosphere. In this study, the authors investigate one of those linkages, the latent heat flux (λE), by exploring how the influence of the factors driving λE varies with changing environmental conditions during drought. The understanding gained through this study may then be used to enhance land surface and atmospheric models so that they are better able to describe drought.
At the most basic level, drought exists when there is not enough water to meet the demand for it (Redmond 2002). Recognizing drought is a relative condition wherein there is an imbalance between the supply of water and the demand for water (Heim 2002), the American Meteorological Society (1997) has defined four categories of drought: meteorological or climatological drought, agricultural drought, hydrological drought, and socioeconomic drought. Meteorological drought may be defined as a prolonged period during which there is an absence or reduction in precipitation. Agricultural drought is defined as a period during which there is insufficient soil moisture to meet the need of plants, most typically nonirrigated crops. Hydrological drought, which can persist long after the meteorological drought has ended, is defined in terms of the effects of meteorological drought on streamflow, groundwater, and the other components of hydrological systems. Finally, socioeconomic drought is defined in terms of the effects on the availability of economic goods and services. This study emphasized meteorological drought and, through its impacts on soil moisture and λE, agricultural drought. However, it is important to recognize that no one category of drought can be considered in isolation from the others because all are interconnected and have the potential to have significant socioeconomic and environments impacts.
According to data available from the National Climatic Data Center (2002), nearly 10% of the total land area of the United States—approximately 89 000 000 ha—was experiencing either severe or extreme drought at any given time during the last century. Furthermore, there have been several notable exceptions when a larger segment of the United States was affected by drought. For example, the drought of the 1930s extended from California through the Intermountain West and the Great Plains into the Great Lakes region of the United States (Skaggs 1975). A second example is the drought of 2002, which, at its peak, encompassed approximately 39% of the land area of the United States (Douglas et al. 2003).
According to Obashi (1994), who cites statistics from the World Meteorological Organization, for the period between 1967 and 1991, drought impacted 1.4 billion people worldwide, the same number of individuals that were negatively affected by floods, hurricanes, and all other weather-related natural disasters combined. During that same period, 1.3 million of the 3.5 million deaths attributable to natural disasters globally were due to the effects of drought.
Drought also can have substantial economic costs; indeed, it has been suggested that drought is the costliest of all weather-related natural disasters (Wilhite 2002). As pointed out by Diaz (1983), Riebsame et al. (1991), and others, drought can affect the water supply, water quality, agricultural and timber productivity, power generation, and recreational activities. Drought also impacts some unexpected industries. For example, the drought in the United States during 1988 resulted in losses of revenue in excess of $1 billion to the barge industry (Changnon 1989). Overall, the drought of 1988 cost 5000 lives (Trenberth and Guillemot 1996) and more than $30 billion in related damages (Svoboda et al. 2002). The total economic cost of all major drought events in the United States since 1980 exceeds $100 billion (Lawrimore et al. 2002).
Additionally, drought has a significant impact on most ecosystems. Drought can be devastating to wetlands and riparian habitats, rangelands, and forested regions (Riebsame et al. 1991). Perhaps the starkest example of the relationship between drought and ecology can be seen in wildfire. For example, the early start and high severity of the 2002 fire season in the western United States can be attributed, at least in part, to drought (Douglas et al. 2003).
Given the effects of drought, it is clear that accurate predictions of the severity, extent, and duration of drought would be beneficial so that the adverse consequences of drought can be minimized. Since the quality of a model’s predictions are dependent on the strength of the underlying knowledge of those physical processes represented by the model, a solid understanding of the linkages between the land surface and the atmosphere is a key prerequisite for model development. A better understanding of the linkages between the land surface and the atmosphere is the first step toward improving the capabilities of land surface models to describe drought.
This study focuses on improving the understanding of one important linkage between the land surface and the atmosphere, λE. By investigating how the environmental factors driving λE vary with changing environmental conditions, the key controls on λE were isolated. In turn, a better understanding of the controls on λE during drought suggested relationships and methods for improving land surface models.
The study site, data collection methods, and postprocessing procedures, and principal component regression analysis are described in section 2. The results of the analyses are presented and discussed in the section 3. Finally, conclusions and a brief discussion of ongoing research are presented in section 4.
2. Methodology
a. Site description
The data used in this study were collected as a part of the International H2O Project (IHOP_2002), a multiagency field campaign conducted in the southern Great Plains of the United States during May and June 2002 (Weckwerth et al. 2004; LeMone et al. 2007). The data were collected between 20 May and 16 June 2002 at IHOP_2002 site 10 located in the panhandle of Oklahoma (36.88°N, 100.61°W; Fig. 1) northwest of the city of Beaver. Although the site, which was dominated by a single species of C4 grass, Andropogon gerardii, was not grazed during the field program, it was heavily grazed prior to the observation period. The vegetation was distributed in an intricate mosaic or patchwork wherein approximately 30% of the surface was covered with clumps of vegetation and the remaining surface was bare soil. Because of the drought conditions, the vegetation remained dormant until 30 May; this is atypically late since the growing season usually begins nearly a month earlier (Gould and Shaw 1983).
