Elsevier

Atmospheric Research

Volume 261, 15 October 2021, 105760
Atmospheric Research

Seasonal variability of daily evapotranspiration and energy fluxes in the Central Andes of Peru using eddy covariance techniques and empirical methods

https://doi.org/10.1016/j.atmosres.2021.105760Get rights and content

Highlights

  • The energy and water vapor fluxes in the Andes Mountain for a period of 1 year are analyzed.

  • Evapotranspiration(ET) is measured with an Eddy Covariance System and estimated with empirical methods.

  • A well-defined evapotranspiration seasonality is found.

  • The daily ET is related to the solar radiation during the wet season and to the soil moisture during the dry season.

Abstract

In this study, we analyze the mechanisms associated to evapotranspiration over the high central Peruvian Andes, a place where evapotranspiration has been poorly characterized. We made use of the eddy covariance system (sonic anemometer and a krypton hygrometer) installed at the Huancayo Observatory (12.04° S, 75.32°, 3330 m.a.s.l.) to document for the first time the hourly, daily and monthly variability of surface energy fluxes from July 2016 to June 2017. The relationship between evapotranspiration and meteorological variables is also examined. Furthermore, we evaluated the performance of three empirical equations that estimate the potential evapotranspiration to explore their adequacy in the central Peruvian Andes. These are the FAO Penman-Monteith (PM), Priestley-Taylor (PT) and Hargreaves. Likewise, the accuracy of the MODIS16A2 evapotranspiration product was also examined.

We show that evapotranspiration over the high central Peruvian Andes is modulated by the water- and energy-limited states during the dry and wet season, respectively. During the wet season (January–March), latent heat flux (LE) is greater than sensible heat flux (H), and the daily evapotranspiration variability is mainly related to incoming solar radiation (SW↓; R2 = 0.76, p-value < 0.01), having a daily mean of 3.45 mm. In contrast, the dry season (June–August) is characterized by a greater H with a daily mean evapotranspiration of 0.95 mm. Furthermore, evapotranspiration is significantly tied to the soil moisture variability on daily time scales (R2 = 0.77, p-value < 0.01).

Of the three equations, PT has a good performance in reproducing the daily evapotranspiration variability (R2 > 0.72; p-value < 0.01). In contrast, strong biases are noticed during the dry season mainly because this empirical equation does not account for the soil water content. Thus, our results show that the inclusion of soil water content or a physiological-plant parameter would be necessary to estimate the real evapotranspiration during the dry season.

Introduction

Of the component of the hydrological cycle, evapotranspiration (ET) is the less documented compared to precipitation or runoff essentially due to the difficulties of its measurement. In the surface energy balance, usually ET is the largest term, indicating its extraordinary climatic relevance. The involved mechanisms include the physical aspect of the water change of phase in the surface-atmosphere interface and the plant physiological processes when vegetation is present. Indeed, the latter is still crudely represented in the conceptual models of ET. For a recent review of the state-of-the-art, see Cuxart and Boone (2020).

On a more local impact, for accurate estimations of water balance in a catchment, monitoring ET is crucial for planning adequate crop irrigation programs, (Qiu et al., 2011; Smith, 2000), ensure conservation and management of water resources (Cleugh et al., 2007), and proper planning of hydraulic works (Jhorar et al., 2011). In Peru, where these issues are of extraordinary importance, no direct measurements of ET were available and decisions are usually based on hydrological modeling.

