# Combining Electrical Resistivity Tomography and Satellite Images for Improving Evapotranspiration Estimates of Citrus Orchards

^{1}

^{2}

^{*}

## Abstract

**:**

_{c}) FAO-56 approach. Sentinel 2-A imagery were used to compute vegetation indices (VI

_{s}) required for spatially estimating ET. The potentiality of the ERT technique was exploited for tracking the soil wetting distribution patterns during and after irrigation phases. The ERT-derived information helped to accurately estimate the wet exposed fraction (f

_{ew}) and therefore the water evaporated from the soil surface into the dual K

_{c}FAO-56 approach. Results, validated by site-specific ET measurements (ET

_{EC}) obtained using the eddy covariance (EC) technique, showed that ERT-adjusted ET estimates (ET

_{ERT}) were considerably reduced (15%) when compared with the original dual K

_{c}FAO-56 approach (ET

_{FAO}), soil evaporation overestimation being the main reason for these discrepancies. Nevertheless, ET

_{FAO}and ET

_{ERT}showed overestimations of 64% and 40% compared to ET

_{EC}. This is because both approaches determine ET under standard conditions without water limitation, whereas EC is able to determine ET even under soil water deficit conditions. From the comparison between ET

_{EC}and ET

_{ERT}, the water stress coefficient was experimentally derived, reaching a mean value for the irrigation season of 0.74. The obtained results highlight how new technologies for soil water status monitoring can be incorporated for improving ET estimations, particularly under drip irrigation conditions.

## 1. Introduction

_{c}), where T and E are considered together, and a more complex methodology (dual K

_{c}FAO-56), where both T and E components are determined separately (i.e., basal crop coefficient K

_{cb}and evaporation coefficient K

_{e}). Both models can be combined with spectral data provided using remote sensors in order to provide spatially distributed ET estimates [15]. Although K

_{cb}represents a specific crop characteristics index that varies only to a limited extent with climate, K

_{e}can vary considerably depending on the time interval between wetting events, the magnitude of the wetting event, and the E power of the atmosphere. Nevertheless, these aspects are not deeply addressed in the FAO-56 approach. Thus, the use of alternative techniques is required in order to obtain more accurate K

_{e}estimates.

_{e}needs to take into account the spatial and temporal distribution of irrigation-wetting patterns that are governed by static and dynamic conditions, such as soil characteristics (e.g., hydraulic parameters, texture, structure, initial water content), irrigation systems (types, emitter spacing, discharge rate, and irrigation frequency), and root distribution. In general, soil wetting patterns can be obtained using in situ soil water (SW) measurements. Caution needs to be applied on the use of SW measurements, since site conditions (compaction layers or surface soil conditions) may be quite site-specific, and installation of instrumentation can affect the soil wetting patterns being measured [16]. Undisturbed methods, such as dye tracers, often combined with flow and/or transport modeling, have been used to describe infiltration from a point/line source; however, models do not at present fully reflect the current state of process understanding and empirical knowledge of preferential flow [17]. A number of near surface geophysics methods have been adopted for image irrigation wetting patterns [18]. Among these minimally invasive methods, electrical resistivity tomography (ERT) has the main advantage of being sensitive in monitoring soil-plant interactions in terms of SW relationships with high-resolution scale both in two and/or three dimensions [19,20,21,22]. Furthermore, ERT can provide useful information on soil wetting patterns both spatially and temporally distributed if applied in time-lapse mode.

_{c}FAO-56 approach for estimating E and assessing its influence on ET by comparison with EC measurements.

## 2. Materials and Methods

_{c}FAO-56 approach is proposed. A schematic summary of the adopted methodology is reported in Figure 1.

_{c}FAO-56 approach [14] was adjusted with ERT-derived data that provided site-specific information on the exposed wetting fraction (f

_{ew}). This parameter represents the fraction of soil that is both exposed and wetted (from which most evaporation occurs). As reported in [14], f

_{ew}contributes in the definition of the soil evaporation coefficient (K

_{e}, Equation (2)) and as well in ET determination (see Equation (1)). ERT surveys provided f

_{ew}that was incorporated as a parameter into the soil water balance (SWB) model (Equation (4)) within the dual K

_{c}FAO-56 approach, permitting the obtainment of an ERT-adjusted evaporation coefficient (K

_{e,ERT}) and evapotranspiration estimates (ET

_{ERT}).

_{c}FAO-56 (ET

_{FAO}) and the ERT-adjusted model (ET

_{ERT}) was performed with EC-based ET measurements (ET

_{EC}) collected in situ. In the following sub-sections, the materials and methods are described in detail.

#### 2.1. Original Model Description

_{c}FAO-56 approach determines crop evapotranspiration (ET

_{FAO}) based on the concepts of reference evapotranspiration (ET

_{0}) and separate coefficients for crop T (represented by the basal crop coefficient, K

_{cb}) and E (K

_{e}), as follows:

_{0}(mm d

^{−1}) is estimated by using the Penman–Monteith equation with hourly weather data (see 1.2) supplied by a weather station located close to the study site.