The site experienced a protracted period of severe to extreme drought prior to and through the duration of the field campaign. According to the index developed by the U.S. Drought Monitor and used to characterize drought in this study, severe drought is characterized by water shortages, moderate crop and pasture losses, very high fire risk, and precipitation levels 50%–60% below normal for the preceding 3–4-month period. Similarly, extreme drought is characterized by widespread water shortages, major crop and pasture losses, extreme fire risk, and precipitation levels 60%–70% below normal for the preceding 4–5-month period (Svoboda et al. 2002). (Further information regarding the Drought Monitor may be found online at http://www.drought.unl.edu.)
According to the Oklahoma Water Resources Board (OWRB), the panhandle region of Oklahoma, which includes IHOP_2002 site 10, received only 30%—approximately 216.0 mm—of its long-term average precipitation during the 7 months prior to the beginning of the IHOP_2002 field campaign (OWRB 2002a). Additionally, for the period leading up to and including the IHOP_2002 field campaign, 1 March to 1 July 2002, the panhandle of Oklahoma received 112.5 mm of total rainfall, which is 45% of the long-term average (OWRB 2002b). For comparison, this same region received more than twice the 2002 total for the same period during 2003 when the total rainfall was 251.0 mm (OWRB 2003). According to the Oklahoma Climatological Survey (OCS), the air temperature was slightly above the long-term average for both May and June 2002. The mean monthly air temperature for Beaver County, Oklahoma, during May 2002 was 20.4°C, approximately 2.1°C above average (OCS 2002a); during June 2002 the mean monthly air temperature was 25.0°C, approximately 1.0°C above average (OCS 2002b).
The drought conditions experienced at the study site are common in the southern Great Plains of the United States. Historic records, as well as tree-ring analyses, indicate that this region has experienced extended periods of drought at least three times since the mid-1800s (Woodhouse et al. 2002) and numerous short-duration droughts, such as the 1988 drought (Riebsame et al. 1991). The 2002 drought, which extended across 39% of the land area of the United States at its peak, was among the 10 driest on record. According to the OCS (2002c), it resulted in water shortages, multiple intense wildfire outbreaks, and economic losses exceeding $250 million for the state of Oklahoma alone.
b. Micrometeorological measurements
The micrometeorological data were collected using an eddy covariance micrometeorological station positioned 150 m from both the northern and eastern edge of the research site. The station was equipped with an array of instruments, and the data were stored as 30-min block averages in a datalogger (model CR23X, Campbell Scientific, Logan, Utah). The system was powered via a 12-V, 100 A-h battery trickle-charged using a solar panel.
The micrometeorological measurements included wind speed and virtual air temperature (Schotanus et al. 1983) using a sonic anemometer (model CSAT3, Campbell Scientific) mounted facing due east, the direction of the prevailing wind, at a height of 3 m above the ground. Water vapor density was measured using a krypton hygrometer (model KH2O, Campbell Scientific) mounted facing east at a height of 3 m with a horizontal displacement of 15 cm from the sonic anemometer. Both instruments operated at a sampling frequency of 10 Hz.
A standard suite of transformations and corrections was applied during postprocessing in order to determine the sensible heat flux (H) and λE. The first of these transformations was a coordinate rotation of the wind components such that both the mean crosswind (
Additional measurements included net radiation (Rnet; model Q*7, Radiation Energy Balance Systems, Seattle, Washington) and incident solar radiation (model Eppley pyranometer, Eppley Laboratory, Newport, Rhode Island). Both of these instruments were mounted at a height of 3 m facing due south. Radiative surface temperature was measured via an infrared thermometer (model IRT-4000, Everest Interscience, Tuscon, Arizona) mounted facing due west at a height of 3 m and oriented at a 45° angle such that the sensor measured the temperature over a 1-m2 sampling area representative of the site as a whole. Finally, precipitation was measured using a tipping-bucket rain gauge (model TE525WS, Texas Electronics, Dallas, Texas).
Additional measurements used in this research were collected at IHOP_2002 site 3, which was located approximately 1.25 km southeast of site 10. These measurements, which were overseen by the Atmospheric Technology Division of the National Center for Atmospheric Research, included atmospheric pressure, mixing ratio, and photosynthetically active radiation (PAR) measured using a digital barometer (model PTB 220, Viasala, Helsinki, Finland), an integrated humidity and temperature sensor (model Hummiter 50Y, Viasala), and a PAR sensor (model LI-190SA Quantum Sensor, LI-COR Biosciences, Lincoln, Nebraska), respectively.
c. Soil measurements
Soil properties, including temperature (Tsoil), volumetric moisture content (θ), and heat flux (G), were measured with a number of instruments buried approximately 2 m due north of the micrometeorological tower at a range of depths from 2.5 to 10.0 cm (Fig. 2). This position was chosen so as to minimize interference with the measurement of both net radiation and infrared surface temperature. The depths were chosen in order to maximize the accuracy of G and to create a near-surface soil moisture profile.
Soil temperature was measured with a soil temperature probe (model STP-1, Radiation Energy Balance Systems, Seattle, Washington) buried at a depth of 2.5 cm. Measurements of G were taken with a pair of soil heat flux plates (model HFT-3, Radiation Energy Budget Systems) buried at a depth of 5 cm and separated by a horizontal distance of 30 cm. The measured G at the heat flux plates was corrected to the surface by accounting for heat storage in the overlying soil layer as described by Oke (1987).
