CLOUD RADIATIVE FORCING AND EFFECTS ON ...

1 downloads 0 Views 98KB Size Report
M. Nardino, A. Orsini, R. Pirazzini, F. Calzolari, T. Georgiadis and V. Levizzani. ISAO-CNR, Via Gobetti 101, 40129 Bologna, Italy. Tel. +39 051 6399586 Fax.
CLOUD RADIATIVE FORCING AND EFFECTS ON LOCAL CLIMATE M. Nardino, A. Orsini, R. Pirazzini, F. Calzolari, T. Georgiadis and V. Levizzani ISAO-CNR, Via Gobetti 101, 40129 Bologna, Italy Tel. +39 051 6399586 Fax. +39 051 6399652 E-mail [email protected]

Abstract Cloud radiative forcing is one of the main regulating factors of the Earth's climate. The effects of cloud type, coverage and microphysical properties on local climate have been recognised of paramount importance especially when climate models are applied to forecast future trends for planning land-use and availability of resources. Global change studies have identified the modifications in the cloud coverage and occurrence as an important factor. Well established climatic regimes will experience modifications in a not too distant future because of the changes in the solar radiation reaching the surface. The different scenarios of the radiative transfer itself lead to different thermodynamic regimes both in-cloud and at the surface. In presence of living organisms or water in its different phases the surface properties, like the spectral albedo, are strictly related to micrometeorological quantities and local thermodynamic parameters. The radiative forcing as a function of cloud transmissivity and surface albedo is modelled for different experimental conditions. 1. Introduction The influence of the cloud radiative forcing on the surface radiation balance is important for the effects that the clouds could produce on the Earth’s climate. The impact of the clouds on the radiative fluxes is matter of several studies for climatic sensitivity and global changes because the great uncertainties that are present. For example, satellite studies have shown that clouds tend to warm the earth-atmosphere system over the polar regions (Nakamura and Oort, 1988), while recent studies show, in contrast, that over polar regions clouds have a cooling effect (Schweiger and Key, 1994). This discrepancy could be attribute to the fact that the behaviours of the surface radiative components as function of the cloud amount, are different. In fact, the clouds have a cooling effect because of the shortwave radiation and a warming effect because of the longwave radiation. The climatic variations could cause changes in the radiative fluxes with consequences on the global sea level. Because the clouds and their characteristics are influenced by the climatic variations, the study of the clouds effects on the radiative fluxes over the snow and ice surfaces become important in order to evaluate the global changes and feedback. Several studies, conducted in polar regions, showed that over highly reflecting snow surfaces the net radiation increases with cloud amount and this fact is known as the “radiative paradox” (Ambach, 1974). Effectively this happens when the increase of the longwave radiation amount dominates the decrease of the shortwave radiation. The study reported in this work, is focalised to the evaluation of the cloud radiative forcing on the net radiation for two Antarctic and an Arctic sites. The total effective transmission τ, defined as the ratio of incoming solar radiation at the surface and at the top of the atmosphere, has been computed for all sites through radiation measurements. This parameter, during clear sky conditions, depends principally on the amount and vertical distribution of water vapour and ozone, while during overcast conditions is determined mainly b the cloud proprieties which are dependent on the cloud optical thickness (Bintanja and van den Broeke, 1996).

1

Following the parameterisation proposed by Bintanja and van den Broeke (1996) according to Konzelmann et al. (1994) it was studied as function of the cloud amount: τ = τ cl + (τ ov − τ cl )N 2 (1) where τcl and τov are the transmissivities under clear sky and overcast conditions and N is the cloud cover index. In order to analyse the dependence of the net radiation on cloud amount in terms of surface albedo and effective cloud transmissivity the parameterisation proposed by Bintanja and van den Broeke (1996) was considered: ∂R ∂Lwn = 2 SwTOA( 1 − α )( τ ov − τ cl ) N + (2) ∂N ∂N where SwTOA is the daily mean incoming shortwave radiation at the top of the atmosphere and ∂Lwn / ∂N is the slope of the linear dependence of the net longwave radiation on N. Through this formula, based essentially on experimental data, the dependence of R on N can be investigated. Finally, the shortwave (CRFs ) and longwave (CRF L) radiative forcing at the surface were computed to quantify the influence of clouds on radiation budget and on local climate. They are defined as: CRFS = N [S ov − S cl ] (3) CRFL = N [Lov − Lcl ] where the subscripts cl and ov represent clear sky and overcast conditions, respectively. 2. Methods In this study radiation data collected in two Antarctic and an Arctic site are used. The three locations differ mainly to the height a.s.l., and, consequently, the surface temperature, the vertical structure of the atmosphere, the cloud amount and the cloud type. At Reeves Névé (Lat. 74° 39’S, Long. 161° 35’E, 1200 a.s.l.) the measurements have been carried out from 24th November 1994 until 7th January 1995 without interruptions for the radiometric measurements. The station was equipped with: a pyrradiometer (0.3-60 µm, Schenk 8111, Austria), an albedometer (0.3-30 µm, Schenk 8104, Austria), a soil heat flux plate placed at 5 cm depth and four thermocouples, three of them buried in the snow (10, 50, 100 cm depth) and the latter sensor placed at 50 cm over the snow surface and shielded from the solar radiation. The other two stations, Dome Concordia (Lat. 75° 06.06’S, Long. 123° 20.74’E, 3306 m a.s.l.) and Ny-Ålesund (Lat. 78°55’N, Long. 11°56’E 13 m a.s.l.), were equipped with the same instrumentation: a radiometer model CNR1 (Kipp and Zonen), and three thermocouples. At Dome Concordia the acquisition started on 20th January and ended on 2nd February 1997, while at Ny-Ålesund it started on 19th March and stopped on 14th April 1998. The three thermocouples, used for the computation of the subsurface heat flux, were buried into the snow at 5 cm and 10 cm depth at Dome Concordia and at 8 cm and 17.5 cm at NyÅlesund. The other temperature sensor was used for the computation of the skin temperature. For all sites the radiometric measurements were recorded by a Campbell CR10 ET datalogger every minute, averaged every 10 minutes and then stored in a Campbell SM192 memory module. The cloud cover index has been computed from observational measurements at NyÅlesund downloaded from the data base of Koldewey station (Alfred Wegener Institute in cooperation with Norwegian Polar Institute) and from visual observations or by the daily longwave radiation for the other two sites.