_{e}depends on the water available in the surface layer of the topsoil and its estimation requires a daily water balance (SWB) computation for the surface soil layer in order to determine the cumulative E or depletion from the wet condition, as follows:

_{c,max}is the maximum crop coefficient value following rain or irrigation while K

_{r}represents the reduction applied to E depending on the amount of water evaporated from the soil, as follows:

_{FC}− 0.5 θ

_{WP}) Z

_{e}is the total evaporable water (i.e., maximum depth of water that can be evaporated from the soil surface layer), with θ

_{FC}and θ

_{FC}measured soil field capacity and wilting point (see 2.3) at the study site and Z

_{e}equal to 0.1, as reported in [14]; REW is the readily evaporable water (fixed at 10 mm for the study site soil, i.e., sandy loam [14]); and D

_{e,i}is the cumulative depth of evaporation from the topsoil (mm) at the end of the 1-th day, and it is solved as follows:

_{e,i−1}is the cumulative depth of E following complete wetting from the exposed and wetted fraction of the topsoil at the end of day i − 1 (mm); P

_{i}is precipitation on day i (mm); RO

_{i}is precipitation runoff from the soil surface on day i (mm); I

_{i}is irrigation depth on day i that infiltrates the soil (mm); E

_{i}is evaporation on day i (i.e., E

_{i}= K

_{e}ET

_{0}) (mm); T

_{ew,i}is depth of T from the exposed and wetted fraction of the soil surface layer on day i (mm); DP

_{e,i}is deep percolation loss from the topsoil layer on day i if SW exceeds θ

_{FC}(mm); f

_{w}is the fraction of soil surface wetted by irrigation (0.35 for drip irrigation [14]); f

_{ew}is exposed and wetted soil fraction, computed as the lowest value between the average exposed soil fraction not covered (or shaded) by vegetation and f

_{w}[14]. Therefore, f

_{ew}calculation within the SWB depends also on the occurrence of irrigation and precipitation and it is calculated differently for each scenario as: (1) if the surface is wetted by irrigation, then f

_{w}is the f

_{w}for the irrigation system, and therefore f

_{ew}is equal to 0.35 (this value is obtained from [14]); (2) if the surface is wetted by significant rain (i.e., >3 to 4 mm) with no irrigation, f

_{w}= 1 and therefore f

_{ew}is (1 − f

_{c}); and (3) if there is neither irrigation nor significant precipitation, f

_{ew}is the f

_{ew}of the previous day. However, the value of 0.35 considered in this approach, as it is a general theoretical f

_{ew}procedure, may not reflect the site-specific conditions, which could be more accurately determined performing local measurements (as the ERT-adjustment described in Section 2.2 “ERT-adjusted model parameter”).

#### 2.2. ERT-Adjusted Model Parameter

_{ew}values proposed in [14] can be adjusted in order to account for the site specificities (such as soil type and depth, and dripper flow and density, among others). This is the case of the novel ERT-adjusted dual K

_{c}FAO-56 approach proposed herein. In this approach, K

_{r,ERT}(Equation (3)) and K

_{e,ERT}(Equation (2)) were calculated by solving the SWB model (Equation (4)) using the f

_{ew}information provided by ERT instead of using the FAO-56 proposed f

_{ew}value as in the original FAO-56 model, thus ET

_{ERT}was derived by including the modified K

_{e,ERT}term in Equation (1).

_{r}, Equation (5)) between the electrical resistances collected before and after irrigation:

_{t}and d

_{0}are the electrical resistance values (Ω) at time t and time 0 (initial condition), and F(σ

_{ohm}) is the electrical resistance (Ω), obtained by running the forward model for an arbitrary ER of 100 Ω m. This calculation was performed simultaneously for T1 and T2 using a 5% error level [22].

_{ew}parameter used to set the SWB model (Equation (4)) within the ERT-adjusted dual K

_{c}FAO-56 approach was retrieved from the volume derived by ERT. From the entire volume, the ER ratio values corresponding uniquely to the first 10 cm were extracted, since it was assumed that it is mainly at this depth where soil evaporation occurs. Once the data corresponding to the first 10 cm were extracted, f

_{ew}was determined by recognizing the soil wetting patterns (i.e., extracting all the values corresponding to the fixed threshold) at the different acquisition times after the irrigation beginning (Figure 2a and Table 1). Finally, the constant f

_{ew}value used in the ERT-adjusted approach was obtained as an average of all instantaneous values retrieved in each acquisition time (Figure 2a and Table 1).

#### 2.3. Satellite-Based Dual Kc Approach

_{c}FAO-56 approach (original and ERT-adjusted), applied in this study, incorporates data derived from remote sensing in order to obtain spatially distributed estimates of K

_{e}(K

_{e,FAO}and K

_{e,ERT}), K

_{cb}(Equations (2) and (6)), and ET (ET

_{FAO}and ET

_{ERT}) (Equation (1)).

