10.9 IMPACTS OF CLOUDS ON GPP AND RESPIRATION OF CONTRASTING TERRESTRIAL

Lianhong Gu* and Dennis D. Baldocchi, University of California, Berkeley Shashi B. Verma, University of Nebraska, Lincoln T. A. Black, University of British Columbia, Vancouver Timo Vesala, University of Helsinki, Helsinki

1. INTRODUCTION* (http://public.ornl.gov/FLUXNET/). Here we report The presence or absence of clouds in a region is a results of analyses conducted for three tower sites: a consequence of complex interactions among many Scots pine in Finland, an aspen forest in Canada, atmospheric, oceanic and terrestrial processes. It is a and a native tallgrass prairie in Oklahoma. For physical expression of changes in a number of information about these sites, readers are referred to environmental factors, such as radiation, heat, Gu et al. (2002). Gross (GPP) and temperature, moisture, precipitation, etc. All these ecosystem respiration are inferred from net ecosystem factors are vital for the functioning of terrestrial exchange (NEE) measurements through a multiple ecosystems, and influence the exchange of energy and response model which uses diffuse photosynthetically mass of terrestrial ecosystems with the overlying active radiation (PAR), direct beam PAR, soil and air atmosphere. In particular, it has been observed that for temperatures as variables (Gu et al. 2002). Sky many forest ecosystems, the maximal net uptake of conditions are measured by relative irradiance, which is (CO2) often occurs on cloudy rather than the ratio of global solar radiation to clear-sky global on sunny day. In a previous study, we showed that a solar radiation at the surface (Gu et al. 1999). The boreal deciduous forest and a temperate deciduous details about the method of analysis can be found in Gu forest had maximal carbon uptake under sky conditions et al. (1999) and Gu et al. (2002). with a solar radiation level equivalent to about 70-80% of the corresponding clear-sky solar irradiance. We also 3. RESULTS demonstrated that the two could tolerate a Typical patterns of the changes of GPP with relative cloud-induced sunlight reduction of as much as 50% irradiance are featured by a concave shape (Figure 1), without lowering the capacity to sequester carbon as indicating that GPP peaks under cloudy conditions. compared with sunny days (Gu et al. 1999). However, Note that under cloudy conditions, relative irradiance uncertainties remain in regard to the mechanism of this may exceed 1. This is caused by sunlight reflections phenomenon. Since variations in net carbon uptake can from cloud walls, a phenomenon termed the ‘cloud-gap be achieved through changes in , or effect’ in Gu et al. (1999) and analyzed in detail in Gu et respiration or both, an interesting question is: does al. (2001). In general, the cloud-gap effect leads to moderate cloudiness increase net carbon uptake by enhanced canopy photosynthesis. enhancing canopy photosynthesis or by reducing Following Gu et al. (1999), we use optimal relative ecosystem respiration or both? Canopy photosynthesis irradiance and critical relative irradiance to characterize can be enhanced by increased diffuse radiation (Gu et the influence of clouds on GPP. Optimal relative al. 2002) or reduced water stress (Freedman et al. irradiance is the relative irradiance that has the highest 2001). Ecosystem respiration can be reduced by GPP for a given solar elevation angle. Critical relative decreased temperature under cloudy conditions. In this irradiance is the relative irradiance that has a GPP study, we have investigated the responses of equal to the clear-sky GPP for a given solar elevation contrasting terrestrial ecosystems to cloudiness by angle. Both optimal relative irradiance and critical analyzing tower flux measurements. We have found relative irradiance tend to decrease with solar elevation that moderate cloudiness tends to enhance canopy angle (Figure 2). For the aspen forest, however, both photosynthesis. For the influences of clouds on indices reached the minimum around a solar elevation o ecosystem respiration, the responses are more angle of 47 . The optimal relative irradiance and critical complex. Some ecosystems show reduction in relative irradiance of the tallgrass prairie, which is ecosystem respiration in the presence of clouds, but for dominated by C4 species, converge towards unity at others the response is weak. Possible explanations for low solar elevation angle. Also they are higher than this difference include temperature - moisture corresponding values for the aspen and Scots pine interactions on soil respiration. forests. This difference reflects that the photosynthesis of C4 species respond more linearly to variations in 2. MATERIAL AND METHODS sunlight than C3 species does. For comparison, Figure This study utilizes tower flux data collected and 2 includes the optimal and critical relative irradiance archived through the Fluxnet project determined for NEE. They closely match those for GPP. The influence of clouds on ecosystem respiration is * more complicated (Figure 3). It ranges from a strong Corresponding author address: Lianhong Gu, 151 dependence (tallgrass prairie) to weak dependence Hilgard Hall, Department of Environmental Science, (aspen forest) to no response (Scots pine forest). This Policy and Management, University of California, may be due to the compounding effects of soil moisture Berkeley, CA 94720; on ecosystem respiration. Further studies are needed. email:[email protected] 1.0 0 Aspen forest Aspen forest o 0.9 -5 Solar elevation angle: 55 - 60 0.8 ) -1 -10 s