Before the progressive implementation of eddy-covariance (EC) systems, still scarce, the experimental estimation of the actual values of ET was made traditionally with evaporation tanks in meteorological stations, essentially abandoned nowadays, and with weighing lysimeters for agronomical purposes in some research-oriented locations (Girona et al., 2011). Therefore, empirical approaches were needed for the estimation of ET using standard meteorological data. The most popular ones, providing a value for the potential Evapotranspiration in well-watered surface conditions (ETp), are currently the Food and Agriculture Organization (FAO) adaptation (Allen et al., 2006) of the Penman-Monteith equation (Penman, 1948; Monteith, 1965) and a simplification of it, the Priestley-Taylor equation (Priestley and Taylor, 1972), both of which allow the estimation of ETp at the daily and subdaily time step. An important obstacle in using the mentioned equations to calculate daily value of ETp is the partial absent of some meteorological information, and in this case the FAO recommends the use of the Hargreaves equation (Hargreaves and Samani, 1985).

The EC technique has been used to estimate real ET with a high degree of accuracy, as shown in many studies (Matsumoto et al., 2011; Kosugi et al., 2007), and the values that it provides are taken as reference to compare with estimation from equations and satellite measurements (Liu et al., 2013). However, the method presents challenges which may result in some loss of the total flux essentially related to the presence of hectometric circulations (Mauder et al., 2020), and the values show discrepancies with those from lysimeters, especially at the subdaily scale (Hirschi et al., 2017). Therefore, the comparison of the result of empirical methods with EC values should take the uncertainties of the EC systems into account.

In South America the EC technique has been used essentially in the Amazon basin (Shuttleworth, 1988; Costa et al., 2010; Hasler and Avissar, 2007). Juárez et al. (2007) showed that the ET in the southern Amazon region is mainly related to soil water content, while Souza-Filho et al. (2005) found that the ET in the northern Amazon region is strongly associated with variations in net radiation. The use of the technique at high altitudes is relatively recent (Coners et al., 2016), including the equatorial Andes (Carrillo-Rojas et al., 2019; Ochoa-Sánchez et al., 2019, Ochoa-Sánchez et al., 2020).

The Andes extend along western South America from the Patagonia region (55°S) to Venezuela (11°N). In the tropical zone, they are surrounded to the east by the warm and moist Amazon region, while the western slopes are influenced by dry, cool air coming from the eastern Pacific Ocean (Houston and Hartley, 2003). Seasonal precipitation in the southern tropical Andes (Central and southern Peru, Bolivia, and northern Chile) is marked by a wet season during the December–March period and a dry season during the June–August period (Vuille and Keimig, 2004; Garreaud, 2009; Segura et al., 2019; Espinoza et al., 2020, Kumar et al., 2020) This region distinguishes from the area closer to the Equator, characterized by the absence of a marked dry season (Espinoza et al., 2009, Laraque et al., 2007, Segura et al., 2019).

Agriculture has great economic importance in Peru, and several studies have analyzed the water balance to determine the partition of precipitation between runoff and ET (Lavado et al., 2009; Zubieta et al., 2015). The complex topography of the southern tropical Andes induces a great spatial variability of precipitation (Garreaud, 1999; Houston and Hartley, 2003; Garreaud, 2009; Segura et al., 2019; Espinoza et al., 2020) and the low density of meteorological stations makes challenging the analysis of the hydrological cycle in this Andean region. A similar characterization for ET in the region is still pending and this work intends to be a contribution to it. There is a previous study of ET in the Mantaro valley but only in a dry month (Callañaupa, 2016).

So far, only empirical or hydrological model estimates of ET have been provided for the Peruvian Andes (Lavado et al., 2009; Zubieta et al., 2015). Hellström and Mark (2006) calculated ET in the Llanganuco Valley (Central Peruvian Andes) for the wet (December) and dry (June–July) seasons using the BROOK90 model, finding a strong seasonality, with mean ET values of 2.63 mm/day during the wet season and 0.03 mm/day during the dry season.

To tackle the obvious need of ET reference measurements in the central Andes, the Geophysical Institute of Peru (IGP) installed in 2015 an Eddy-Covariance (EC) system at the Huancayo Observatory. This system of EC consisting of a krypton hygrometer and a sonic anemometer, recording temperature, humidity and wind at a frequency of 10 Hz to compute turbulent sensible and latent heat fluxes. Measurements of the radiation budget at the surface and ground heat flux (obtained from upper soil temperature and moisture) are also available. This equipment shall allow to document the surface energy balance in a point representative of the area, including the actual ET.