_{cb}(Equations (1) and (2)) as a function of the soil adjusted vegetation index (SAVI) following the methodology proposed by [25,26]:

_{max}and SAVI

_{min}refer to the maximum and minimum SAVI values for each image, and F

_{c,max}is the maximum value of fractional vegetation cover (f

_{c}) within the study site for which K

_{cb}reaches its maximum value (as in [14]).

_{NIR}and ρ

_{RED}are the infrared and red reflectance of Sentinel images and L is a soil normalization factor, generally taken to be 0.5 [27].

_{c,max}(Equation (6)) and f

_{ew}[14], f

_{c}is calculated as reported in [28]:

_{NIR}− ρ

_{RED})/(ρ

_{NIR}+ ρ

_{RED}) [29]. The value of NDVI

_{max}(set to 1) corresponds with the NDVI when f

_{c}is maximum (f

_{c}= 1) whereas NDVI

_{min}(set to 0) refers to the NDVI value when the surface is without vegetation (f

_{c}≈ 0). Within the study period, the f

_{c}at the study site ranged between 0.337 and 0.646, with an average value of 0.483.

#### 2.4. Ancillary Weather and Soil Data

_{c}FAO-56 model consists of weather observations and soil hydraulic characteristics referring to an experimental orange orchard of 0.7 ha located in southern Italy (Lentini, SR) and managed by Centro di Ricerca Olivicoltura, Frutticoltura e Agrumicoltura of the Italian Council for Agricultural Research and Agricultural Economics Analyses (CREA-OFA, Acireale). The orange orchard has been treated by deficit irrigation strategies, including partial root-zone drying (PRD) and regulated deficit irrigation (RDI), since 2010. The complete description of the experimental site and the irrigation strategies applied are reported in [21,30,31].

_{air}, °C; relative humidity, RH, %; precipitation, P, mm; wind speed, u, m s

^{−1}; and reference evapotranspiration, ET

_{0}, mm), were analyzed in order to initiate/calibrate and implement the SWB model within the dual K

_{c}FAO-56 model.

_{0}, maximum T

_{air}and P) is shown in Figure 3.

#### 2.5. Evapotranspiration Validation Using EC

_{2}O and CO

_{2}concentrations, respectively. The sample frequency for the raw data was 10 Hz (high frequency data) [31]. Low frequency data (30-min) were obtained for: net radiation (R

_{n}, W m

^{−2}, net radiometer CNR-1 Kipp & Zonen, located 7 m above the ground) and soil heat flux (G, W m

^{−2}), obtained using self-calibrated soil heat flux plates (HFP01SC, Hukseflux) placed in the exposed, half-exposed, and shadowed soil at a depth of about 0.05 m.

_{EC}) by the direct measurements of latent heat flux (λET, W m

^{−2}) exchanged within the soil-plant-atmosphere continuum, using the following equation:

^{−}

^{1}) is the latent heat of vaporization and σ

_{wq}(g m

^{−2}s

^{−1}) is the covariance between the vertical wind speed and water vapour density.

^{−2}) is computed as:

^{−3}) is the air density, c

_{p}(1004 J g

^{−1}K

^{−1}) is the air specific heat capacity at constant pressure, and σ

_{wT}(m s

^{−1}K) is the covariance between the vertical wind speed and air temperature.

_{n}is greater than 100 W m

^{−2}. In this study, the CR was forced according to the procedure proposed by [37], in order to maintain the observed Bowen ratio between H and LE as constant.

^{−2}, were then transformed to equivalent depth of ET (mm d

^{−1}). In this study, ET

_{EC}measurements were used as a reference to compare the ET estimates obtained by the original satellite-based dual K

_{c}FAO-56 (ET

_{FAO}) approach and the ET estimates (ET

_{ERT}) obtained by the adjusted model with ERT-derived parameters (f

_{ew}).

_{EC}measures (mm d

^{−1}) and ET

_{0}by weather station (mm d

^{−1}), the crop coefficient using EC (K

_{c,EC}) was estimated, as in the following:

#### 2.6. Water Stress Coefficient Determination

_{c}FAO 56 approaches (both the original and ERT-adjusted) compute ET under standard conditions (i.e., water stress coefficient, K

_{s}, equal to 1), whereas ET measured using the EC technique (ET

_{EC}) incorporates soil water stress condition (ET

_{EC}= K

_{s}K

_{c,EC}ET

_{0}; K

_{c,EC}being the hypothetical K

_{c}value that will be measured by EC in the absence of water stress). Thus, assuming that K

_{c,ERT}is equal to K

_{c,EC}, K

_{s}can be empirically derived as:

## 3. Results

#### 3.1. Evapotranspiration Rates using EC

_{EC}rates within the reference period (June–September 2017) is shown in Figure 4. ET

_{EC}rates ranged between 0.94 and 3.50 mm day

^{−1}, with a mean value of 2.24 mm d

^{−1}. Prior to the CR adjustment, the slope of the regression forced through the origin of the CR was around 0.82, with a determination coefficient (R

^{2}) of about 0.90.

