Analysis of global and net radiation fluxes in relation to surface albedo at DACCIWA site in Ile-Ife, southwest Nigeria

Measurements of global solar radiation (RG) and net radiation (RN) fluxes were made above the grass-covered surface at DACCIWA site, Ile-Ife, southwest Nigeria, from 2017 to 2019. The datasets were obtained from a four-component net radiometer (model NR01, Hukseflux Thermal Sensors B.V., the Netherlands). Observations were made and separated for cases of clear sky and cloudy conditions during the measurement period. The results showed considerable fluctuations for both radiation fluxes. For clear sky conditions, the magnitudes of RG and RN were higher than those observed for cloudy conditions due to attenuation. For the period of observation, LWup fluxes were higher in the dry season than in the wet season due to dry and hotter surface conditions while LWd dominated the wet season at the site. The highest radiation values occurred in 2018, while the lowest were observed in 2017. Daily surface albedo (α) ranged from 0.16 to 0.22. Empirical relationships obtained for RG and RN are RN = 0.754 RG – 17.4 Wm−2 and RN = 0.657 RG – 32.7 Wm−2 for wet and dry seasons, respectively. Based on the empirical relationships, daily RN and RG can be obtained when measurements like these are not available. Linear relationships between RN and RG show that for all days (cloudy and clear sky conditions), daily average RN is about 0.75 RG, and about 0.60 RG for clear sky conditions at the location. However, the results reflect the characteristics of the site and can be applied for gap-filling or careful extension of the measurement period.


Introduction
Land surface atmosphere exchanges which result into various atmospheric processes and are used in describing the weather and climate of a location or region depend to a large degree on the Earth's Radiation Budget System (ERBS). The importance of ERBS therefore cannot be overemphasized as it contributes significantly to the general circulation system and enhances the modulation of the earth's climate system (García et al. 2017;Mueller et al. 2009;Liang et al. 2010).
The sun is the primary and major source of radiant energy to the earth's surface. The amount of energy emitted by the sun to the earth's surface depends on factors such as land use, nature and area of surfaces, vegetation and cloud cover, and the presence of aerosols and particulate matters for its spatiotemporal distributions (Dewitte and Clerbaux 2017;Kondratyev 1969;Nage 2018;Ndulue et al. 2019;Wang and Liang 2009). The spatial distribution and temporal variation of such radiant energy across the continents can either be in the form of global or net radiation (Ogunjobi et al. 2002;Jiang et al. 2019a, b;Wild et al. 2017).
Global radiation (R G ) flux consists of direct solar radiation flux and diffuse sky radiation flux in the wavelength range 0.1 μm and 4.0 μm. It can be measured directly using a pyranometer or estimated from empirical models developed on the basis of measured values (Balogun et al. 2003;Nage 2018). It is given in watts per square meter (W. m −2 ). Net radiation (R N ) flux is the total energy flux available at the ground surface and can be measured directly or estimated from some other meteorological variables (Selirio et al. 1971;Amarakoon and Chen 1999;Samani et al. 2005Samani et al. , 2007Jegede et al. 2006;Ayoola et al. 2014). R N can be deduced from the earth's energy budget equation, i.e., summation of shortwave and longwave radiation parameters (Polavarapu 1970;Giambelluca et al. 1997;Kounouhéwa et al. 2013), and is given by where R N is the net radiation parameter, S ↓ and S ↑ are the incoming and outgoing shortwave radiation parameters, and L ↓ and L ↑ are the incoming and outgoing longwave radiation (1) R N = S ↓ − S ↑ + L ↓ − L ↑ parameters. R N is related to turbulent heat fluxes by the surface energy balance equation: where H s and H L are the sensible and latent heat fluxes, and G and ∆S are the ground heat flux and storage term, respectively.
Equation (2), when simplified further, becomes The available energy is the sum of the turbulent fluxes (H s + H L ), and (G + ∆S) represents the energy storage in the system (Mamadou et al. 2014;Wohlfahrt et al. 2017;Majozi et al. 2017). Other terms in Eq. (2) can be obtained directly from routine measurements. R N can be measured with the aid of net radiometers which are either component type or radiation energy balance system (REBS) depending on sensitivity and accuracy. It is given in watts per square meter (Wm −2 ).
In spite of the fact that dynamics and diurnal variation of radiation fluxes have significant influences on the energy and hydrological cycles of West African Monsoon (WAM), accurate and comprehensive knowledge of their variability in association with climate change within the region is still inadequate due to lack of sufficient data and detailed studies on the subject matter (Alados et al. 2003;Kounouhéwa et al. 2013;Nage 2018). However, such knowledge is useful for predictions in agricultural systems and site selection for photovoltaic power generation, among others (Nage 2018;Jiang and Lu 2019;Jiang et al. 2019a, b).
Many observations on global and net radiation fluxes based on in-situ measurements and models' application have been reported in the temperate region, but only fewer studies have been in the  tropical region. Therefore, more studies need to be conducted in the tropical region in order to relate the dynamics and variability of radiation to climate change in the area. This will help in providing relevant solutions to reduce the devastating effects of environmental degradation, such as soil erosion, flood disasters, pollution, and other related unforeseen occurrences ravaging the region. Li and Barnes (1980) revealed a linear relationship between net and global radiation fluxes in a radiation balance study over Lake Albert in South Australia. Linear relationships were also established between net and global radiation fluxes in a study on net radiation measurements and its components over a natural prairie grass in the semi-arid climate of Baghdad by Riahi et al. (2003). A study by Hu et al. (2012) at Lhasa and Haibei in the Tibetan Plateau showed a good linear correlation between net radiation (R n ) and broadband solar radiation (R s ). Samani et al. (2007) estimated daily net radiation over a well-watered grass canopy at Chamberino Weather Station in New Mexico for three different methods with an average error from 0.01 to 0.3 and a standard error from 1.06 to 5.34 MJ. m −2 . day −1 . Investigation of dynamics and diurnal variation of net radiation over agricultural crops at Nalohou village in northern Benin by Kounouhéwa et al. (2013) showed larger net longwave radiation values in the dry months than in the wet months due to prominent strong clouds. A study on diurnal and seasonal variation of incoming solar radiation flux by Soneye et al. (2019) in Ile-Ife, Nigeria, attributed large differences in the values of radiation fluxes between wet and dry seasons to convective cloud drift and aerosol loading effects, respectively. The present study has analyzed hourly averaged data sets of global solar and net radiation fluxes acquired for three consecutive years (January 2017-December 2019) at a meteorological station (DACCIWA site) in Ile-Ife, southwest Nigeria, to investigate the diurnal variation of radiation fluxes. The study will establish empirical relationships to estimate global solar and net radiation fluxes based on the already available meteorological data in the area.