To determine θ, time domain reflectometry–based (TDR) probes (model CS615, Campbell Scientific) were used. Three probes were inserted horizontally into the soil at depths of 5.0, 7.5, and 10.0 cm at the same location as the other subsurface sensors. The soil moisture measurements at 5.0 cm were used in this analysis since it was nearest the surface-atmosphere interface; thus, it should have the strongest relationship with both the total λE and λE from the soil.
d. Surface measurements
In addition to the continuously monitored data described above, both leaf area index (LAI) and multispectral reflectance were measured on four days (19 May, 29 May, 7 June, and 16 June 2002) during the field campaign. LAI was measured using a plant canopy analyzer (model LAI-2000, LI-COR Biosciences) while the reflectance data were collected using a multispectral reflectometer (model MSR5, Cropscan, Rochester, Minnesota). Each of these measurements were made at 10 positions located at 5-m intervals along a pair of 20-m transects. One of the transects ran north–south beginning 10 m south of the micrometeorological station while the other transect ran east–west beginning 10 m west of the micrometeorological station (Fig. 3).
e. Gap-filling method
Because of such factors as instrument failure, inclement weather, and theoretical or practical considerations, micrometeorological datasets are seldom continuous (Falge et al. 2001). In the case of this study, gaps constituted approximately 1% and 4% of the H and λE datasets, respectively. While these gaps made up only a small portion of the total data collected and the gaps would not impact this analysis, a gap-filling procedure was utilized to generate a continuous dataset. Gap filling the data allowed the same dataset to be used in this study as will be used in follow-up studies with land surface models.
f. Analysis techniques
Principal component regression analysis (PCR) was used to determine the influence of the factors controlling λE. This method was selected because it minimized the effects of multicollinearity while retaining the maximum amount of information regarding the relative importance of any given predictor variable (Wilks 1995). Further, PCR allowed the relative influence of the predictor variables to be determined from the loading coefficients associated with the principal components.
PCR takes advantage of the mathematical transformation commonly referred to as principal component analysis (PCA) or empirical orthogonal function analysis (EOF). PCA rotates some number, n, of interrelated variables through n-dimensional space to generate a set of n new independent variables (Fig. 4), that is, principal components, as a weighted combination of the original variables (Jolliffe 2002). While the family of principal components retains all of the information contained within the original variables, principal components have an ordered hierarchy in which greatest amount of information from the original set of n variables is captured by the first PC and then the amount of information contained within each PC decreases with each subsequent PC until the least amount of information is contained in the last or nth PC (Jackson 1991). The hierarchal characteristic of PCs implies that the number of PCs used in further analysis can be reduced with minimal loss of information by eliminating the lower order PCs (Montgomery and Peck 1992).
Five response variables were regressed against the first principal component associated with five environmental factors. The response variables included the total λE, one of its two components, soil evaporation (λEsoil) or transpiration (λEveg), and the surface resistances associated with bare soil (rsoil) or a vegetated surface (rveg). The five predictor variables were vapor pressure deficit (D), which is defined as the difference between the saturation water vapor pressure and ambient water vapor pressure (Oke 1987), horizontal wind speed (U), net radiation (Rnet), soil moisture content (θ), and greenness fraction (Fg). The saturation water vapor pressure used in determining D was calculated as a function of ambient air temperature using the Clausius–Clapeyron equation (Bolton 1980); λE was partitioned into λEsoil and λEveg according to the two-source model proposed by Shuttleworth and Wallace (1985). The environmental factors were selected using the general guidelines provided by Draper and Smith (1981) and Montgomery and Peck (1992). The resulting subset of environmental factors demonstrated a strong practical and theoretical relationship with λE and eliminated the redundancy present in the complete dataset.
3. Results and discussion
a. Temporal variability in the environmental conditions
Figure 5 shows that the components of the surface energy budget varied significantly during the course of IHOP_2002 field campaign. When only rain-free days were considered, as was the case in this analysis, Rnet (Fig. 5a) demonstrated a consistent temporal pattern with an overnight minimum averaging −51 W m−2 and a midday average peak approaching 660 W m−2. The relatively small standard deviations of both the midday peak value (19.1 W m−2) and the overnight minimum (7.2 W m−2) underscore the consistency in the observed Rnet. In contrast, H (Fig. 5b) varied not only diurnally, but also over the observational period. Initially, H exceeded 400 W m−2 at midday; but, after rain events (e.g., on Day of Year 144 and 148), the midday H decreased dramatically to as little as 260 W m−2. The variability in the midday H was also demonstrated by the relatively large standard deviation of 58 W m−2, which was more than 3 times the standard deviation for the midday Rnet. On a diurnal basis, λE had an average range of nearly 160 W m−2 but the diurnal range exceeded 250 W m−2 on days following rain events (Fig. 5c). As was the case with H, the variation in λE is clearly demonstrated by looking at the midday peak values; the midday peak had an averaged value of 135 W m−2 and a standard deviation of 48.9 W m−2. Over the course of the IHOP_2002 field campaign, three distinct periods related to the behavior of λE could be seen: the period prior to day of year (DOY) 148 when the mean daily λE was 40 W m−2, the period from DOY 148 to 155 when a peak flux of 242 W m−2 was measured on DOY 148 followed by a period of gradual decline, and the period from DOY 156 to 166 when a peak λE of 363 W m−2 was measured on DOY 156 followed by a period of gradual decline in the peak daily λE. Finally, G (Fig. 5d) showed a consistent pattern with an average peak flux of approximately 80 W m−2 and a standard deviation of 12.1 W m−2.