#### 3.2. Soil Wetting Distribution Patterns Using ERT

_{ew}) during and after the irrigation phase in T1 and T2. The mean f

_{ew}value obtained from both T1 and T2 was 0.1. This value was used for running the SWB model within the ERT-adjusted dual K

_{c}FAO-56 approach.

#### 3.3. Satellite dual Kc Approach

#### 3.3.1. Maps of original and ERT-adjusted dual K_{c} FAO-56

_{c}(K

_{c,FAO}and K

_{c,ERT}; a,b) and ET (ET

_{FAO}and ET

_{ERT}; c,d) obtained for the study area by the original and ERT-adjusted dual K

_{c}FAO-56 approach for DOY 193 (under ET

_{0}conditions of 8.56 mm day

^{−1}). The dual K

_{c}values derived from FAO-56 and ERT-adjusted approaches were 0.69 and 0.59, respectively, resulting in ET values of 5.91 (ET

_{FAO}) and 5.06 mm (ET

_{ERT}).

#### 3.3.2. ET Comparison: Original and ERT-Adjusted Dual K_{c} FAO-56 vs EC

_{EC}, the original (ET

_{FAO}), and the ERT-adjusted (ET

_{ERT}) dual K

_{c}FAO-56 approaches satellite ET estimates.

_{FAO}and ET

_{ERT}, with average values of 4.07 and 3.46 mm, respectively) resulted in an average 64% and 40% greater than the measured ET

_{EC}fluxes (average value of 2.47 mm), with root mean square errors (RMSE) of 1.74 and 1.17 mm day

^{−1}and coefficients of determination (R

^{2}) of 0.48 and 0.62, respectively. The slope terms were 1.64 and 1.41 for ET

_{FAO}and ET

_{ERT}respectively, reflecting that ET discrepancies with respect to ET

_{EC}were greater for high ET values. The average T component was 3.19 mm for both original and ERT-adjusted dual K

_{c}FAO-56 approaches, whereas E term was 0.88 mm and 0.27 mm, respectively.

#### 3.3.3. Crop Coefficients Comparison and K_{s} Estimation

_{cb}+ K

_{e}) obtained from EC (K

_{c,EC}) and from the original (K

_{c,FAO}) and ERT-adjusted (K

_{c,ERT}) dual K

_{c}FAO-56 approaches. Within the reference period (June–September 2017), the observed K

_{c}using EC (K

_{c,EC}) was 0.40 ± 0.08. For the same period, K

_{c,FAO}and K

_{c,ERT}resulted in 0.64 ± 0.12 and 0.54 ± 0.11, respectively.

_{s}) obtained as Equation (13) is reported in Figure 9. K

_{s}values ranged from 0.55 to 1.00 with an average of 0.74.

## 4. Discussion

_{ew}information retrieved using ERT were herein included in the adjusted satellite dual K

_{c}FAO-56 approach. The obtained results showed that ERT improved ET estimates (and E), with respect to the estimates obtained by the original dual K

_{c}FAO-56 approach, when compared with the site-specific ET

_{EC}rates measured at the experimental site by EC. In fact, the comparison between K

_{c,FAO}and K

_{c,ERT}reveals that K

_{c,ERT}was always lower than K

_{c,FAO}, this discrepancy being due to the more accurate estimation of f

_{ew}performed in the ERT-adjusted approach. Consequently, with the ERT-adjusted approach, the E term, and therefore ET, was considerably reduced (15%) when compared with the original dual K

_{c}FAO-56 approach [14]. Nevertheless, ET obtained from both approaches, even considering the ERT-adjustment, remained substantially higher than ET measured in EC (64% and 40%, respectively). Such overestimations could be due to the assumption taken in this study of considering f

_{ew}constant during the day, since it is well known that f

_{ew}progressively diminish after an irrigation event. Additionally, the presence of different irrigation treatments within the footprint of the EC tower may introduce some uncertainties in the results obtained. In addition, the spatial resolution of Sentinel does not allow the separation of regions irrigated differentially. Therefore, this limitation could be solved by using high spatial resolution images, such as those acquired by unmanned aerial vehicles, which would allow the f

_{ew}of each irrigation treatment to be considered separately instead of averaging both of them, as done for mixed pixels. Similar to the results obtained in this study, several authors have found ET overestimation ranging from 12% to 42% when comparing ET from the FAO 56 approach with ET provided by EC in heterogeneous orchards under drip irrigation [44,45] indicating that the overestimation was even worse when examining only the irrigated period [46]. These authors pointed out that the values of K

_{cb}suggested by [14,45,46] and the high soil evaporation predicted following the FAO-56 approach [45,46] as the main reasons for these overestimations. In general, the magnitude of such overestimations was lower than the one obtained in this study probably due to the water stress expected in our experiment as consequence of the deficit irrigation conditions applied at the experimental site. Deficit irrigation strategy played a strategic role in altering the normal ratio between the energy balance surface fluxes, determining a fairly high sensible heat flux and an ET reduction or underestimation due to the imposed water stress conditions [31]. This behavior was poorly captured by the original and the ERT-adjusted K

_{c}FAO-56 approaches, whereas it was taken into account in the ET

_{EC}value obtained using the EC technique, as indicated by the calculated K

_{s}(≈0.74).

