Box B： Simulation of Ship Navigation along the NSR
WP8： Simulation based on Year round and Seasonal Operation Scenarios
Reply to the reviewer
We appreciate very much Dr. Alfred Tunik's careful review and commentary. Some of the comments indicate a gap between the level of model simulation and limitation of the historical environmental data. In reality, a ship moves on a plane, although a ship moves on a line in our model. In the beginning of the study, we tried to implement a more sophisticated model that enables a course decision looking at spatial distribution of the environmental data. However we realized that we are not at a stage to implement such a detailed analysis. We made an effort to match the level of historical environmental data over an extended period, the ship speed algorithms by means of introducing a concept of ice index and finally, to grasp a trend of cost. Thus, several assumptions were adopted without sufficient verification, but we trust we are able to maintain an acceptable degree of accuracy. We recognize that these comments are well worthy of further study.
The followings are replies for each item.
#Comment 1.
The total export of forest products on the NSR reached 1.3 million tons in 1987 and it comprised 20% of the total cargo flow of 6.6 million tons in the NSR. However the export of timber rapidly decreased in 1993 due to the confusion of the Russian economy, especially high costs for fuel and transportation tariffs imposed on river transport goods. Timber exports to South Korea and Japan in 1997 were less than 100 thousand tons and have not recovered yet. Actual statistics for the timber in 1996 is not listed in the report, although it is estimated in the order of 1% Of the total cargo flow. Detail are referred to in WP3 report. (Ivanov et al., 1998)
#Comment 2.
We corrected as you pointed out and found it was a misprint. The open water speed of the Arktika class is 21 knots and that figure was used in the simulation.
#Comment 3.
Figure 2.5.5 expresses the relation between the navigation speed for SA-15 and the severity of the ice condition. The severity is expressed as a function of multiples of ice concentration and its particular category such as ice free water, gray-white ice, first-year ice, second-year ice and multi-year ice conditions. CASPPR adopted ice numerals that are calculated from ice class and ice conditions. In order to develop the ice numeral, we need the data as depicted in Figure 2.5.5 that shows the ice conditions including ice categories, their ice concentrations and average navigation velocity. Data such as Figure 2.5.5 are useful to verify the relation between ice index and velocities. We attempt to show that these kinds of data shall be collected for verification purposes and try to develop the ice numeral from the actual results. We add an explanation for the relation between ice numeral and data as in Figure 2.5.5.
#Comment 4.
We well recognize the difference between ice conditions along a route and a certain wide area. Our model does not take into account for stuck in ice under compression and we roughly estimated ship speed behind icebreakers. We should have categorized more detailed navigation modes as you pointed out. In this study, we focused on link ice conditions given by the AARI/WP2 and the algorithms developed by WP6. The historical data from the AARI are mainly based on satellite data and their resolution is not sufficient to recognize narrow leads or other small ice features. Even ridge distributions are lacking as we indicated in chapter 2.2. We have to develop probabilistic models to estimate the existence of narrow lead or other ice features that are easy to by-pass in order to fulfill a gap between limited data and the implementation of more detailed simulation. We recognize that the issues you raised are important points to enhance accuracy of the simulation in the future.