There were no clear long-term patterns over the observational period for either D (Fig. 6a) or U (Fig. 6b). On both a diurnal basis and across the observational period, D varied over a range of approximately 54 mb with a mean near 18 mb and a standard deviation of 13 mb. Similarly, U ranged from 0 to nearly 13 m s−1 with a mean value over the entire observational period of 5.1 m s−1 and a standard deviation of 2.5 m s−1. A clear pattern was apparent for θ (Fig. 6c), however, and this pattern was tied to rain events. As can be seen by comparing Fig. 6c with Fig. 7d, although θ was initially less than 10%, it increased after rain events to more than 39% and 32% on DOY 148 and DOY 156, respectively. (Figure 7 provides a time series of several key meteorological variables including air temperature, atmospheric pressure, relative humidity, and rainfall.) Nearly 14 mm of rain fell on DOY 146 and 147 and 13 mm of rain fell overnight on DOY 155 into the morning of DOY 156. Each of the spikes in θ was followed by a period of slow dry down. Finally, Fg (Fig. 6d) increased in an exponential fashion, as would be expected during the early phases of plant phenology (Barbour et al. 1999), from zero at the beginning of IHOP_2002 to nearly 0.30 at the end of the observational period.
b. Principal component regression analyses using the total latent heat flux
To understand the relative influence of the environmental factors, the data were first sorted chronologically and PCR was conducted on each unique time period using the total λE as the response variable. The time periods selected for this analysis included the entire observational period, the portion of the field campaign during May, the portion of the field campaign during June, and each of the four weeks during IHOP_2002 (Fig. 8). With the possible exception of the increasing influence of vegetation as measured via Fg seen in the latter weeks of the field campaign when the grass had transitioned from a dormant to active state, there were no clear temporal patterns observed in the relative influence of the environmental factors driving λE. However, this analysis did show that θ was consistently a key factor in driving λE; overall, during the entire observational period the relative influence of θ was 39%, which is more than 10% greater than the next most significant factor, wind speed, which had a relative influence of approximately 28%. Especially during the earliest weeks of IHOP_2002 when the site was driest, the magnitude of λE was limited by water availability.
When PCR was conducted on the data sorted according to θ, several patterns became evident (Fig. 9). First, this analysis clarified and reconfirmed the results suggested by the PCR analysis when the data were sorted by time, that the influence of θ was greatest when the soils were the driest. Under dry soil conditions, the relative influence of θ was nearly 40% while under wet soil conditions it was merely 15%. These results suggest that decreased water availability was an important control on λE during dry conditions.
Also, θ appeared to impact the role several other environmental factors played in driving λE. For example, U had the greatest influence on λE under dry conditions but it played only a minor role when the soil was wet. This pattern was due to U maintaining a high atmospheric demand for water vapor by continually introducing dry air to the surface. Maintaining a high atmospheric demand, in turn, facilitated the maximum rate of evaporation possible given the limited water supply.
As the soil became moist, the influence of Rnet increased sharply. Under dry soil conditions, Rnet has a relative influence of approximately 5% but it increased to nearly 34% under moist soil conditions and remained high under wet soil conditions. The importance of Rnet under moist and wet conditions suggests a threshold where the amount of energy available to evaporate water is no longer sufficient to evaporate all of the moisture available. Under wet conditions, λE at the site became energy, not water, limited. In general, wet soils may have an albedo only half that of dry soils (Jury and Horton 2004). While this would tend to increase Rnet under wet conditions, the overcast conditions during rain events resulted in an overall reduction in Rnet as θ started to increase. Immediately following rain events, clear-sky conditions resulted in an increase in Rnet as θ started to decrease. Therefore, the relationship between Rnet and θ was relatively weak due to the fast response of the soil to rain events. The weakness of the relationship is evidenced by the correlation coefficient (r) of 0.49.
Finally, while the vegetation was dormant during the period when the soil was driest, it appeared that importance of Fg increased in direct relation to θ. The apparent relationship between Fg and θ suggests that moister soils facilitated the uptake of water by the vegetation through an increase in the hydraulic conductivity, which provided a second pathway for the transfer of water to the atmosphere.
This analysis clarified the role of vegetation as represented by Fg (Fig. 10). When the vegetation was dormant, Fg falls out of the analysis since the relative influence of Fg goes to zero. Once vegetation was present, as D increased, the role of Fg decreased. This inverse relationship between D and the relative influence of Fg is to be expected since plants respond to a high D by closing their stomata to minimize water loss (Hopkins 1999). Since transpiration should be minimal when there is a high D, the contribution of transpiration to the total λE is reduced and variations in Fg have less impact on the total moisture flux.