_{ew}to better characterize changes in soil water conditions and subsequently evaporation, in order to incorporate dynamic f

_{ew}values into the approach instead of using a constant one. At this stage, ERT may be considered a useful tool for precision irrigation strategies, in particular for identifying the soil wetting patterns distribution and also allowing a better characterization of the wet bulb, which may therefore improve the efficiency of irrigation [21]. Currently, the scope of ERT is limited to scientific research or as a validation method for calibrating other methods that can be more easily incorporated into the daily activities of farmers and technicians. In the future, more commercially oriented applications of ERT technologies could be derived in order to facilitate the implementation of this technique for agriculture water management applications.

## 5. Conclusions

- Spatially distributed ET rates can be obtained by incorporating VI
_{s}computed using remote sensing technologies into the dual K_{c}FAO-56 approach. - The integration of 3-D ERT methodology into the dual K
_{c}FAO-56 approach considerably reduced errors in ET estimates. This technology allowed the tracking of the wetting distribution patterns, helping to accurately estimate f_{ew}and therefore the water evaporated from the soil surface. - The dual K
_{c}FAO-56 approach determines ET under standard conditions where no limitations are placed on crop growth or ET, whereas EC measures ET even for non-standard conditions (e.g., under soil water stress conditions). From the comparison between the ET measured from the EC tower and the ET estimated from the ERT-adjusted dual K_{c}FAO-56 approach, the K_{s}term can be experimentally derived.

