Wireless sensor network to identify the reduction of meteorological gradients in greenhouse in subtropical conditions Thais

Spatial and temporal monitoring of temperature and relative humidity is essential for greenhouse management, therefore, wireless sensor networks (WSN) can offer crucial advantages. The objective of this work was to use a WSN to characterize and map the horizontal and vertical variability of air temperature and relative humidity inside a greenhouse using five different configurations. The configurations were based on combinations between the following actuating mechanisms: i) mechanical ventilation (by two exhaust fans); ii) natural ventilation (through the roof vent openings); iii) shading through the use of thermo-reflective screen. The WSN was designed with 45 spatially distributed measuring points, and the air temperature and relative humidity were recorded automatically every 30 seconds, for ten consecutive days, for each configuration. Our results show that the horizontal and vertical homogeneity of the meteorological elements depends on the actuating mechanism used in the greenhouse. Mechanical ventilation approximated the temperature and relative humidity of the indoor and outdoor air, with a homogeneous horizontal distribution throughout the environment. Opening the roof vent reduced vertical gradients of temperature and relative humidity. Our observations also showed that the combination of the use of roof vent openings with mechanical ventilation is an effective way to achieve horizontal homogeneity of meteorological elements. Introduction Knowing the behaviour of meteorological elements, such as temperature and relative humidity in a greenhouse, is essential to exploit the benefits of protected cultivation. As the air moves inside the greenhouse, gradients of temperature and relative humidity are formed throughout the structure (Chen, 2003). These gradients can result in a heterogeneous microclimate, which may be undesirable for crops (Teitel et al., 2010) and for workers (García-Ruiz et al., 2018; López -Martínez et al., 2018). The wireless sensor networks (WSN) offer several advantages, both for monitoring the greenhouse microclimate, mapping the heterogeneity of meteorological elements, or controlling the actuation mechanisms. However, the location of sensors inside the greenhouse is essential, as their measurements can be used to control all actuation mechanisms. Also, it is possible to obtain different values at different points in the internal microclimate over time (Narasimhan et al., 2007). The gradients of meteorological elements are related to several factors such as solar radiation, mass transfer by convection, and dynamics of air movement due to differences in internal and external temperature. The internal conditions of the greenhouse microclimate can affect crops, with an unequal growth of plants (Teitel et al., 2010), and the safety and productivity of workers. Therefore, measurements of thermal heterogeneity along the vertical and horizontal axes are taken to assess working conditions (García-Ruiz et al., 2018; López -Martínez et al., 2018). In literature, several studies reported different results regarding the distribution of temperature and relative humidity, which may be ascribed, among other reasons, to discrepancies between crops and greenhouse configurations, with different microclimate conditions and technological mechanisms, such as ventilation and cooling systems. Several studies evaluated the horizontal distribution in environments with mechanical ventilation systems (Balendonck et al., 2014), natural ventilation (Ferentinos et al., 2017; García-Ruiz et al., 2018) and natural ventilation associated with mechanical ventilation (Kittas et al., 2012; López et al., 2013). As to the vertical distribution, other studies also evaluated greenhouse microclimates with different systems, such as natural ventilation (Suay et al., 2008), or mechanical ventilation and evaporative cooling (Kutta and Hubbart, 2014; Zorzeto and Leal, 2017). The position of the greenhouse sensors can reflect different values at different points in the internal microclimate over time (Narasimhan et al., 2007). Patterns of heterogeneity were found as a function of the incidence of sunlight and time of day, while the majority of homogeneous days was correlated with cloudy days (García-Ruiz et al., 2018). In a natural ventilated greenhouse, in Almería, the authors found air temperature heterogeneity mainly in the central hours of the day and in the horizontal dimension rather than vertically (measured from 0.23 m to 1.56 m from the floor). Correspondence: Thais Queiroz Zorzeto Cesar, School of Agricultural Engineering (FEAGRI), University of Campinas (UNICAMP), Brazil. E-mail: thaisqzc@unicamp.br