DeepDyve requires Javascript to function. Please enable Javascript on your browser to continue.
Evaluation of the Operational Simplified Surface Energy Balance Model for Pastureland Evapotranspiration Mapping and Drought Monitoring in North Central Kentucky
Evaluation of the Operational Simplified Surface Energy Balance Model for Pastureland...
Gebremedhin, Maheteme;Ries, Ian;Senay, Gabriel B.;Matisoff, Martin;Amusan, Ibukun;Sandifer, Jeremy;Gyawali, Buddhi
2020-06-11 00:00:00
Hindawi Advances in Meteorology Volume 2020, Article ID 1386468, 11 pages https://doi.org/10.1155/2020/1386468 Research Article Evaluation of the Operational Simplified Surface Energy Balance Model for Pastureland Evapotranspiration Mapping and Drought Monitoring in North Central Kentucky 1 1 2 1 Maheteme Gebremedhin , Ian Ries, Gabriel B. Senay, Martin Matisoff, 3 1 1 Ibukun Amusan, Jeremy Sandifer, and Buddhi Gyawali College of Agriculture, Communities, and the Environment, Kentucky State University, 400 East Main Street, Frankfort, KY 40601, USA United States Geological Survey, Earth Resources Observation and Science Center, North Central Climate Adaptation Science Center, Boulder, CO 80303, USA College of Business and Computational Sciences, Kentucky State University, 400 East Main Street, Frankfort, KY 40601, USA Correspondence should be addressed to Maheteme Gebremedhin; maheteme.gebremedhin@kysu.edu Received 19 August 2019; Revised 22 April 2020; Accepted 25 May 2020; Published 11 June 2020 Academic Editor: Nir Y. Krakauer Copyright © 2020 Maheteme Gebremedhin et al. (is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. (e use of remotely sensed evapotranspiration (ET) for field applications in drought monitoring and assessment is gaining momentum, but meeting this need has been hampered by the absence of extensive ground-based measurement stations for ground validation across agricultural zones and natural landscapes. (is is particularly crucial for regions more prone to recurring droughts with limited ground monitoring stations. A three-year (2016–2018) flux ET dataset from a pastureland in north central Kentucky was used to validate the Operational Simplified Surface Energy Balance (SSEBop) model at monthly and annual scales. Flux and SSEBop ET track each other in a consistent manner in response to seasonal changes. (e mean bias error (MBE), root mean squared error (RMSE), mean absolute percentage error (MAPE), and coefficient of determination (R ) were 5.47, 21.49 mm −1 mon , 30.94%, and 0.87, respectively. (e model consistently underestimated ET values during winter months and overestimated them during summer months. SSEBop’s monthly ET anomaly maps show spatial ET distribution and its accurate representation. (is is particularly important in areas where detailed surface meteorological and hydrological data are limited. Overall, the model estimated monthly ET magnitude satisfactorily and captured it seasonally. (e SSEBop’s functionality for remote ET estimation and anomaly detection, if properly coupled with ground measurements, can significantly enhance SSEBop’s ability to monitor drought occurrence and prevalence quickly and accurately. balance (ET = P-R-S-I, where P = precipitation, R = runoff, 1. Introduction S = soil storage, and I = infiltration) and is the single most Evapotranspiration (ET), the turbulent transfer of water important predictor of seasonal crop water consumptive use from the ground (evaporation; E) and plant surfaces drought prevalence. ET estimation and drought monitoring (transpiration; T), is a key component of the hydrologic are determined by the availability and proliferation of re- cycle, which is responsible for returning 60–90% of pre- motely sensed data. Furthermore, the need for improved cipitation (P) to the atmosphere[1–3]. Previous studies (e.g., methods for monitoring and modeling the water cycle has [4–6]) showed that ET exhibits considerable variation in fueled interest in the rapid, widespread use of remote-based space and time and, as a result, obtaining reliable and ac- ET data. curate measurements continues to present challenges [7]. ET Traditional land-based ET measurements have largely is the second largest component in the terrestrial water relied on in situ ground observation methods, such as 2 Advances in Meteorology lysimtery, energy balance, Bowen ratio (FR), eddy covari- for drought monitoring and identifying drought years/ ance (EC), scintillometry, and soil water balance [8–12]. seasons using rainfall as a surrogate variable. (e surface-based EC method for small areas of coverage