## Author Contributions

## Funding

## Acknowledgments

## Conflicts of Interest

## References

- French, R.J.; Schultz, J.E. Water use efficiency of wheat in a Mediterranean type environment. I. The relation between yield, water use and climate. Aust. J. Agric. Res.
**1984**, 35, 743–764. [Google Scholar] [CrossRef] - French, R.J.; Schultz, J.E. Water use efficiency of wheat in a mediterranean-type environment. II. Some limitations to efficiency. Aust. J. Agric. Res.
**1984**, 35, 765–775. [Google Scholar] [CrossRef] - Capra, A.; Consoli, S.; Russo, A.; Scicolone, B. Integrated agro-economic approach to deficit irrigation on lettuce crops in Sicily (Italy). J. Irrig. Drain. Eng.
**2008**, 134, 437–445. [Google Scholar] [CrossRef] - Consoli, S.; Papa, R. Corrected surface energy balance to measure and model the evapotranspiration of irrigated orange orchards in semi-arid Mediterranean conditions. Irrig. Sci.
**2013**, 31, 1159–1171. [Google Scholar] [CrossRef] - Zhao, P.; Kang, S.; Li, S.; Ding, R.; Tong, L.; Du, T. Seasonal variations in vineyard ET partitioning and dual crop coefficients correlate with canopy development and surface soil moisture. Agric. Water Manag.
**2018**, 197, 19–33. [Google Scholar] [CrossRef] - Yau, S.K.; Nimah, M.; Farran, M. Early sowing and irrigation to increase barley yields and water use efficiency in Mediterranean conditions. Agric. Water Manag.
**2011**, 98, 1776–1781. [Google Scholar] [CrossRef] - Yan, A.; Gao, C.; Ren, Y.; Zong, R.; Ma, Y.; Li, Q. Effects of pre-sowing irrigation and straw mulching on the grain yield and water use efficiency of summer maize in the North China Plain. Agric. Water Manag.
**2017**, 186, 21–28. [Google Scholar] [CrossRef] - Eastham, J.; Gregory, P.J.; Williamson, D.R.; Watson, G.D. The influence of early seeding of wheat and lupin crops on evapotranspiration and evaporation from the soil surface in a Mediterranean climate. Agric. Water Manag.
**1999**, 42, 205–218. [Google Scholar] [CrossRef] - Liu, C.M.; Zhang, X.Y.; Zhang, Y.Q. Determination of daily evaporation and evapotranspiration of winter wheat and maize by large-scale weighing lysimeter and micro-lysimeter. Agric. For. Meteorol.
**2002**, 111, 109–120. [Google Scholar] [CrossRef] - Photiades, A.; Hadjichristodoulou, A. Sowing date, sowing depth, seed rate and row spacing of wheat and barley under dryland conditions. Field Crops Res.
**1984**, 9, 151–162. [Google Scholar] [CrossRef] - Lascano, R.J.; Baumhardt, R.L.; Hicks, S.K.; Heilman, J.L. Soil and plant water evaporation from strip-tillage control, measurement and simulation. Agron. J.
**1994**, 86, 987–994. [Google Scholar] [CrossRef] - Stagnari, F.; Galieni, A.; Speca, S.; Cafiero, G.; Pisante, M. Effects of straw mulch on growth and yield of durum wheat during transition to Conservation Agriculture in Mediterranean environment. Field Crops Res.
**2014**, 167, 51–63. [Google Scholar] [CrossRef] - Prosdocimi, M.; Jordan, A.; Tarolli, P.; Keesstra, S.; Novara, A.; Cerda, A. The immediate effectiveness of barley straw mulch in reducing soil erodibility and surface runoff generation in Mediterranean vineyards. Sci. Total Environ.
**2016**, 547, 323–330. [Google Scholar] [CrossRef] [PubMed] [Green Version] - Allen, R.G.; Pereira, L.S.; Raes, D.; Smith, M. Crop Evapotranspiration-Guidelines for Computing Crop Water Requirements-FAO Irrigation and Drainage Paper 56; FAO: Rome, Italy, 1998; Volume 300, p. D05109. [Google Scholar]
- González-Dugo, M.P.; Neale, C.M.U.; Mateos, L.; Kustas, W.P.; Prueger, J.H.; Anderson, M.C.; Li, F. A comparison of operational remote sensing-based models forestimating crop evapotranspiration. Agric. For. Meteorol.
**2009**, 149, 1843–1853. [Google Scholar] [CrossRef] - Subbaiah, R. A review of models for predicting soil water dynamics during trickle irrigation. Irrig. Sci.
**2013**, 31, 225–258. [Google Scholar] [CrossRef] - Jarvis, N.; Koestel, J.; Larsbo, M. Understanding preferential flow in the vadose zone: Recent advances and future prospects. Vadose Zone J.
**2016**, 15, 15. [Google Scholar] [CrossRef] - Hardie, M.; Ridges, J.; Swarts, N.; Close, D. Drip irrigation wetting patterns and nitrate distribution: Comparison between electrical resistivity (ERI), dye tracer, and 2D soil–water modelling approaches. Irrig. Sci.
**2018**, 36, 97–110. [Google Scholar] [CrossRef] - Cassiani, G.; Boaga, J.; Vanella, D.; Perri, M.T.; Consoli, S. Monitoring and modelling of soil–plant interactions: The joint use of ERT, sap flow and eddy covariance data to characterize the volume of an orange tree root zone. Hydrol. Earth Syst. Sci.
**2015**, 19, 2213–2225. [Google Scholar] [CrossRef] - Cassiani, G.; Boaga, J.; Rossi, M.; Putti, M.; Fadda, G.; Majone, B.; Bellin, A. Soil–plant interaction monitoring: Small scale example of an apple orchard in Trentino, North-eastern Italy. Sci. Total Environ.
**2016**, 543, 851–861. [Google Scholar] [CrossRef] - Consoli, S.; Stagno, F.; Vanella, D.; Boaga, J.; Cassiani, G.; Roccuzzo, G. Partial root-zone drying irrigation in orange orchards: Effects on water use and crop production characteristics. Eur. J. Agron.
**2017**, 82, 190–202. [Google Scholar] [CrossRef] - Vanella, D.; Cassiani, G.; Busato, L.; Boaga, J.; Barbagallo, S.; Binley, A.; Consoli, S. Use of small scale electrical resistivity tomography to identify soil-root interactions during deficit irrigation. J. Hydrol.
**2018**, 556, 310–324. [Google Scholar] [CrossRef] - Binley, A.M.; Kemna, A. DC resistivity and induced polarization methods. In Hydrogeophysics; Rubin, Y., Hubbard, S.S., Eds.; Springer: Dordrecht, The Netherlands, 2005; Volume 50, pp. 129–156. [Google Scholar] [CrossRef]
- Binley, A. Tools and techniques: DC electrical methods. In Treatise on Geophysics, 2nd ed.; Schubert, G., Ed.; Elsevier: Amsterdam, The Netherlands, 2015; Volume 11, pp. 233–259. [Google Scholar] [CrossRef]
