MODELING INTERCEPTED SOLAR RADIATION FOR TWO DIFFERENT TYPES OF VEGETATION (RAIN FOREST OF REBIO-JARU-RO AND MANGROOVE -PA) MODELING INTERCEPTED SOLAR RADIATION FOR TWO DIFFERENT TYPES OF VEGETATION (RAIN FOREST OF REBIO-JARU-RO AND MANGROOVE -PA) Moura 1, R.G.; Tota 1, J.; Manzi 1, A. O.; Gu 2, L. 1 CPTEC - INPE, Cachoeira Paulista-SP, Brasil 2 Uc Berkeley Biometeorology Lab, DESPM, Berkeley, CA , US ABSTRACT Measurements of solar radiation were made over a terra firme forest, at the Biological Reserve of Jaru-RO, as part of the LBA and over a mangrove area in the City of Bragança-PA, as part of the DESMATA project. Data of solar radiation flux were collected with Kipp & Zonen pyranometers, and photosynthetically active radiation (PAR) flux, with LICOR quantum sensor, in the top of the towers installed at each site, and in the ground of the forests. In the first site, information regarding the leaf area index (LAI) were provided by a previous work (Moura, 2001). The results show that, on average, the fraction of solar radiation and PAR fluxes that reach the ground are smaller at the rain forest than at the mangrove. In this work the observations of solar radiation and PAR at the top of the forests of terra-firme and mangroove are used to evaluate the performance of the radiative transfer model proposed initially by Sellers (1985) and modified by Gu (1999). The results of the model showed that it is capable to reproduce the solar radiation fluxes that reach the ground at both sites quite well, when forced with the average values at the top of the towers. Nonetheless, the model estimates of PAR at the forest ground compare with the reference values but not at the mangroove site. INTRODUCTION The knowledge of the interaction of the solar radiation with the vegetation is essential for the understanding of the processes of vegetation physiology, biomass productivity and turbulent exchanges of energy and mass with the forests. The attenuation of the solar radiation in canopy is a physical phenomenon that depend on the type and the characteristics of the vegetal cover, specially the space distribution of the elements of the studied vegetation, which presents an irregular concentration of available energy in different stratus of the vegetation with consequences to the vertical profile of the radiation fluxes. Because of this, a task of difficult execution in the last years, some alternatives, as the use of mathematical models, have been tested, using leaf area index and profile of the radiation in forest environment data. Studies of this nature had been carried through in vegetal covers of medium and dense forests in the Amazônia, consisting on the determination of the LAI and the vertical distribution of the function a(z) (leaf area density) using indirect methods based in the penetration of solar radiation in the canopy (Honzák et al., 1996; Wandelli and Marques Filho, 1999), providing important alternatives to the destructive and laborious methods of direct measure of all existing leaf area in a chosen amostral area (McWilliam et al., 1993). On the other hand, Moura et al., 2000, analyzed the attenuation of the radiation in forest environment using a simplified version of the model of radiative transference modified by Gu et al. (1999). In this context, the present study has the following objectives: to present and to analyze the data of solar radiation, in two distinct areas of forest, observed during the projects The Large Scale Biosphere-Atmosphere Experiment in Amazonia-LBA”, and “Impacto do Desmatamento Junto ao Litoral Atlântico da Amazônia-DESMATA”. The solar radiation will be evaluated above and below the forest canopies, focusing its daily variability and the factors that determine its attenuation in the vertical direction; comparing with the values obtained from the model of radiative transfer modified by Gu et al. (1999), with the purpose of estimating the values of incident solar radiation that reaches the surface close to the forest ground. DATA USED AND OBSERVED ASPECTS Data of solar radiation and photosynthetically active radiation (PAR), at the top of two different types of vegetation cover, on period 01 to 15 of June of 2002, were used in this work. The first one is a tropical rain forest, located in the Biological Reserve of Jaru (Rebio-Jaru-RO), (Lat.:10° 05’ S; Long.: 61° 56’ W). This forest site presents a vegetation with characteristics of terra-firme forest, with average height of the canopy of approximately 35 m with some emergent trees reaching 45 m. The second site is a mangroove forest, located in the bragantina zone (Bragança-Pará), (Lat.: 00° 50’ 31” S; Long.: 46° 38’ 56” W). In this forest the predominant species are Rhizophora Mougle, Avicennia Germinaus and Laguncularia Racemoso, with average canopy height around 18 m. Additional information about the instruments used in this study may be found at table 1. Data of PAR at the ground of the mangroove were directly measured by a quantum sensor and the data of solar radiation at mangroove and solar radiation and PAR at forest ground were indirectly obtained using the extinction fractions measured by Leitão (1994), on an area presumably similar to the mangroove, and Moura (2001) on the very same rain forest site (Rebio-Jaru). METHODOLOGY From the collected data of solar radiation and photosynthetically active radiation of (PAR), in the top of the towers, on each different types of vegetation cover, and the photosynthetically active radiation (PAR) data collected only on the surface closed ground of mangroove the mean diurnal behavior for each component in the period was analyzed, in each one of the previously cited micrometeorological experiments. After that, the hourly average values of radiation in the top of the towers, together with other parameters