Also, while there was no correlation between D and U (r = 0.12), there appeared to be a threshold relationship between D and the relative influence of U with U taking on a much more influential role when there was a high D. The relative influence of U increased from 18% for the low and intermediate subgroups to 36% when D was at its greatest. This relationship between D and U suggests that the turbulent transport of water vapor is critical when there is a high D, a period that corresponds to low or intermediate levels of θ.
While the results show, quite reasonably, that the relative influence for Fg increased from zero when there is no green vegetation to 28% once the vegetation was well established late in the study period, it also showed the relationship between the role of D and Fg (Fig. 11). The relative influence of D increased from less than 3% when Fg was at a bare or intermediate level to nearly 12% when the vegetation was established. Given the role D plays in controlling water loss via transpiration, the increasing importance of D with increasing amounts of vegetation cover is not altogether unexpected. However, the relationship between the influence of D and the magnitude of Fg also suggests that the main influence of D on total λE is through the impact of D on water transfer via transpiration.
c. Analysis of the soil evaporation
After partitioning λE into λEsoil and λEveg using the two-source model developed by Shuttleworth and Wallace (1985), it was found that λEsoil was the primary source for the moisture flux over the entire observation period. (Days with rain events were omitted so that the evaporation of intercepted water could be excluded from the analyses.) Between 53% and 100% of the total λE was due to λEsoil (Fig. 12), and on average, λEsoil contributed 86% of the total moisture flux; λEsoil accounted for all of λE during the period from DOY 140 to 148 because there was no significant green vegetation at the study site.
Because of the large soil evaporation component (λEsoil), PCR was conducted using only the component of λE from soil evaporation with the data sorted according to θ as described above. As shown in Fig. 13, the results of this analysis reinforce those of the earlier analysis and suggest that θ was the key control particularly under dry or moist soil conditions. Under dry soil conditions the relative influence of θ was nearly 50%, which is more than twice the relative influence of the next most influential environmental factor, Rnet. Under moist soil conditions, θ remained the most influential environmental factor with a relative influence exceeding 40%. Even under wet soil conditions when Rnet was slightly more influential in driving soil evaporation, θ maintained a relative influence of approximately 22%.
Since it might seem counterintuitive that Fg has an influence on λEsoil, it is important to recall that as Fg varied, the fraction of the surface area that is bare also varied in a complementary fashion (the fractions of bare soil and vegetated ground must sum to one). Since evaporation is proportional to the surface area over which moisture exchange can take place, for a given level of θ, Fg influenced the amount of λEsoil through its impact on the amount of the surface area that is bare soil.
The relationship between θ and λE (or λEsoil) may be explained using the physical processes governing soil evaporation. The initial sharp increase in λE may be due to the transition in soil evaporation resulting from diffusion from deeper soil layers (Hornberger et al. 1998) to soil evaporation limited by moisture availability and soil properties (Wallace et al. 1999; Suleiman and Ritchie 2003). Once θ exceeded approximately 10%, λE (or λEsoil) increased in an exponential fashion. During the latter phase of the soil evaporation process, soil evaporation is proportional to the hydraulic diffusivity of the soil, which, in turn, increases exponentially with increasing θ (Hillel 1998).
4. Conclusions
Based on the results of this study, it is clear that θ was an important control on both the total λE and the primary pathway for moisture exchange with the atmosphere, soil evaporation. Particularly, under dry and moist soil conditions, when evaporation from the site was essentially water limited, θ had a strong control on the moisture flux. However, other environmental factors, such as Rnet and Fg, also had a major influence on λE when soil conditions were wet, that is, when λE from the site was energy limited.
Given the high relative influence of θ in controlling λE, which exceeded 40% in some cases, a thorough understanding of soil hydrology is a prerequisite for accurately describing or modeling the exchange of moisture with the atmosphere. To accurately describe λE during drought conditions, it is first necessary to accurately describe the soil hydrology and other physical processes controlling θ. This conclusion is highlighted by the strong empirical relationships developed between θ and the total λE, λEsoil, and rsoil.
This study has important implications for the modeling community. Currently, many land surface models, such as the Noah land surface model (Ek et al. 2003), the Simple Biosphere II (Sellers et al. 1996), and the Community Climate Model (Oleson et al. 2004), determine rsoil either directly or indirectly as an exponential decay function of θ. These results suggest that a sigmoidal relationship may yield a more accurate description of rsoil and ultimately the total λE in sparse-canopy environments. Such a relationship could be especially useful when describing λE at the soil moisture extremes as was the case during the IHOP_2002 field campaign.
Given the social, economic, and environmental effects of drought, it would be beneficial to be able to accurately model the extent, duration, and severity of drought so that its adverse effects can be minimized. This study reemphasized the importance of θ and the need to describe accurately soil hydrology to predict λE correctly. This study also provided a practical means of improving land surface models by more accurately describing the relationship between θ and rsoil under extremes in soil moisture conditions.
These conclusions are particularly salient to the developers of land surface models. While experimental and field studies have resulted in a number of methods for estimating soil evaporation, these studies have often been conducted over bare soil (e.g., Mahfouf and Noilhan 1991) with little consideration of the role of vegetation and other environmental factors. As a result, many land surface models estimate soil evaporation without fully accounting for all of the factors that drive the moisture flux.
