Relationships between single tree canopy and grass net radiations

https://doi.org/10.1016/j.agrformet.2006.11.003Get rights and content

Abstract

It is reported a simple approach to transform daily values of grass net (all-wave) radiation (Rn, MJ m−2 day−1), as measured over standard grass surface at meteorological stations, into whole tree canopy net radiation (A, MJ tree−1 day−1). The revolving Whirligig device [McNaughton, K.G., Green, S.R., Black, T.A., Tynam, B.R., Edwards, W.R.N., 1992. Direct measurement of net radiation and photosynthetically active radiation absorbed by a single tree. Agric. For. Meteorol. 62, 87–107] describing a sphere about the tree measured A in five trees of different species (walnut, dwarf apple, normal apple, olives and citrus), with leaf area LA varying from 8.65 to 40 m2. For each tree, A and Rn were linearly related (A = bRn), as previously reported elsewhere, but it was found that the slope of such regression was also a linear function of LA or, b = 0.303 (±0.032) LA. Consequently, the hypothesis that total daily tree canopy net radiation per unit leaf area is linearly related to grass net radiation could not be rejected after 86 days of measurements in five locations, and the empirical relationship is A = 0.303 (±0.032) RnLA (R2 = 0.9306).

Introduction

Orchard is an assemblage of fruit trees whose leaf area does not evenly cover the ground surface. The discontinuous spatial distribution of trees with inter-plant spacing of about the height of the plants increases the difficulty in characterizing the net radiation regime of orchards and its use in energy balance studies. Measurements of tree canopy net radiation do not follow a standard experimental setup and they can be grouped into three classes of design: (i) above; (ii) above-and-below; and (iii) around the tree or hedgerow canopy.

The above canopy measurements group includes experiments with one, two or more sensors. Single sensor experiments has it positioned either well above the canopy row (about two times the plant height), to include the tree crown and the inter-row space in the view angle (Ben-Asher and Sammis, 1978, Trambouze et al., 1998), or immediately above the tree (Thorpe, 1978, Judd et al., 1986, Caspari et al., 1993, Azevedo et al., 2003, Green et al., 2003a, Green et al., 2003b, Pereira et al., 2003, Williams et al., 2004). Experiments with two (Testi et al., 2004) or four (Rana et al., 2005) sensors have them evenly positioned above the trees row and the inter-row.

The above-and-below measurements group can have two sensors above and two below sensing the row and the inter-row space, positions of maximum and minimum ground cover (Heilman et al., 1996, Villalobos et al., 2000), or only one sensor well above the row and four evenly distributed below the canopy (Daamen et al., 1999). Yunusa et al. (2004) combined measurements of all-wave radiation well above a vineyard row with shortwave radiation above and a series of three linear solarimeters below a vine.

The around-the-canopy group uses an array of eight net radiometers arranged at equally spaced fixed positions around an imaginary volume enclosing the tree foliage. Static linear sensors (0.9 m long) were used around apple trees in hedgerows (Thorpe, 1978), and windbreak trees (Smith et al., 1997). A dynamic system (named Whirligig) of eight regular sensors rotating in a circle around the tree was used for isolated trees (McNaughton et al., 1992, Green, 1993, Green and McNaughton, 1997, Pereira et al., 2001, Green et al., 2002, Green et al., 2003a, Green et al., 2003b).

Besides the operational difficulties with any measurement of tree net radiation the results represent only the sampled trees and rarely can be taken as representative of others with different canopy sizes (leaf areas) in the same orchard. Thorpe (1978) found that the around-the-canopy net radiation per unit leaf area represented about 50% of the net radiation of the single sensor well above the tree row. McNaughton et al. (1992) reported that daily integrated net radiation (A) absorbed by the 10.6 m2 of leaf area (LA) of a false acacia tree (Robinia pseudoacacia L.) was nearly equal to the net radiation absorbed by 8 m2 of grass (Rn). Thus, tree net radiation per unit leaf area was 0.75 (=8/10.6) Rn. Pereira et al. (2001) found that a citrus tree (Citrus latifolia Tan.) with LA = 40 m2 absorbed the equivalent of 12.66 m2 of grass surface, or A = 0.32 (±0.02) Rn on a leaf area basis. Inverting the Penman–Monteith equation, using the sap flow as inputs, Zhang et al. (1997) found that the net radiation per unit leaf area of poplar trees (Populus spp.) was equal to 0.484 (±0.046) of Rn measured 7 km away.

This paper reports the results of a study of around-the-tree canopy measurements of net radiation compared with net radiation of standard weather station to determine an empirical transfer function including the tree leaf area. The hypothesis that tree net radiation per unit leaf area is a unique function of grass net radiation was tested and accepted.