Methodology
The study was conducted at a meteorological station (OAU Met-station) located in the Teaching and Research Farm of Obafemi Awolowo University (OAU), Ile-Ife, southwest From 2004 to date, data acquired from the station has served the university community, research institutes, and industries which are in need of such. The site is situated in the "tropical wet and dry" climate zone of West Africa with an average annual rainfall of 1000 to 1500 mm. The region has two alternating seasons classified as wet and dry, as determined by the movement of the inter-tropical discontinuity line (ITD). The wet season spans from April to October and the dry season from late November to February. As part of the peculiarities of the dry season, the supply of water is limited and evapotranspiration rates are higher than in the wet season. The dry season is characterized with harmattan dust which has found its origin in the subtropical highs in the Sahara deserts. Convective activities and frequent rainfall characterize the wet season, and the surface airflow (maritime origin) which flows over the Gulf of Guinea is usually warm and moist (Charney 1975  Kipp & Zonen B. V., Netherlands for comparison purpose) on a 1.7-m mast situated within the OAU meteorological station for routine measurements of radiation components and net all-wave radiation. NR01 has two pyranometers (model SR01) and two pyrgeometers (model IR01) which are backfacing for incoming and outgoing shortwave and longwave radiation components, respectively. Studies have shown that measurements of radiation taken by NR01 are much more accurate than those of NR-LITE, a radiation energy balance system (REBS), especially for nighttime scenarios when the values are low (Foken 2008;Ayoola et al. 2014). Additionally, NR01 gives more detailed information than NR-LITE. It is robust and only requires limited maintenance. The sensitivity of each component sensor of NR01 is provided in Table 1. Both NR01 and NR-LITE radiometers were connected to a CR1000 datalogger acquired from Campbell Scientific. In order to achieve better accuracy, differential input channels connection was made for voltage measurements.
Data acquisition software was also provided by Campbell Scientific (Campbell Scientific Inc. 2010). All raw data were sampled at 10-s intervals and stored as 1-min average values. Quality assurance and quality control (QA/QC) test was carried out on the datasets to get rid of spikes and other bad data values. Continuous measurements of radiation at the study location have been ongoing since 2016, but the datasets for this study cover the period from January 2017 to December 2019 (Fig. 2).