2

3. Results In figure 1 is reported the behaviour of the total transmission as function of the cloud cover index for all sites. The curve represents equation (1) and the best fit values are reported in the legend. It can be noted the good agreement between the data and the parameterisation proposed by Bintanja and van den Broeke (1996). Through the best fit parameters the transmissivity during clear sky and overcast conditions have been computed and the results are reported in Table 1. The τcl values are quite similar at Dome Concordia and reeves Névé and the highest value of Dome Concordia could be attributed to the location of this station, where the air is drier and more rarefied (the station is situated at 3306 m a.s.l) and probably the clouds had thinner optical thickness. The values of 0.702 resulted at Ny-Ålesund could be due to the relatively warm and moist conditions of this coastal site. The values of τov are different between the three locations and this could be due principally to the differences in the shortwave cloud properties as explained by Bintanja and van den Broeke (1996). For Reeves Névé and Dome Concordia τov is larger than 0.6: this can mainly be attributed to the occurrence of multiple reflection between surface and cloud base, as showed by Wendler et al. (1981). The low values founded at Ny-Ålesund shows that the optical thickness is very high causing a lower amount of shortwave radiation that reaches the surface. Probably, the clouds present at Ny-Ålesund during the experiment were principally constituted by stratus and cumulus. In figure 2 is shown the trend of albedo as function of the cloud cover index for NyÅlesund and Reeves Névé. The curve represents the best fit of the following equation: α= a + b Nc (4) This fit has been applied to Ny-Ålesund data and the c values was resulted equal to 2.3 and consequently set to 2. The intercepts of the curves represent the albedo under clear sky conditions for both sites. The differences can be due to the presence of a more reflected snow surfaces at Reeves Névé; anyway, the values of albedo are comparable and characteristic of polar surfaces. In figure 3 is reported the dependence of ∂Rn / ∂N on the transmissivity under overcast conditions and albedo according to equation (2). The values of N, τcl and ∂Lwn / ∂N were obtained by the average of the three experimental campaign in order to have a comparison between the three sites. The term ∂Lwn / ∂N represent the slope of the best fit line of the net longwave radiation as function of N as shown in Pirazzini et al. (2000). The graph shows that and increase of R with increasing N occurs when both albedo and τov are large and the opposite when both are small. In this way both albedo and transmissity under overcast condition have the same importance with regard to the increase of R with N. Comparing this graph with the results of Pirazzini et al. (2000) where only the data of Reeves Névé and Dome Concordia were considered, it can be noted that the values of albedo and τov for which dR/dN starts to be positive, are lower. This means that at Ny-Ålesund the warming effect of the clouds on the net radiation is more pronounced respect to the Antarctic sites. Nevertheless, the cloud cover and characteristics of the three sites are quite different because the different locations, the results show that an increase in net radiation with increasing cloud amount occurred only in the accumulation zone (high albedo) and not in ablation zone in according to Bintanja and van den Broeke (1996). Finally, in figure 4 are reported the value of the cloud radiative forcing obtained through equation (3) for all sites. Because CRFL is normally positive, clouds warm the surface in the longwave region and they cool the surface in the shortwave region. Looking to the graph it is clear that at Reeves Névé and Ny-Ålesund the clouds warm the surface and this effect is stronger at Ny-Ålesund, while at Dome Concordia the increase of the longwave radiation with N is near equal to the decrease of the shortwave radiation with N. For the Dome Concordia local climate the useful data are too few to firmly states this result, while at Reeves Névé and Ny-Ålesund 3