has proven invaluable in providing direct measurement of 2. Materials and Methods turbulent water vapor fluxes between the ground and the 2.1. Site Microclimate. (e study site is located at the Harold atmosphere. For more than two decades, a network of R. Benson Research and Demonstration Farm, 12 km systems [11] has provided the high-quality data needed to southwest of the main campus of Kentucky State University, validate remotely sensed ET estimates. (e EC method ° ° Frankfort, Kentucky (38.11 N, 84.88 W; 242 m above mean directly measures ET based on statistical covariance be- sea level). (e study period had marked differences in total tween vertical water-vapor flux and scalar concentration precipitation among the years, ranging from the lowest fluctuations [13]. Scaling up observational ET information −1 −1 (1012 mm·yr ) in 2016 to 1855 mm·yr in 2018. 2018 was a is a major challenge of this method because it involves a 150-year record, which gave 697 mm (60%) more rain than complex and multidimensional process that varies with the historical (1980–2010) average. Based on the 30-year landscape, vegetation type, soil moisture, and crop man- average of 1159 mm for the region, the observation years can agement. It is also affected by underlying environmental be classified as dry in 2016 (1012 mm), average in 2017 factors [14, 15]. In this regard, space-borne ET estimates (1219 mm), and wet in 2018 (1855 mm). Air temperature (T ) [16–18] are ideal to circumvent some of the stringent re- a from 2016 through 2018 at the site showed no major de- quirements (e.g., sensor placement and calibration) and parture relative to the thirty-year (1981–2010) average of limitations (smaller footprint) of land-based ET observa- 12.8 C. Averaged over the three years, monthly T varied tions [1, 7]. ° ° between −1 C in January and 24.2 C in July, with annual For the past 18 years, simplified surface energy balance averages of 13.5, 13.8, and 13.3 C in 2016, 2017, and 2018, (SSEBop) has generated and issued ET maps for the con- respectively. (e only exception was January 2018, which tinental United States on a monthly time interval at a 1 km was the coldest month (−5.0 C; Figure 1). (e soil at the site spatial resolution scale. It also provides an efficient and is in the well-drained McAfee silt loam soil series, commonly timely data delivery system that is critical for operational found in a range of 6 to 12 percent slopes, slightly acidic in decision-making by water resource planners and managers pH, and moderately drained with high permeability [23]. and hydraulic engineers. (e model’s use and application have successfully reported ET on different scales, i.e., wa- tershed, basin, United States, and global scale [19, 20]. In 2.2. Pasture Management. Kentucky’s pasture agro- addition, the model has been used to generate ET and ecosystem is extensive and largely rain-fed. Approximately anomaly maps over very large extents (the coterminous 25% (2.5 million acres) of the total 13.4 million land area is United States) using low spatial resolution (1 km) MODIS used for pasture agroecosystems. (e study area at the data. Harold R. Benson Research and Demonstration Farm One key attribute of satellite-based ETestimate models is (pasture) was seeded in 2005 and consisted of mixed grasses, model validation, which is defined as the process of de- predominantly fescue (Festuca arundinacea), as well as termining the degree to which a model is an accurate Johnson grass (Sorghum halepense), green foxtail (Setaria representation of the real world from the perspective of the viridis), hemlock (Conium maculatum), common chicory intended uses of the model [21]. SSEBop at basin and global (Cichorium intybus), and red clover (Trifolium pratense). scales is a good example of the successful application of During the observation years, the same mix of grass species satellite-based ET drought monitoring [22]. SSEBop in- was used and meat goats rotationally and intermittently creases our understanding of the global water cycle by grazed the pasture for a maximum of two weeks beginning in mapping seasonal and year-to-year changes in ET across the late June through late summer (August). Grazing was landscape and providing complete spatial ET coverage, allowed until the stub height reached 0.1 m. supplying input data aiding simulating irrigation require- ments, and assessing trends of ET. Monthly maps of actual and anomaly ET and products are available publicly via the 2.3. Latent and Sensible Heat Measurements. Vertical ex- Famine Early Warning System (https://app.climateengine. change of H O vapor, sensible heat, and latent heat were org/climateEngine). measured using the EC technique. (e EC flux tower is Few studies have evaluated SSEBop across wide geo- located on a pasture with a sufficiently wide and horizontal graphical areas (see noted exceptions [19]) and less fre- fetch of at least 200–300 m in the major wind (westerly and quently on ecosystem scales. (erefore, the goal of this study southwesterly) direction. (e EC tower was instrumented was to evaluate the performance of SSEBop in estimating ET with a fast-response 3D sonic anemometer (WindMaster from a pasture agroecosystem under the climate and soil Pro; Gill Instruments, Lymington, Hampshire, UK) that conditions of north central Kentucky. Although the SSEBop measures wind speed (m s ) and sonic air temperature (T s, model has been in use for 18 years, validation studies on its C). (e wind sensor is located 2.82 m above the ground use over varied ecosystems are limited. (e objectives of this level. Water vapor concentration was measured in situ using study were (i) to evaluate SSEBop against EC data and its a closed path infrared CO /H O gas analyzer (LI-7200; LI- 2 2 comparison to reference evapotranspiration (ET ) as its COR Inc., Lincoln, NE, USA). (e two sensors are installed ref upper boundary and (ii) to assess the application of SSEBop such that they are within the mixed surface layer, i.e., Advances in Meteorology 3 30 250 –5 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Month Month 1981–2010 2017 2016 2016 2018 2017 (a) (b) Figure 1: Average air temperature ( C) (a) and monthly annualized precipitation (mm) (b) at the Harold R. Benson Research and Demonstration Farm for the observation period of 2016–2018. Precipitation records are from the Franklin County Mesonet Weather Data ° ° (Lat: 38 :12 N; Lon: −84.88 W, which is located about 670 m from the flux station. Long-term (normal, gray bars) data are from the National ° ° Weather Service Capital City Airport, Station no. US W00053841 (Lat: 38.1847 N; Lon: −84.9033 W) (National Oceanic and Atmospheric Administration, National Environmental Satellite Data and Information Service (NOAA/NCDC)) (https://gis.ncdc.noaa.gov/maps/ncei/ normals)) [24]. z > 5z + d, where z is the measurement height (m), z is annually). We assessed the relative energy balance by h m h m the surface roughness length (m), and d is the zero plane plotting the sum of H and LE against the available energy displacement height (m; the theoretical height above the (i.e., R -G) and the mathematical expression for a complete ground where the logarithmic wind profile vanishes to zero). energy closure assumes (e measurement height z was nearly 9 times the average z h h y + a R − G � H + LE + G + ε, (4) of the grass canopy height (0.30 m at peak developmental stage). where a and y are the slope and intercept of the linear re- (e turbulent sensible (H) heat flux was computed as gression, respectively, and G is ground heat flux. Unit for R , −2 follows: G, LE, and H terms is W·m . (e last term, ε, is the error term, which includes both instruments and random errors. G ′ ′ (1) H � ρ C w T , a P s was directly measured using a soil heat plate at a 10-cm depth. −3 where ρ is dry air density (kg m ) and C is the specific heat a p −1 −1 capacity of the air at constant pressure (1004 J kg K ). (e 2.4. Penman–Monteith (P–M) Method. Reference evapo- overbar indicates the averaging period, in this case 30 min. −2 transpiration (ET , mm/d) measures the rate at which water Similarly, latent heat (LE, W ) was calculated as ref vapor is released from actively growing vegetation (typically, ′ ′ (2) LE � λw q , grass height of vegetation is assumed to be 0.12 m) unlimited −1 by water with a fixed crop surface resistance of 70 s·m and −1 where λ is the latent heat of vaporization (2.45 MJ·kg ) and an albedo of 0.23 [25, 26]. (e FAO-56 (Food and Agri- q is the fluctuation about the mean of density of water vapor ′ culture Organization) Penman–Monteith (P–M) method is −3 −3 ’ (kg m ), ρ , in air (1000 kg m ), and w and q are the currently considered as the standardized method of com- turbulent vertical wind speed (m/s) and water vapor density puting reference ET from an assumed uniform grass ref- −3 (kg m ), respectively. Overbar indicates the averaging erence surface worldwide [26]. (e method computes period, which in this case was 30 min. Daily ET (mm/day) −1 reference evapotranspiration as daily totals (mm day ) was estimated using the following relationship: using the following equations: LE ET � . (3) ρ λ 0.408Δ