We describe the area, data and methods in Section 2, the observed annual cycle of ET using the experimental estimations in Section 3, in which the adequacy of the empirical approaches for the site are also discussed. Furthermore, conclusions from the result section are stated in Section 4.

Section snippets

Study area

The Huancayo Observatory of the Geophysical Institute of Peru (12.04° S, 75.32°, 3330 masl) is located in the Mantaro River Basin, in the Junín region of the Central Peruvian Andes (Fig. 1). The main valley is oriented in the NW-SE direction, having a gentle slope of 0.2° between the cities of Jauja and Huancayo, with mountain ranges at both sides reaching altitudes above 5500 masl, and the Cunas tributary to the west of the Mantaro River. The Observatory is located on the western side of the

Wet and dry season at the Huancayo Observatory

The daily precipitation time series for the July 2016–June 2017 period is shown in Fig. 2a. Following previous studies (see introduction), our results show a strong seasonality in the annual cycle of precipitation. Rain events are observed from September to May, with a dry spell in November. The September–November season is catalogued as the onset of the rainy season with precipitation values of 66 mm, while the core of the wet season is observed during the January–March season with

Discussion and conclusion

In this study, we used the eddy covariance technique to analyze seasonal evapotranspiration and energy fluxes in the high central Peruvian Andes for the first time, at the tower installed in the Mantaro Valley. The estimated annual amount of ET for the period July 2016–June 2017 is approximately of 800 mm/year (from Section 3.4), very similar to the amount of precipitation. However, it is important to remember that while precipitation is measured in a single point, the eddy covariance technique

Author statement

Individual contributions for the present work, the contributions were distributed as follows: Conceptualization: Stephany Callañaupa Gutierrez and Hans Segura Cajachagua; methodology: Stephany CallañaupaGutierrez and José Flores Rojas; software: Stephany Callañaupa Gutierrezand Miguel Saavedra Huanca; validation: Stephany Callañaupa Gutierrez, Hans Segura Cajachaguaand Joan Cuxart; formal analysis: Stephany Callañaupa Gutierrezand Miguel Saavedra Huanca; investigation: Stephany Callañaupa

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This work was supported by the project, “Study of the physical processes that control the superficial fluxes of energy and water for the modeling of frosts, intense rains and evapotranspiration in the central Andes of Peru” (400-PNICP-PIBA-2014). The authors thank NASA for the NDVI (MOD13Q1) product and the evapotranspiration (MOD16A2) product. JC thanks IGP for the support to his visit to the Huancayo Observatory. Also, the authors appreciate the work of Luis Suarez and Lucy Giraldez,

References (78)

  • S.M. Liu et al.

    Measurements of evapotranspiration from eddy-covariance systems and large aperture scintillometers in the Hai River Basin, China

    J. Hydrol.

    (2013)
  • Q. Mu et al.

    Improvements to a MODIS global terrestrial evapotranspiration algorithm

    Remote Sens. Environ.

    (2011)
  • P.L. Nagler et al.

    Predicting riparian evapotranspiration from MODIS vegetation indices and meteorological data

    Remote Sens. Environ.

    (2005)
  • P.L. Nagler et al.

    Evapotranspiration on western US rivers estimated using the enhanced vegetation index from MODIS and data from eddy covariance and Bowen ratio flux towers

    Remote Sens. Environ.

    (2005)
  • R. Qiu et al.

    Energy partitioning and evapotranspiration of hot pepper grown in greenhouse with furrow and drip irrigation methods

    Sci. Hortic.

    (2011)
  • M. Saavedra et al.

    Physical controls on frost events in the Central Andes of Peru using in situ observations and energy flux models

    Agric. For. Meteorol.