during the considered polar summers the clouds tend to warm the surface according with other Antarctic and Arctic studies (Bintanja and van den Broeke 1996; Ambach, 1974). 4. Conclusion The main result of the present work is that during Arctic and Antarctic summer the net radiation increased with the cloud cover index. This phenomenon is the result of the asymmetry in the dependence of the net shortwave and longwave radiation on the radiative cloud properties and cloud amount. The results show that the warming effects of the clouds is observed when both cloud transmissivity and albedo are highs. The τov high value is attributable to the combined effect of a low cloud optical thickness and multiple reflection between surface and cloud base. The effect of clouds warming and cooling of the three sites has been studied and for Dome Concordia, situated on Antarctic plateau, it can be said that the surface is in a near equilibrium but the small data quantity do not permit to accept this result. The coastal zone of the Reeves Névé glacier is one of the principal zones of the entry of the low pressure air masses that characterise the Antarctic climate. Consequently, an increase in the cloud amount could cause an increase of the surface temperature and a change in the surface mass balance of the ice sheets causing the glacier melt and consequences on the global sea level. On the other hand, in the Arctic zone the climatic variations due to a greater presence of water vapour in atmosphere could produce a feedback that induce a surface warming and an environmental susceptibility.

Acknowledgements The authors wish to acknowledge ENEA–Progetto Antartide and PNRA (National Program for Research in Antarctica) for financing and supporting this research; a particular thank to Dr. M. Zucchelli for the logistic and technical assistance during the whole experiment. Thanks to the AWI (Alfred Wegener Institute) for the possibility to use the Koldewey station data. Finally, a special thank to Fabrizio Ravegnani, Ubaldo Bonafè and Giuliano Trivellone for their contribution to the Arctic experimental campaign.

References Ambach, W., 1974: The influence of cloudiness on the net radiation balance of snow surface with high albedo, J. Glaciol., 13, 73-84. Bintanja, R. and M.R. Van den Broeke, 1996: The influence of clouds on the radiation budget of ice and snow surface in Antartica and Greenland in summer, Int. J. of Climatology, 16, 1281-1296. Konzelmann, T., R.S.W. van de Wal, W. Greuell, R. Bintanja, E.A.C. Henneken and A. AbeOuchi, 1994: Parameterization of global and longwave incoming radiation for the Greenland Ice Sheet, Global Planet. Change, 9, 143-164. Nakamura, N. and A. H. Oort, 1988: Atmospheric heat budget of the polar regions, J. Geophys. Res., 93, 9510-9524. Pirazzini, R., S. Argentini, F. Calzolari, V. Levizzani, M. Nardino, A. Orsini, G. Trivellone and T. Georgiadis: Study of the cloud forcing at two Antarctic sites: Reeves Névé and Dome Concordia. Proceeding of 8th Workshop Italian Research on Antarctic Atmosphere, Bologna, Italy, 20-22 October 1999 (in press). Schweiger, A.J. and J.R. Key, 1994: Arctic Ocean radiative fluxes and cloud forcing estimated from the ISCCP C2 cloud dataset, 1983-1990, J. Appl. Meteorol., 33, 948-963. Wendler, G., F.D. Eaton and T. Ohtake, 1981: Multiple reflection effects on irradiance in the presence of Arctic stratus cloud, J. Geophys. Res, 86, 2049-2057.

4

Total transmission

(a) Ny-Alesund 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.0

τ

=0.702-0.606 N R=0.94

2

0.2

0.4

0.6

0.8

1.0

0.8

1.0

0.8

1.0

Cloud Amount

Total transmission

(b) Reeves Nèvè 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.0

τ

=0.87-0.26 N R=0.87 0.2

2

0.4

0.6

Cloud Amount

Total transmission

(c) Dome Concordia 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.0

τ

=0.89-0.14 N R=0.97 0.2

2

0.4

0.6

Cloud Amount

Figure 1. Dependence of the daily mean total transmission on cloud amount for (a) Ny-Ålesund (b) Reeves Névé and (c) Dome Concordia.

5

Table 1. Mean transmissivity in clear-sky (τcl) and overcast (τov) conditions at the three measurement sites. τcl

τov

Ny-Ålesund

0.702

0.096

Reeves Névé

0.87

0.61

Dome Concordia

0.89

0.75

(a) Ny- Alesund 1.00 2

0.95

a=0.76+0.204 N R=0.95

albedo

0.90

0.85

0.80

0.75

0.70 0.0

0.2

0.4

0.6

0.8

1.0

0.6

0.8

1.0

Cloud Amount

(b) Reeves Névé 1.00 2

0.95

a=0.83+0.06 N R=0.63

albedo

0.90

0.85

0.80

0.75

0.70 0.0

0.2

0.4

Cloud Amount

Figure 2. Dependence of the daily mean albedo on cloud amount for (a) Ny-Ålesund (b) Reeves Névé.

6

0.6

0.55

0.5

dR/dN>0

0.45

0.4

0.35

dR/dN