Although this study was conducted at a specific location in the panhandle of Oklahoma, the grazed land cover and environmental conditions at the site are not atypical of the southern Great Plains and American West as a whole. Therefore, the general findings of this research are likely applicable to regions with a similar land use, soil, vegetation, and vulnerability to drought conditions. Further research should confirm if other sparsely vegetated ecoregions, such as boreal forest, behave similarly to this study site. To test the broader applicability of this study, follow-up research is ongoing. The soil moisture–soil resistance relationship is being tested over the entirety of the IHOP_2002 domain by implementing the relationship in the Noah land surface model. After the sigmoidal function for rsoil has been successfully validated over the IHOP_2002 domain, it will be tested over other ecological regions.
Acknowledgments
The authors thank M. LeMone, R. Grossman, and M. O’Connell for their assistance during the IHOP_2002 field campaign. The authors thank F. Chen, T. Horst, S. Oncley, G. McLean, J. Mecikalski, R. Cuenca, and D. Gochis for their discussions and insight during the course of this research. The authors thank S. Grant for her assistance with preparing this manuscript. The authors thank the three anonymous reviewers and the editor of the Journal of Hydrometeorology for their thoughtful comments and suggestions that greatly improved the quality and clarity this paper. The authors would like to acknowledge the financial support of the NCAR Water Cycle Initiative, and the National Science Foundation (Awards ATM-0236885 and ATM-0296159).
REFERENCES
American Meteorological Society, 1997: Meteorological drought— Policy statement. Bull. Amer. Meteor. Soc., 78 , 847–849.
Barbour, M. G., Burk J. H. , Pitts W. D. , Gilliam F. S. , and Schwartz M. W. , 1999: Terrestrial Plant Ecology. 3d ed. Addison Wesley, 649 pp.
Blanken, P. D., and Coauthors, 1997: Energy balance and canopy conductance of a boreal aspen forest: Partitioning overstory and understory components. J. Geophys. Res., 102 , 28915–28927.
Bolton, D., 1980: The computation of equivalent potential temperature. Mon. Wea. Rev., 108 , 1046–1053.
Campbell Scientific, 1998: Eddy Covariance System Operator’s Manual CA27 and KH2O. 29 pp.
Carlson, T. N., and Ripley D. A. J. , 1997: On the relationship between NDVI, fractional vegetation cover, and leaf area index. Remote Sens. Environ., 62 , 241–252.
Changnon, S. A., 1989: The 1988 drought, barges, and diversion. Bull. Amer. Meteor. Soc., 70 , 1092–1104.
Diaz, H. F., 1983: Drought in the United States. J. Climate Appl. Meteor., 22 , 3–16.
Dingman, S. L., 2002: Physical Hydrology. 2d ed. Prentice-Hall, 646 pp.
Douglas, A., Gleason K. , Phillips D. , and Waple A. M. , 2003: State of climate in 2002. Bull. Amer. Meteor. Soc., 84 , S1–S68.
Draper, N. R., and Smith H. , 1981: Applied Regression Analysis. John Wiley and Sons, 709 pp.
Ek, M. B., Mitchell K. E. , Lin Y. , Rogers E. , Grunmann P. , Koren V. , Gayno G. , and Tarpley J. D. , 2003: Implementation of Noah land surface model advances in the National Centers for Environmental Prediction operational mesoscale Eta Model. J. Geophys. Res., 108 .8851, doi:10.1029/2002JD003296.
Falge, E., and Coauthors, 2001: Gap filling strategies for long term energy flux data sets. Agric. For. Meteor., 107 , 71–77.
Gould, F. W., and Shaw R. B. , 1983: Grass Systematics. 2d ed. Texas A&M University Press, 397 pp.
Gutman, G., and Ignatov A. , 1997: Satellite-derived green vegetation fraction for the use in numerical weather prediction models. Adv. Space Res., 19 , 477–480.
Heim, R. R., 2002: A review of twentieth-century drought indices used in the United States. Bull. Amer. Meteor. Soc., 83 , 1149–1165.
Hillel, D., 1998: Environmental Soil Physics. Academic Press, 771 pp.
Hopkins, W. G., 1999: Introduction to Plant Physiology. John Wiley and Sons, 512 pp.
Hornberger, G. M., Raffensperger J. P. , Wiberg P. L. , and Eshleman K. N. , 1998: Elements of Physical Hydrology. John Hopkins University Press, 312 pp.
Horst, T. W., 2006: Attenuation of scalar fluxes measured with horizontally-displaced sensors. Extended Abstracts, 17th Symp. on Boundary Layers and Turbulence, San Diego, CA, Amer. Meteor. Soc., CD-ROM, 7.5.
Jackson, J. E., 1991: A User’s Guide to Principal Components. John Wiley and Sons, 569 pp.
Jolliffe, I. T., 2002: Principal Component Analysis. Springer-Verlag, 516 pp.
Jury, W. A., and Horton R. , 2004: Soil Physics. John Wiley and Sons, 370 pp.