    (2017)
  • T. Shimizu et al.

    Estimation of annual forest evapotranspiration from a coniferous plantation watershed in Japan (2): comparison of eddy covariance, water budget and sap-flow plus interception loss

    J. Hydrol.

    (2015)
  • M. Smith

    The application of climatic data for planning and management of sustainable rainfed and irrigated crop production

    Agric. For. Meteorol.

    (2000)
  • A. Trepekli et al.

    Seasonal evapotranspiration, energy fluxes and turbulence variance characteristics of a Mediterranean coastal grassland

    Agric. For. Meteorol.

    (2016)
  • T.E. Twine et al.

    Correcting eddy-covariance flux underestimates over a grassland

    Agric. For. Meteorol.

    (2000)
  • K. Wilson et al.

    Energy balance closure at FLUXNET sites

    Agric. For. Meteorol.

    (2002)
  • R. Zubieta et al.

    Impacts of satellite-based precipitation datasets on rainfall–runoff modeling of the Western Amazon basin of Peru and Ecuador

    J. Hydrol.

    (2015)
  • R.G. Allen et al.

    Evapotranspiración del cultivo: guías para la determinación de los requerimientos de agua de los cultivos

    Roma: FAO

    (2006)
  • M. Aubinet et al.

    Estimates of the annual net carbon and water exchange of forests: the EUROFLUX methodology

    Adv. Ecol. Res.

    (2000)
  • S.M. Callañaupa

    Caracterización de la evapotranspiración en los cultivos alrededor del observatorio de Huancayo usando la técnica Eddy Covariance (Tesis para optar el título profesional de Ingeniero Geógrafo)

    (2016)
  • M.H. Costa et al.

    Atmospheric versus vegetation controls of Amazonian tropical rain forest evapotranspiration: are the wet and seasonally dry rain forests any different?

    J. Geophys. Res. Biogeosci.

    (2010)
  • N.C. Cristea et al.

    Revised coefficients for Priestley-Taylor and Makkink-Hansen equations for estimating daily reference evapotranspiration

    J. Hydrol. Eng.

    (2013)
  • J. Cuxart et al.

    Evapotranspiration over land from a boundary-layer meteorology perspective

    Bound.-Layer Meteorol.

    (2020)
  • J. Cuxart et al.

    Evaluation of the surface energy budget equation with experimental data and the ECMWF model in the Ebro Valley

    J. Geophys. Res. Atmos.

    (2015)
  • J. Cuxart et al.

    Estimation of the advection effects induced by surface heterogeneities in the surface energy budget

    Atmos. Chem. Phys.

    (2016)
  • J.C. Espinoza et al.

    Hydroclimate of the Andes part I: main climatic features

    Front. Earth Sci.

    (2020)
  • J.C. Espinoza et al.

    Spatio-temporal rainfall variability in the Amazon basin countries (Brazil, Peru, Bolivia, Colombia, and Ecuador)

    Int. J. Climatol.

    (2009)
  • J.L. Flores-Rojas et al.

    Seasonal and diurnal cycles of surface boundary layer and energy balance in the Central Andes of Perú, Mantaro Valley

    Atmosphere

    (2019)
  • T. Foken

    The energy balance closure problem: an overview

    Ecol. Appl.

    (2008)
  • T. Foken et al.

    Micrometeorology

    (2008)
  • T. Foken et al.

    Energy balance closure for the LITFASS-2003 experiment

    Theor. Appl. Climatol.

    (2010)
  • O. Garay et al.

    Primera aproximación para la identificación de los diferentes tipos de suelo agrícola en el valle del río Mantaro

    Editado por el IGP, Lima

    (2010)
  • R. Garreaud

    Multiscale analysis of the summertime precipitation over the Central Andes

    Mon. Weather Rev.

    (1999)
  • R.D. Garreaud

    The Andes climate and weather

    Adv. Geosci.

    (2009)
  • View full text