The rest of this paper evaluates 

 modelled iceberg deterioration by 

 examining environmental terms 

 used in the formulae. The envi- 

 ronmental assumptions regarding 

 each melt process are also re- 

 viewed. Building on White's 

 research (White et al, 1980), 

 various observed thermal and 

 velocity parameters are independ- 

 ently compared to determine 

 which of each best represent the 

 terms in the formulae. Using 

 Anderson's operational computer 

 model (Anderson, 1983), melt 

 estimates are calculated from 

 operational data center inputs. 

 Figure E-6 depicts the contribution 

 of each deterioration process to 

 illustrate its relative importance 

 and the changes in contribution 

 caused by using different parame- 

 ters to represent terms in formu- 

 lae. The implications of error 

 estimates for various model inputs 

 on IIP operations are then dis- 

 cussed. 



WARM AIR CONVECTION 



Melt from warm air convection is 

 ignored in the model. For March 

 through mid-May, no melt is 

 estimated for air convection. For 

 July through September, the 

 average melt is estimated at 

 Scm/day, assuming an average 

 daily air temperature of 1 0°C and 

 average wind of 37km/hr. 



The daily average air temperature 

 warmed during the study period 

 from 6°C on 1 5 June to 8°C on 21 

 June. The average wind speed for 

 the study period was 33km/hr. 

 Warm air heat convection was 

 estimated at 4cm/day. 



Climatological average air tem- 

 peratures for the IIP region could 

 be used to make monthly melt 

 estimates. Likewise, daily global- 

 scale air temperature values could 

 be requested from an operational 

 data center; however, this level of 

 effort for a relatively insignificant 

 deterioration process seems 

 impractical for operational fore- 

 casting purposes. 



SOLAR RADIATION 



The modelled melt due to solar 

 radiation is fixed at 2cm/day, 

 which represents the minimum 

 melt rate for the period March 

 through August (Anderson, 1983). 

 The model assumes cloudy 

 conditions. 



The daylight (0800Z to 2400Z) 

 was obscured (100%) by cloud 

 cover or fog every day of the study 

 except for the afternoon of 17 

 June and rrwming of 19 June. For 

 those half day periods the skies 

 were partly (averaged 50%) 

 cloudy. Assuming a 35% albedo 

 for an iceberg, the average melt 

 rate for the June study period was 

 4cm/ day (White et al, 1980). 



The model could be adapted to 

 the monthly melt estimates 

 derived by White, although the 

 benefit would be minimal. Like- 

 wise, global-scale radiation 

 estimates could be requested from 

 operational data centers (COMNA- 

 VOCEANCOM, 1986); however, 

 the level of effort to identify those 

 periods of clear skies would only 

 provide an additional melt of 

 2cm/day. 



Again this level of effort for a 

 relatively insignificant deterioration 

 process seems impractical for 

 operational forecasting purposes. 



WATER TEMPERATURE 



The model uses sea surface 

 temperature to estimate both 

 buoyant and forced convection 

 contributions to iceberg melt. The 

 melts due to buoyant and forced 

 convection were computed as a 

 function of observed sea surface 

 temperature (T^) and as a func- 

 tion of the in-situ temperature of 

 the water column integrated over 

 the estimated draft of the iceberg 



(fTo)- 



Buoyant Vertical Convection 



Buoyant convection is considered 

 solely dependent upon the "rela- 

 tive" temperature between a near 

 vertical wall of ice and the water 

 column. The cluster's average 

 daily melt due to buoyant convec- 

 tion using Jt, was estimated at 

 2cm/day with average values for 

 individual icebergs ranging from 

 1 cm/day (#787) to Scm/day 

 (#785). The melt rate as a func- 

 tion of Tg averaged 7cm/day 

 greater; with daily differences 

 ranging from +3cm/day (#787/21 

 June) to -f-1 IcnVday (#747/20 

 June). These differences were 

 associated with surface tempera- 

 tures which were approximately 

 1 .5°C warmer than the averaged 

 temperature for the first ten meters 

 of the water column. 



121 



