82 | | GloSea5 data have been exposed as two different virtual datasets through ECOMS-UDG, based on their different forecast extents, initializations and member configuration. Members are defined from all combinations of the members and lagged runtimes. |
83 | | |
84 | | In particular, in UDG members are ordered considering all combinations of member `x` runtime, being member the factor that varies fastest. For instance, a call to `loadECOMS` using `members = 1:5` in `dataset = Glosea5_seasonal_12` will return, in this order, the following 5 members: |
85 | | |
86 | | Member1: [member1,runtime1] |
87 | | |
88 | | Member2: [member2,runtime1] |
89 | | |
90 | | Member3: [member3,runtime1] |
91 | | |
92 | | Member4: [member4,runtime1] |
93 | | |
94 | | Member5: [member1,runtime2] |
95 | | |
96 | | |
97 | | The main characteristics of both GloSea5 hindcasts are are next enumerated: |
98 | | |
| 82 | GloSea5 data have been exposed as two different virtual datasets through ECOMS-UDG, based on their different forecast extents, initializations and member configuration: |
| 98 | Members are defined from all combinations of the members and lagged runtimes. In particular, in UDG members are ordered considering all combinations of member `x` runtime, being member the factor that varies fastest. For instance, a call to `loadECOMS` using `members = 1:5` in `dataset = Glosea5_seasonal_12` will return, in this order, the following 5 members: Member1: [member1,runtime1], |
| 99 | Member2: [member2,runtime1], |
| 100 | Member3: [member3,runtime1], |
| 101 | Member4: [member4,runtime1], |
| 102 | Member5: [member1,runtime2]. |
| 103 | |
| 104 | |