Kaimal, J. C., and Finnigan J. J. , 1994: Atmospheric Boundary Layer Flows. Oxford University Press, 289 pp.
Korgan, F. N., 1997: Global drought watch from space. Bull. Amer. Meteor. Soc., 78 , 621–636.
Lawrimore, J., Heim R. R. , Svoboda M. , Swail V. , and Englehart P. J. , 2002: Beginning a new era of drought monitoring across North America. Bull. Amer. Meteor. Soc., 83 , 1191–1192.
LeMone, M. A., and Coauthors, 2007: NCAR/CU surface, soil, and vegetation observations during the international H2O project 2002 field campaign. Bull. Amer. Meteor. Soc., 88 , 65–81.
Mahfouf, J. F., and Noilhan J. , 1991: Comparative study of various formulations of evaporations from bare soil using in situ data. J. Appl. Meteor., 30 , 1354–1365.
Montgomery, D. C., and Peck E. A. , 1992: Introduction to Linear Regression Analysis. John Wiley and Sons, 461 pp.
National Climatic Data Center, 2002: U.S. national percent area severely to extremely dry and severely to extremely wet. [Available online at http://www.ncdc.noaa.gov/oa/climate/research/2002/may/uspctarea-wetdry.txt.].
Obashi, G., 1994: WMO’s role in the international decade for natural disaster reduction. Bull. Amer. Meteor. Soc., 75 , 1655–1661.
OCS, 2002a: Oklahoma monthly climate summary, May 2002. University of Oklahoma, 23 pp. [Available from Oklahoma Climatological Survey, University of Oklahoma, 100 E. Boyd St., Ste. 1210, Norman, OK 73019.].
OCS, 2002b: Oklahoma monthly climate summary, June 2002. University of Oklahoma, 23 pp. [Available from Oklahoma Climatological Survey, University of Oklahoma, 100 E. Boyd St., Ste. 1210, Norman, OK 73019.].
OCS, 2002c: Oklahoma event summary: The Oklahoma drought of 2001-2002. University of Oklahoma, 29 pp. [Available from Oklahoma Climatological Survey, University of Oklahoma, 100 E. Boyd St., Ste. 1210, Norman, OK 73019.].
Oke, T. R., 1987: Boundary Layer Climates. 2d ed. Routledge, 464 pp.
Oleson, K., and Coauthors, 2004: Technical description of the Community Land Model (CLM). NCAR Tech. Note NCAR/TN-461+STR, National Center for Atmospheric Research, 173 pp.
OWRB, 2002a: Oklahoma water resources bulletin and summary of current conditions, May 8, 2002. Oklahoma Water Resource Board, 8 pp. [Available from Oklahoma Water Resources Board, 3800 N. Classen Blvd., Oklahoma City, OK 73118.].
OWRB, 2002b: Oklahoma water resources bulletin and summary of current conditions, July 3, 2002. Oklahoma Water Resource Board, 8 pp. [Available from Oklahoma Water Resources Board, 3800 N. Classen Blvd., Oklahoma City, OK 73118.].
OWRB, 2003: Oklahoma water resources bulletin and summary of current conditions, July 3, 2003. Oklahoma Water Resource Board, 8 pp. [Available from Oklahoma Water Resources Board, 3800 N. Classen Blvd., Oklahoma City, OK 73118.].
Pond, S., Phelps G. T. , Paquin J. E. , McBean G. , and Stewart R. W. , 1971: Measurement of turbulent fluxes of momentum, moisture and sensible heat over ocean. J. Atmos. Sci., 28 , 901–917.
Redmond, K. T., 2002: The depiction of drought: A commentary. Bull. Amer. Meteor. Soc., 83 , 1143–1147.
Riebsame, W. E., Changnon S. , and Karl T. R. , 1991: Drought and Natural Resource Management in the United States: Impacts and Implications of the 1987-89 Drought. Westview Press, 174 pp.
Schotanus, P., Nieuwstadt F. T. M. , and DeBruin H. A. R. , 1983: Temperature measurements with a sonic anemometer and its application to heat and moisture fluxes. Bound.-Layer Meteor., 26 , 81–93.
Sellers, P. J., and Coauthors, 1996: A revised surface parameterization (SiB2) for atmospheric GCMs. Part I: Model formulation. J. Climate, 9 , 676–705.
Shuttleworth, W. J., and Wallace J. S. , 1985: Evaporation from sparse crops—An energy combination theory. Quart. J. Roy. Meteor. Soc., 111 , 839–855.
Skaggs, R. H., 1975: Drought in the United States, 1931-40. Ann. Assoc. Amer. Geogr., 65 , 391–402.
Suleiman, A. A., and Ritchie J. T. , 2003: Modeling soil water redistribution during second-stage evaporation. Soil Sci. Soc. Amer. J., 67 , 377–386.
Svoboda, M., and Coauthors, 2002: The drought monitor. Bull. Amer. Meteor. Soc., 83 , 1181–1190.
Trenberth, K. E., and Guillemot C. J. , 1996: Physical processes involved in the 1988 drought and 1993 floods in North America. J. Climate, 9 , 1288–1298.
Wallace, J. S., Jackson N. A. , and Ong C. K. , 1999: Modelling soil evaporation in an agroforestry system in Kenya. Agric. For. Meteor., 94 , 189–202.