Section snippets

Material and methods

Four data sets are from experiments in New Zealand. One set is from a 3.5 m tall and isolated walnut tree (Jugans spp.) at Massey University Research Orchard near Palmerston North (40.2°S, 175.4°E), during 9 days of January 1992 (see Green, 1993). Tree leaf area (LA) was estimated as 26.4 m2 by measuring 10% of the total number of leaves.

Another data set is from a 4-year-old orchard olive tree (Olea europaea L. cv. Barnea) at the Ponder Estate (41.5°S, 173.9°E) in Marlborough between January and

Results and discussion

The results reported in this paper are based on 86 daily integrations of Whirligig measurements of tree net radiations (A, MJ tree−1 day−1) around five trees of four different species and canopy sizes, and the corresponding grass net radiations (Rn, MJ m−2 day−1), in five different locations. As indicated by McNaughton et al. (1992) it was possible to develop the simple linear relationship A = bRn. The small spread of the points around the regression line indicates the goodness of each fit (Fig. 1);

Conclusions

Considering the paucity of detailed measurements of tree net radiation there is a need for a transfer function to use the more readily available grass net radiation of standard weather stations. Results for five trees of four different species with canopy leaf area up to 40 m2, in five locations support the hypothesis that daily integrated tree net radiation per unit leaf area is a unique linear function of grass net radiation. This empirical approach compensates for not having to use any

Acknowledgement

A.R.P. and N.A.V.N. were partially supported by the Brazilian National Research Council.

References (24)

Cited by (13)

  • A combined tracer/evapotranspiration model approach estimates plant water uptake in native and non-native shrub-steppe communities

    2015, Journal of Arid Environments
    Citation Excerpt :

    However, tracer studies do not provide good estimates of the amount of water uptake because water uptake is a function of soil water availability and transpiration, and therefore water uptake needs to be estimated using a different approach. While many approaches have been used to estimate ET (Garcia et al., 2013), the Penman-Monteith (PM) model remains one of the most commonly used and most detailed for field-scale ET estimation (Allen and Pereira, 2009; Allen et al., 1998; Bond-Lamberty et al., 2011; Fisher et al., 2011; Pereira et al., 2007; Tian et al., 2011; Wilske et al., 2010). The PM model is particularly appealing for use in species-specific modeling because it includes parameters for stomatal conductance and aerodynamic resistance.

  • Relationship between tree row LIDAR-volume and leaf area density for fruit orchards and vineyards obtained with a LIDAR 3D Dynamic Measurement System

    2013, Agricultural and Forest Meteorology
    Citation Excerpt :

    This is presently defined as half the leaf area per unit of ground surface (Chen and Black, 1992). The leaf area is an extremely important parameter because it is strongly related to processes such as evapotranspiration, radiation interception, and CO2 fixation (Testi et al., 2004; Cohen et al., 2005; Williams and Ayars, 2005; Goodwin et al., 2006; Orgaz et al., 2006; Pereira et al., 2007; Pereira and Green, 2007; Lopez-Lozano et al., 2011). The leaf area can be calculated by direct measurement which involves a lot of time and expense or it can be indirectly estimated using remote sensors and without any physical contact with the leaves (Zheng and Moskal, 2009; Rosell and Sanz, 2012).

  • A general algorithm for automated scheduling of drip irrigation in tree crops

    2012, Computers and Electronics in Agriculture
    Citation Excerpt :

    This is especially relevant in tree crops where factors such as tree size, plantation frame, training system and crop load may affect water requirements. To deal with some of these factors, Kc can be refined according to traits such as ground cover and plant height (Pereira et al., 2007). Indirect measurements like light interception by the canopy can also be used (Girona et al., 2011).

  • Automated irrigation of apple trees based on measurements of light interception by the canopy

    2011, Biosystems Engineering
    Citation Excerpt :

    As the seasonal pattern of air temperature lagged behind that of Rs, the expected effect of a temperature-dependent KRI would be to increase irrigation in late summer and decrease it in spring and autumn. Air temperature and absorption of solar radiation by canopies are related to tree water needs through biophysical relationships such as the Priestley–Taylor model for estimating evapotranspiration (Pereira, Green, & Nova, 2007; Priestley & Taylor, 1972), suggesting that there is a sound background for a general biophysical effect of temperature on KRI in most species. Further work will be necessary to confirm the effectiveness of expressing KRI as a function of weather variables and to assess whether a common solution can be developed for different horticultural crops or whether case-specific calibrations are required.

View all citing articles on Scopus
View full text