- Consoli, S.; Vanella, D. Comparisons of satellite-based models for estimating evapotranspiration fluxes. J. Hydrol.
**2014**, 513, 475–489. [Google Scholar] [CrossRef] - Consoli, S.; Vanella, D. Mapping crop evapotranspiration by integrating vegetation indices into a soil water balance model. Agric. Water Manag.
**2014**, 143, 71–81. [Google Scholar] [CrossRef] - Huete, A.R. A soil-adjusted vegetation index (SAVI). Remote Sens. Environ.
**1988**, 25, 295–309. [Google Scholar] [CrossRef] - Gutman, G.; Ignatov, A. The derivation of the green vegetation fraction from NOAA/AVHRR data for use in numerical weather prediction models. Int. J. Remote Sens.
**1998**, 19, 1533–1543. [Google Scholar] [CrossRef] - Rouse, J.W.; Haas, R.H.; Schell, J.A.; Deering, D.W.; Harlan, J.C. Monitoringthe Vernal Advancement and Retrogradation of Natural Vegetation; Type III, Final Report; NASA/GSFC: Greenbelt, MD, USA, 1974; pp. 1–371.
- Consoli, S.; Stagno, F.; Roccuzzo, G.; Cirelli, G.; Intrigliolo, F. Sustainable management of limited water resources in a young orange orchard. Agric. Water Manag.
**2014**, 132, 60–68. [Google Scholar] [CrossRef] - Vanella, D.; Consoli, S. Eddy Covariance fluxes versus satellite-based modelisation in a deficit irrigated orchard. Ital. J. Agrometeorol.
**2018**, 2, 41–52. [Google Scholar] [CrossRef] - Aiello, R.; Bagarello, V.; Barbagallo, S.; Consoli, S.; Di Prima, S.; Giordano, G.; Iovino, M. An assessment of the Beerkan method for determining the hydraulic properties of a sandy loam soil. Geoderma
**2014**, 235, 300–307. [Google Scholar] [CrossRef] - Baldocchi, D.D. Assessing the eddy covariance technique for evaluating carbon dioxide exchange rates of ecosystems: Past, present and future. Glob. Chang. Biol.
**2003**, 9, 479–492. [Google Scholar] [CrossRef] - Aubinet, M.; Grelle, A.; Ibrom, A.; Rannik, U.; Moncrieff, J.; Foken, T.; Kowalski, A.S.; Martin, P.H.; Berbigier, P.; Bernhofer, C.; et al. Estimates of the annual net carbon and water exchange of European forests: The EUROFLUX methodology. Adv. Ecol. Res.
**2000**, 30, 113–175. [Google Scholar] - Kaimal, J.C.; Finnigan, J.J. Atmospheric Boundary-Layer Flows: Their Structure and Measurement; Oxford University Press: Oxford, UK, 1994; 289p. [Google Scholar]
- Prueger, J.H.; Hatfield, J.L.; Kustas, W.P.; Hipps, L.E.; MacPherson, J.I.; Parkin, T.B. Tower and aircraft eddy covariance measurements of water vapor, energy and carbon dioxide fluxes during SMACEX. J. Hydrometeorol.
**2005**, 6, 954–960. [Google Scholar] [CrossRef] - Twine, T.E.; Kustas, W.P.; Norman, J.M.; Cook, D.R.; Houser, P.; Meyers, T.P.; Prueger, J.H.; Starksh, P.J.; Wesely, M.L. Correcting eddy-covariance flux underestimates over a grassland. Agric. For. Meteorol.
**2000**, 103, 279–300. [Google Scholar] [CrossRef] [Green Version] - Wang, K.; Liang, S. An improved method for estimating global evapotranspiration based on satellite determination of surface net radiation, vegetation index, temperature, and soil moisture. J. Hydrometeorol.
**2008**, 9, 712–727. [Google Scholar] [CrossRef] - Knipper, K.; Hogue, T.; Scott, R.; Franz, K. Evapotranspiration estimates derived using multi-platform remote sensing in a semiarid region. Remote Sens.
**2017**, 9, 184. [Google Scholar] [CrossRef] - Sun, G.; Alstad, K.; Chen, J.; Chen, S.; Ford, C.R.; Lin, G.; Liu, C.; Lu, N.; McNulty, S.G.; Miao, H.; et al. A general predictive model for estimating monthly ecosystem evapotranspiration. Ecohydrology
**2011**, 4, 245–255. [Google Scholar] [CrossRef] - Consoli, S.; Licciardello, F.; Vanella, D.; Pasotti, L.; Villani, G.; Tomei, F. Testing the water balance model criteria using TDR measurements, micrometeorological data and satellite-based information. Agric. Water Manag.
**2016**, 170, 68–80. [Google Scholar] [CrossRef] - Kim, J.H.; Yi, M.J.; Park, S.G.; Kim, J.G. 4-D inversion of DC resistivity monitoring data acquired over a dynamically changing earth model. J. Appl. Geophys.
**2009**, 68, 522–532. [Google Scholar] [CrossRef] - Busato, L.; Boaga, J.; Perri, M.T.; Majone, B.; Bellin, A.; Cassiani, G. Hydrogeophysical characterization and monitoring of the hyporheic and riparian zones: The Vermigliana Creek case study. Sci. Total Environ.
**2018**, 648, 1105–1120. [Google Scholar] [CrossRef] - Paço, T.A.; Ferreira, M.I.; Conceição, N. Peach orchard evapotranspiration in a sandy soil: Comparison between eddy covariance measurements and estimates by the FAO 56 approach. Agric. Water Manag.
**2006**, 85, 305–313. [Google Scholar] [CrossRef] - Er-Raki, S.; Chehbouni, A.; Guemouria, J.; Ezzahar, J.; Khabba, S.; Boulet, G.; Hanich, L. Citrus orchard evapotranspiration: Comparison between eddy covariance measurements and the FAO-56 approach estimates. Plant Biosyst.
**2009**, 143, 201–208. [Google Scholar] [CrossRef] [Green Version] - Maestre-Valero, J.F.; Testi, L.; Jiménez-Bello, M.A.; Castel, J.R.; Intrigliolo, D.S. Evapotranspiration and carbon exchange in a citrus orchard using eddy covariance. Irrig. Sci.
**2017**, 35, 397–408. [Google Scholar] [CrossRef]