Webb, E. K., Pearman G. I. , and Leuning R. , 1980: Correction of flux measurements for density effects due to heat and water vapor transfer. Quart. J. Roy. Meteor. Soc., 106 , 85–100.
Weckwerth, T., and Coauthors, 2004: An overview of the International H2O Project (IHOP 2002) and some preliminary highlights. Bull. Amer. Meteor. Soc., 85 , 253–277.
Wilhite, D., 2000: Drought as a natural hazard: Concepts and definitions. Drought: A Global Assessment, D. Wilhite, Ed., Routledge, 3–18 pp.
Wilks, D. S., 1995: Statistical Methods in the Atmospheric Sciences. Academic Press, 467 pp.
Woodhouse, C. A., Lukas J. J. , and Brown P. M. , 2002: Drought in the Western Great Plains, 1845–56. Bull. Amer. Meteor. Soc., 83 , 1485–1493.
The location of the research site, IHOP_2002 site 10, along with the land use/land cover of the surrounding region, which includes the entire domain of the International H2O Project 2002.
Citation: Journal of Hydrometeorology 8, 2; 10.1175/JHM569.1
The type and position of each of the soil sensors are shown. STP refers to the soil temperature probe; HFT3 refers to the soil heat flux plates; and CS615 refers to the soil moisture probes.
Citation: Journal of Hydrometeorology 8, 2; 10.1175/JHM569.1
The position relative to the micrometeorological station (S) of each of the measurement points that make up the pair of transects.
Citation: Journal of Hydrometeorology 8, 2; 10.1175/JHM569.1
A simple two-dimensional example of principal component analysis is shown. The original variables, X and Y, are rotated in two dimensions to yield the principal components X′ and Y′. The principal components contain all of the information contained within the original variables; however, most of that information is contained within the first principal component, X′.
Citation: Journal of Hydrometeorology 8, 2; 10.1175/JHM569.1
Temporal variability is shown for each of the components of the surface energy budget including (a) net radiation, (b) the sensible heat flux, (c) the latent heat flux, and (d) the soil heat flux.
Citation: Journal of Hydrometeorology 8, 2; 10.1175/JHM569.1
The time series of the measurements of (a) water vapor deficit, (b) wind speed, (c) soil moisture content, and (d) greenness fraction.
Citation: Journal of Hydrometeorology 8, 2; 10.1175/JHM569.1
Times series for several common meteorological measurements including (a) air temperature, (b) atmospheric pressure, (c) relative humidity, and (d) daily total precipitation.
Citation: Journal of Hydrometeorology 8, 2; 10.1175/JHM569.1
The PCR results when the total latent heat flux was used as the response variable and the data were divided temporally using periods ranging from a single week to the entire study period. The investigation periods were as follows: Study: 20 May–16 June; May: 20–31 May; June: 1–16 June; week 1: 20–26 May; week 2: 27 May–2 June; week 3: 3–9 June; week 4: 10–16 June.
Citation: Journal of Hydrometeorology 8, 2; 10.1175/JHM569.1
The PCR results when the total latent heat flux was used as the response variable and the data were sorted by soil moisture content. The criteria for partitioning the data are given in Eqs. (8a)–(8c).
Citation: Journal of Hydrometeorology 8, 2; 10.1175/JHM569.1
The PCR results when the total latent heat flux was used as the response variable and the data were sorted by vapor pressure deficit. The criteria for partitioning the data are given in Eqs. (9a)–(9c).
Citation: Journal of Hydrometeorology 8, 2; 10.1175/JHM569.1
The PCR results when the total latent heat flux was used as the response variable and the data were sorted by greenness fraction. The criteria for sorting the data are given in Eqs. (10a)–(10c).
Citation: Journal of Hydrometeorology 8, 2; 10.1175/JHM569.1
The daytime mean total latent heat flux partitioned into the flux from the soil, i.e., soil evaporation, and the flux from the vegetation, i.e., transpiration. The gaps are days during which rain events occurred.
Citation: Journal of Hydrometeorology 8, 2; 10.1175/JHM569.1
The PCR results when soil evaporation was used as the response variable and the data were sorted by soil moisture content. The criteria for sorting the data are given in Eqs. (8a)–(8c).
Citation: Journal of Hydrometeorology 8, 2; 10.1175/JHM569.1
The empirical relationships between soil moisture content and both the total latent heat flux and the latent heat flux from the soil, i.e., soil evaporation. The histogram shows the number of rain-free days for which the mean daytime soil moisture content was in a given of 5% bin.
Citation: Journal of Hydrometeorology 8, 2; 10.1175/JHM569.1
The empirical relationship between soil moisture content and soil resistance. The locations of the wilting point θw, field capacity θfc and saturation (θsat) soil moisture content are labeled.
Citation: Journal of Hydrometeorology 8, 2; 10.1175/JHM569.1
A sample lookup table used in the gap-filling process for the sensible heat flux. The values represent the 5-day average centered on local noon for DOY 152.
Regression coefficients for the best-fit sigmoidal relationships between soil moisture content, and both the total latent heat flux (λEtot) and soil evaporation (λEsoil) are given.