-2 0.7

-15 0.6 mol m mol µ

-20 0.5 GPP ( Critical or optimal relative irradiance 0.4 -25 1.0 Scots pine forest

-30 0.9 0 Scots pine forest 0.8 Solar elevation angle: 45 - 50o 0.7 -5 )

-1 0.6 s -2 0.5 -10 Critical or optimal relative irradiance mol m mol µ 0.4 1.0

GPP ( Tallgrass prairie -15 0.9

0.8 -20 0 0.7 Tallgrass prairie o 0.6 N E E critical relative irradiance Solar elevation angle: 65 - 70 N E E optim al relative irradiance G P P critica l re la tive irra d ia n ce -10 0.5

) G P P optim al relative irradiance -1 Critical orCritical optimal irradiancerelative s

-2 0.4 30 40 50 60 70 -20

mol m mol Solar elevation angle µ Figure 2. Changes of optimal and critical relative irradiance with solar elevation angle GPP ( -30 12 o ) Tallgrass prairie, 65 - 70 -1 s

-2 10 -40

mol m mol 8 0.0 0.2 0.4 0.6 0.8 1.0 1.2 µ

Relative Irradiance 6

Figure 1. Typical patterns of GPP as a function of relative irradiance. 4 Occassionally, surface irradiance under cloudy conditions exceeds the corresponding level under clear sky due to reflection from cloud walls (Gu et al. 2001). These points are denoted by open circles in the plots. 2 Ecosystem respiration ( respiration Ecosystem 0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 8

) A s p e n fo re s t

-1 Relative irradiance o s Solar elevation angle: 55 - 60 -2 Figure 3, continue. 6 REFERENCES mol m mol µ Freedman, J.M., D.R. Fitzjarrald, K.E. Moore, and R.K. 4 Sakai, 2001. Boundary-layer clouds and vegetation- atmosphere feedbacks. J. Climate, 14, 180-197. 2 Gu, L., J. D. Fuentes, H. H. Shugart, R. M. Staebler, and T. A. Black, T.A., 1999. Responses of net Ecosystem respirationEcosystem ( 0 ecosystem exchanges of carbon dioxide to changes 8 )

-1 in cloudiness: results from two North American s -2 deciduous forests. J. Geophys. Res., 104, 31421- 6

mol m mol 31434. µ Gu, L., J. D. Fuentes, M. Garstang, J.T. da Silva, R. 4 Heitz, J. Sigler, and H. H. Shugart, 2001. Cloud modulation of surface solar irradiance at a pasture 2 site in southern Brazil. Agric. For. Meteor., 106, 117- S c o ts p in e fo re s t Ecosystem respiration ( Ecosystem Solar elevation angle: 45 - 50o 129. 0 0 .0 0 .2 0 .4 0 .6 0 .8 1 .0 1 .2 Gu, L., D.D. Baldocchi, S. B. Verma, T. A. Black, T.

Relative irradiance Vesala, E. M. Falge, P.R. Dowty. 2002. Advantages of diffuse radiation for terrestrial ecosystem Figure 3. Changes of ecosystem respiration with relative irradiance. productivity, J. Geophys. Res. (in press).