ET-E0 Relations under Drought
The actual flux of ET that supplies moisture from the land surface to the atmosphere and the idealized flux of E0 that demands it are linked across the land surface-atmosphere interface, particularly in drought. Here we discuss the use of these linkages to monitor and predict drought.
The relationships between long-term mean water balance components and their water and energy limits can be illustrated across the hydroclimatic spectrum using the Budyko (1974) framework shown in Figure 11.1. This framework is also informative for understanding their relationships under transient dry anomalies or droughts, which push regions toward the water limit (i.e., toward the right in Figure 11.1). How ET and E0 respond to drying anomalies or drought depends on the regional mean behavior or on the regional hydroclimatic state at drought onset. For the case of regions that are energy-limited either climatologically or as a transient anomaly, recall that E0 defines the upper energy limit on ET so variations in E0 are matched by variations in ET (until ET becomes constrained by water availability, at which point the dynamic changes). Until this point, ET and E0 vary together (see Figure 11.1b) with forcing from E0 to ET, as both increase under increasing energy availability or advection. This is a parallel relationship, increasing in the case of developing drought and decreasing in drought mediation. This dynamic dominates in flash (or rapid onset) drought under energy-limited initial conditions.
On the other hand, in water-limited conditions (or when energy limitations on ET give way to tighter water limitations) ET is constrained by the availability of water: further decreasing the availability of water results in decreasing ET and less of the energy available at the land surface being expended as latent heat, so leaving more energy available for H. This increased H flux
Interrelations of ET and E0 in drought expressed in the context of the Budyko (1974) concept. In (a), solid lines denote the water limit on ET (horizontal line) and energy limit on ET (sloped line), to which the behavior of ET is asymptotic in arid and humid hydroclimatic extremes, respectively. The dashed curve shows Budyko's (1974) idealized relation between dryness index (Ф) and the evaporative index (e). Vertical arrows represent the ET and Runoff portions of Prcp, respectively. Circles represent typical observed climatological annual behavior, from 229 Australian basins (from Donohue, R.J., et al., J. Hydrol, 390, 23-34, 2010). (Adapted from Hobbins, M. T., and J. Huntington, Handbook of Applied Hydrology, 42-1-42-18, McGraw-Hill Education, New York, 2016.) The interrelations of drying anomalies from either extreme of the hydroclimatic spectrum are shown in (b) for an energy-limited basin and (c) for a water- limited basin.
raises the temperature of the overpassing air, increasing its vapor pressure deficit, and thereby increases E0. Thus, as ET decreases, E0 increases in what is known as the complementary relationship (see Figure 11.1c); it was first proposed by Bouchet (1963) and observed across CONUS by Hobbins et al. (2004). This complementarity is also evident at larger space and time scales: lower ET leads to less cloudy conditions with attendant increases in surface energy and H, which again results in higher E0. In these conditions, forcing is from ET to E0, with the fluxes varying oppositely in both drying (drought intensification) and wetting (drought mediation). This dynamic dominates in sustained drought.
In summary, we observe a complementary relationship between ET and E0 developing in sustained droughts and parallel relationships developing in flash drought onset (Hobbins et al. 2016; McEvoy et al. 2016a). The key here is that E0 increases in both types of drought, whereas ET increases in flash drought onset and decreases thereafter, and is suppressed in sustained drought. This suggests the opportunity to treat E0 as a robust precursor of both types of drought (as in the EDDI tool described in Section 11.3.2). It should be noted that high-E0 events will always precede actual onset of stress, but not all high-E0 events develop into drought, nor should we expect each event to, since meteorological and radiative factors also control the persistence of a high-E0/low-Prcp regime. Nonetheless, it may be useful to conceptualize the persistence of high E0 as a stage-setter for moisture deficits or vegetative stress, and depressed ET as reflecting the onset of such stress. The interactions of ET and E0 with energy and water availability are summarized in Figure 11.2.
Figure 11.3 shows a hypothetical evolution of agricultural and/or meteorological drought. In this case, sufficient SM conditions prevail in a meteorological regime highlighted by below-normal Prcp and above-average atmospheric demand for ET (e.g., from high temperatures, solar loading, or wind speeds). In this case, one would expect a metric that estimates anomalies in atmospheric demand to start to show the potential for SM stress developing if large-scale meteorological conditions were to remain the same. As the meteorological
Interrelations of ET and E0 summarized with respect to drought. The sides of the triangles represent the constraints, or drivers, of each flux.
Schematic showing the temporal evolution of agricultural and/or meteorological drought and typical remote sensing observational timescales for drought early warning capabilities.
forcing continues, surface SM conditions begin to degrade, a condition that can be remotely sensed by active and passive microwave sensors. As surface SM conditions continue to degrade, root-zone SM conditions will also begin to show below-normal conditions and, at this point, the onset of vegetation stress due to inadequate moisture supplies will occur. The onset of actual vegetation stress can be determined through the use of methods that determine actual ET from satellite observations of TIR LST. Finally, as conditions continue, surface and root-zone SM conditions are well below normal, and damage to vegetation health commences. Anomalies in vegetation health, usually determined by below-normal "greenness," are usually observed through vegetation indices based on visible and near-infrared wavelengths (e.g., the Normalized Difference Vegetation Index [NDVI]). In summary, the introduction of E0- and ET-based tools for drought monitoring provide a necessary augmentation to current drought-monitoring systems, especially in terms of providing an understanding of early drought evolution that can aid in the further development of drought early warning systems.