Diurnal variation of global and net radiation fluxes at the study location
The diurnal variations of global and net radiation fluxes as observed at OAU Met-station, Teaching and Research Farm,

Longwave radiation components
The observed longwave radiation components exhibited daily patterns, as shown in Fig. 7a, b, respectively. In Fig. 7a, the values of longwave downwelling radiation (LW d ) were low in dry season, especially in December and January, as compared to other months for the period of measurements. Radiation values fell below 400.0 Wm −2 at nighttime. This is due to high turbid conditions and the presence of aerosols and particulate matters attenuating radiation from the atmosphere before reaching the surface.    Similarly, there was a close relationship between global and net radiation fluxes for the dry season, as indicated by the scatter plots. From Eqs. (5) and (6), it can be inferred that for the wet season, average net radiation is about 75 % of global radiation and about 60 % for the dry season. These values are higher when compared to the previous values earlier reported for West African stations and fourteen other stations in the world by Davies (1967), as shown in Table 2. Table 3 gives the slope (b), intercept (a), and correlation coefficients for the regression of R N on R G for the period of observations. For wet and dry seasons in Nigeria, linear correlation coefficients, r 2 is more than 0.9 in both cases.

Summary and conclusions
Hourly averaged datasets of radiation fluxes acquired for three consecutive years (January 2017-December 2019) at DACCIWA site, situated on the main campus of Obafemi Awolowo University in Ile-Ife, southwest Nigeria, were analyzed in this study to derive empirical relationships for estimating global solar and net radiation fluxes. The radiation fluxes showed considerable fluctuations from sunrise to sunset influencing surface albedo. α values ranged from 0.17±0.03 to 0.24±0.02, 0.17±0.01 to 0.23±0.02, and 0.16±0.05 to 0.22±0.02 for 2017, 2018, and 2019, respectively, at the location. Higher α values in the early morning are adduced to the contribution of high reflectance at a high zenith angle. Low values of α recorded for the wet season could be attributed to the presence of strong convective clouds and frequent rain showers. Higher albedo values were recorded in the dry season due to surface dryness and high reflectance at a high zenith angle.
Large values of radiation fluxes observed in the dry season were adduced to surface dryness and clear sky conditions, while low radiation values in the wet season were attributed to the presence of strong cloud cover and frequent  Author contribution All authors contribute to the study's conception and design. Material preparation, data collection, and analysis were performed by Dr. Adewale Iyiola Ajao. The first draft of the manuscript was written by Dr. Adewale Iyiola Ajao, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript before the final submission was made.
Funding The Nigeria Micrometeorological Experiment (NIMEX) project led by Prof. Oluwagbemiga Jegede had been responsible for the running and maintenance of the Obafemi Awolowo University meteorological station (OAU Met-station) before the start of DAC-CIWA project in 2014. Partial financial support was received from the European Union Seventh Framework Programme (FP7/2007(FP7/ -2013 under grant agreement number 603502 for the upgrade of the existing meteorological station to a supersite and reimbursements for attending conferences during the DACCIWA project.

Data availability
The data for this study will be made available to any individual or organization on reasonable requests via a.ajao@oauife.edu.ng or iyiolamercy2005@yahoo.com. The data were acquired at the DACCIWA site after the intensive observational studies (IOPs) had ended in 2016, but data before this period are available on SEDOO (BAOBAB) database.

Declarations
Ethics approval The manuscript is a result of the research ongoing at the Obafemi Awolowo University Ile-Ife, southwest Nigeria. The work reported here is in its original form and has not been submitted or published elsewhere. Proper acknowledgements have been given to all the works consulted in the course of its preparation.
Consent to participate All authors are core members of the Atmospheric Physics Research Group (APRG) at the Department of Physics and Engineering Physics and the Center for Energy Research and Development (CERD) in Obafemi Awolowo University (OAU), Ile-Ife, southwest Nigeria. The group has been responsible for the maintenance and running of the existing meteorological station in OAU, where the data for the study was acquired. No human or animal participation was directly involved in the research.

Consent for publication
All authors agreed with the content of the final copy of the manuscript, and appropriate authorities were consulted and acknowledged before the work was submitted to the Journal.

Conflict of interest
The authors declare no competing interests.