**Figure 2.**Evolution of the exposed wetted area (f

_{ew}) (

**a**) and evolution of the infiltration front depth (m) in T1 and T2 during and after an irrigation phase (

**b**). Time is expressed in minutes after the irrigation start. The grey area represents the irrigation phase period.

**Figure 3.**Temporal evolution of weather parameters used to implement the soil water balance (SWB) model within the dual K

_{c}FAO-56 approach. Values refer to the reference period (April 2016–October 2017). P is the precipitation (mm); ET

_{0}is the reference evapotranspiration (mm); and T

_{air}is the air temperature (°C).

**Figure 4.**Temporal evolution of daily evapotranspiration measurements (ET

_{EC}) rates using eddy covariance (EC).

**Figure 5.**Time-lapse electrical resistivity (ER) ratio imagery (at the surface:

**a**–

**e**; transects under the irrigation pipeline:

**f**–

**j**; and 3-D volumes:

**k**–

**o**) for T1 with respect to the initial condition (no irrigation).

**Figure 6.**Time-lapse electrical resistivity (ER) ratio imagery (at the surface:

**a**–

**e**; transects under the irrigation line:

**f**–

**j**; and 3-D volumes:

**k**–

**o**) for T2 with respect to the initial condition (no irrigation).

**Figure 7.**Dual crop coefficient (K

_{c,FAO}and K

_{c,ERT};

**a**,

**b**) and ET (ET

_{FAO}and ET

_{ERT};

**c**,

**d**) estimates obtained by the original and ERT-adjusted dual K

_{c}FAO-56 approach for DOY 193 (under ET

_{0}conditions of 8.56 mm day

^{−1}).

**Figure 8.**Comparison between modelled ET

_{FAO}and ET

_{ERT}versus measured ET

_{EC}. The black solid line represents the 1:1 relationship.

**Figure 9.**K

_{c}from EC (K

_{c,EC}) and from the original (K

_{c,FAO}) and ERT-adjusted dual K

_{c}FAO-56 approach (K

_{c,ERT}), and K

_{s}derived from ET

_{EC}and ET

_{ERT}ratio.

**Table 1.**Three-dimensional (3-D) electrical resistivity tomography (ERT) data collection time (local time).

T1 | T2 | ||||
---|---|---|---|---|---|

Time Id | State | Starting Time | Ending Time | Starting Time | Ending Time |

00 | no irrigation | 9.17 | 9.46 | 9.29 | 10.02 |

01 | during the irrigation phase | 10.42 | 11.11 | 11.01 | 11.35 |

02 | 11.39 | 12.09 | 11.58 | 12.30 | |

03 | after the irrigation phase | 12.55 | 13.24 | 12.57 | 13.30 |

04 | 13.47 | 14.16 | 13.52 | 14.24 | |

05 | 14.41 | 15.09 | 14.43 | 15.17 |

Acquisition Dates | Day of the Year (DOY) |
---|---|

7 June 2017 | 158 |

27 June 2017 | 178 |

12 July 2017 | 193 |

17 July 2017 | 198 |

1 August 2017 | 213 |

6 August 2017 | 218 |

11 August 2017 | 223 |

16 August 2017 | 228 |

26 August 2017 | 238 |

5 September 2017 | 248 |

15 September 2017 | 258 |

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Vanella, D.; Ramírez-Cuesta, J.M.; Intrigliolo, D.S.; Consoli, S.
Combining Electrical Resistivity Tomography and Satellite Images for Improving Evapotranspiration Estimates of Citrus Orchards. *Remote Sens.* **2019**, *11*, 373.
https://doi.org/10.3390/rs11040373

**AMA Style**

Vanella D, Ramírez-Cuesta JM, Intrigliolo DS, Consoli S.
Combining Electrical Resistivity Tomography and Satellite Images for Improving Evapotranspiration Estimates of Citrus Orchards. *Remote Sensing*. 2019; 11(4):373.
https://doi.org/10.3390/rs11040373

**Chicago/Turabian Style**

Vanella, Daniela, Juan Miguel Ramírez-Cuesta, Diego S. Intrigliolo, and Simona Consoli.
2019. "Combining Electrical Resistivity Tomography and Satellite Images for Improving Evapotranspiration Estimates of Citrus Orchards" *Remote Sensing* 11, no. 4: 373.
https://doi.org/10.3390/rs11040373