and the value of the average of leaf area index LAI of the tropical rain forest presented by Moura (2001) were inserted in the model of radiative transference developed by Sellers (1985) and modified by Gu et al. (1999), which after adapted for each site, may be used to estimate the values of incident radiation close to the ground of different types of vegetation cover. This could be done for both sites, as the percentile mean values of the solar radiation that reach the grond of the tropical rain forest are known (Moura, 2001), and the average values of photosynthetically active radiation in mangroove, next to the surface, were directly measured on the period. With this, we will have the possibility to evaluate the degree of representation of the estimates of the radiation close to the surface, presented by model in study, when forced with measured average values of radiation in the top of each one micrometeorological towers. CONCLUSIONS The model modified for Lianhong Gu used in this study, after adapted, was capable to reproduce quite well the daily behavior of the solar radiation and PAR fluxes that reach the ground of the tropical rain forest, when forced with the mean values of incident solar radiation and PAR measured in the top of the tower. However, for the mangroove site, considering the adaptations made in the model and the use of the radiation fluxes measured at the top of the tower, the results were better only for the solar radiation, while for PAR, the model couldn’t reproduce the observations. ACKNOWLEDGEMENTS The first author thanks to the project “Impacto do Desmatamento Junto ao Litoral Atlântico da Amazônia-DESMATA” and CARBONCYCLE/LBA, for providing the data of solar radiation and PAR at the mangroove site located in Bragança-PA and at the forest site, located in Ji-Paraná-RO, respectively. REFERENCES Gu, L.; Shugart, H. N.; Fuentes, J. D.; Black, T. A.; Shewchuk, S. R. Micrometeorology, biophysical exchanges and NEE decomposition in a two- story boreal forest-development and test of an integrated model. Agricultural and Forest Meteorology, v.94, p , Honzák, M.; Lucas, R. M.; Amaral. I.; Curran, P. J.; Foody, G. M.; Amaral, S., Estimation of the leaf area index and total biomass of tropical regenerating forests: comparison of methodologies, In: Gash, J.H.C.; Nobre, C.A.; Roberts, J.M.; Victoria, R. L.(Eds.) Amazonian Deforestation and Climate. Institute of Hydrology, UK. p , Leitão, M. M. V. B. R. Balanço de radiação em três ecossistemas da Floresta Amazônica: campina, campinarana e mata densa. São José dos Campos. 135p.(INPE-5587-TDI/549). (Tese de Doutorado em Meteorologia) – Instituto Nacional de Pesquisas Espaciais, McWilliam, A. L. C.; Roberts, J. M.; Cabral, O. M. R.; Leitão, M. V. B. R.; Costa, A. C. L.; Maitelli, G. T.; Zamparoni, C. A. G. P. Leaf area index and above ground biomass of terra firme rain forest and adjacent clearings in Amazonia. Functional Ecology, 7: , Moura, R. G. Estudo das Radiações Solar e Terrestre acima e dentro de uma Floresta Tropical Úmida. São José dos Campos. (INPE-XXXX- TDI/XXX) (Dissertação de Mestrado em Meteorologia) – Instituto Nacional de Pesquisas Espaciais, no prelo, Moura, R.G., Tota, J., Manzi, A. O., Gu, L. Medidas e Modelagem da Radiação Solar Interceptada pela Vegetação durante a Estação Chuvosa na Floresta da Rebio-jaru-Ro, Primeira Conferência Científica Internacional do LBA, Belém-PA, junho, Wandelli, E. V.; Marques Filho, A. O. Medidas de radiação solar e índice de área foliar de coberturas vegetais. Acta Amazonica, v. 29, n.1, p , RESULTS AND DISCUSSION The figure (1) shows the behavior of solar radiation observed and modeled in the ground of tropical rain forest. The analyses had shown that generally the solar radiation is well reproduced, however, specially on clear days the amplitude of modeled solar radiation is always over estimated near noon. Similar results were founds by Moura et al. (2000). On the other hand, on cloudy days the model reproduces well solar radiation transfer near noon. The figure (2) shows the behavior of photosynthetically active radiation observed and modeled in the ground of tropical rain forest. The results for clear days are very similar to the ones for the solar radiation. But for cloudy days the model under estimates the PAR that reaches the ground.INSTRUMENTMEASUREMENTUNITMODELPyranometer Solar radiation Wm -2 Radiometer CM11 (Kipp & Zonen, Delf, Holland) Sensor Quantum PAR Wm -2 or  molm -2 S - 2 Sensores Quantum (Licor, Nebraska, USA) Table - 1 General Specifications of the used instruments to measure the solar radiation and PAR at the two sites. Figure 1 - Daily behavior of solar radiation observed (red) and modeled (blue) in the ground of tropical rain forest. Figure 2 - Daily behavior of photosynthetically active radiation observed (red) and modeled (blue) in the ground of tropical rain forest. Figure 3 - Daily behavior of solar radiation observed (red) and modeled (blue) in the ground of mangroove. Figure 4 - Daily behavior of photosynthetically active radiation observed (red) and modeled (blue) in the ground of forest mangrove. The figure (3) shows the behavior of solar radiation observed and modeled in the ground of mangrove. The analyses show that generally the solar radiation is well reproduced, even better than the results found for the forest, reforcing the hypothesis of similarity between the vegetation cover campina and the mangroove, respecting the penetration of solar radiation. On the other hand, as presented in figure (4), which shows the behavior of photosynthetically active radiation observed and modeled in the ground of mangrove, the model, even following quite well the variability of the PAR signal, couldn’t repeat the same performance, on average underestimating the PAR, unrespect to cloud cover.