CMIP5 Questionnaire

Common Problems and Sollutions / Comments

A place to collect the problems that beta testers find with the questionnaire. These issues may require a ticket to be raised and/or an entry into the questionnaire FAQ documentation.

InstituteBeta Testers (Metafor Guide)
GFDLBalaji et al (Balaji)
IPSLSandrine Bony, Pascale Braconnot, Olivier Marti (Mark M & Eric)
UREADLen Shaffrey (Gerry)
MOChris Jones, John Hughes, Matt Collins (Mark E)
CERFACSSoline Bielli (Sophie & Marie-Pierre)
CNRMDavid Salas y Melia (Antoinette)
NCARGarry Strand (?)
MPI-MHans, Heiner Widmann, Stephanie Legutke, Nebojsa Balic

Following table is for harvesting of beta testers feedbacks:

Contributions history:

  • CERFACS (marie-pierre) - 14/12/09
  • UREAD (gerry) - 15/12/09
  • CURATOR (sylvia) 16/12/09
  • MOHC (mark) 18/12/2009
  • Climpact (michel kolasinski) 21/12/2009
  • DKEZ (hans) 27/07/2010
InstitutePersonType of problemPriorityPart of the questionnaire involvedProblemSolution (Comment)Ticket requiredFAQ entry
1CERFACSS. Bielli (marie-pierre)unclear (meaning)Summary Tab: Models Framewhat does mean "status"?"The status column provides an indication of how much of the model description has been completed"NoNo
2CERFACSS. Bielli (marie-pierre)improvement (simple)Summary Tab: Models Frameno mean to know WHO has edited a particular model templatewould be useful to show up ContactName in the Models part of the Summary page#490No
3CERFACSS. Bielli (marie-pierre)unclear (road path)Model Tab: "Responsible parties" ParamInformation on how to enter a new party is not straightforwardSee slide 4 of the roadmap#491No
4CERFACSS. Bielli (marie-pierre)improvement (structure)Platformwhy Platform does not appear in a tab like Model and Simulation? We have to navigate Summary page when we want to add a new platformmake Platform accessible in a Tab. CLP: The Platform does not appear in a tab because we don't expect users will need to access the platform very often.NoNo
5CERFACSS. Bielli (marie-pierre)unclear (meaning)Model Tab: at top level<Untick the box if there is no representation of model in your model>suppress this sentence when at top level of the model hierarchy#492No
6CERFACSS. Bielli (marie-pierre)spelling errorModel Tab: at top level and realm level"geneology"to be replaced by: "genealogy"#493No
7CERFACSS. Bielli (marie-pierre)unclear (meaning)Model Tab: all components"General Attributes"to be replaced by: "General Properties" (attributes sounds "software")NoNo
8 (id.44)CERFACSS. Bielli (marie-pierre)unclear (meaning)DiscussionModel Tab: top level model: Genealogy Frame: Previous version Param<If this version is a substantial upgrade to a previous version> (what does "substantial" mean? upgrade=new version)CLP: Yes, "substantial" is a subjective term and Yes, "upgrade" implies a new version. Hopefully "upgrade" also implies that the main changes are not so "substantial" that they can't be described.#537Done
9CERFACSS. Bielli (marie-pierre)inconsistencyModel Tab: References Page (add a new reference)a URL is requested even is "offline" is selected for Reference Type#494No
10CERFACSS. Bielli (marie-pierre)improvement (simple)Parties TabURL is mandatoryshould rather be e-mail address mandatory#495No
11CERFACSS. Bielli (marie-pierre)improvement (structure)Simulation Tab: Simulation (edit)Strange to have Simulation page in yellow when no simulation is yet defined (user try to click on without success). When first adding a simulation nothing occurs: you need to click on "Edit" button to have the simulation (but this is not noticed, so the user waits at previous stage wondering why nothing appear...).Make Simulation Tab appear only when a first simulation has been defined. Make Edition page appear directly each time a new simulation have been added (when copying an existing simulation, the Edit page appear directly. So the request is just implementing the same dynamical behaviour for "add new" as for "copy")NoAwaiting
12CERFACSS. Bielli (marie-pierre)improvement (simple)All Tabswhen leaving a page without saving (although there is a permanent red message in the top left corner of each page it is very easy to forget to click on the "save" button)Make a warning message appear before leaving the page: "Are you sure you want to quit without saving?"#496Done
13CERFACSS. Bielli (marie-pierre)unclear (road path)Model Tab: top level model: "General Attributes" Paramthis rubric is empty (no predefined attribute). The user does not know a-priori that he will be asked to describe thing within sub-components, so he wonders want he has to do when arriving there….add a sentence explaining what the user is expected to do a this stage.#497Done
14CERFACSS. Bielli (marie-pierre)unclear (meaning)Model Tab: (title)Different titles are used, e.g.: "Model Component AerosolSpaceConfig", "Model Realm Aerosols", "Model CERFACS_VM1". But nowhere explanation is given (only "models" and "components" in Introduction on the Summary page)add a sentence explaining the differences between "Model", "Realm " and "Component". CLP: Good Idea.
Why not inform more clearly the user where he is? E.g.: "Your are editing the page related to the top level model", "Your are editing the page related to Aerosols realm component", "Your are editing the page related to AerosolSpaceConfig basic component"
CLP: We do this already.
MPM: Yes you have a page title, e.g. "Component !AtmChem2D-Sources" and suggestion was only about the title phrasing (for hilight)
15CERFACSS. Bielli (marie-pierre)confusing (words)Model Tab : Needed Inputs Page (title)<Inputs required by this component are: >"Inputs required" to be replaced by "Inputs Needed"NoNo
16CERFACSS. Bielli (marie-pierre)tricky (structure, road path, concepts)Simulation/Model? Tab : Needed Inputs + Resolve Inputs + Filesjuggling between "Needed Inputs", "File", "Resolve Inputs" and "Conformance" is complex (both concept and road path)CLP: Yes this is not obvious but the Roadmap will help usersNoAwaiting
17CERFACSS. Bielli (marie-pierre)improvement (simple)All Tabs: "other" choicewhen entering a value in case of "other" choice, after saving, "other" selection disappear (look strange: as if user had not make a choice…)make "other" being kept selected. CLP: I can't recreate this problemNoNo
18CERFACSS. Bielli (marie-pierre)unclear (road path)Model/Simulation? Tab: Resolve Inputs Page (new file)If the user only define a new file (File Tab) without attaching Variable(s), then this file won't appear in the list "From" in Resolve Inputs Page (in case of AncilliaryFile).This is normal (one can "Bind" only "to variable(s)" and not "to files" but it is disconcerting for the user not to see his file appear in the list. This is explained in the Edit-File Page (<This column may not be necessary for files being used for ancillary files, but it will be necessary to identify variables for binding to boundary and initial conditions >) but not really obvious to catch this point…#500Awaiting
19CERFACSmarie-pierreimprovement (structure)Model/Simulation? Tab: Resolve Inputs (BoundaryCondition-Coupling)Not possible to "Bind to Variables" when Needed Input is of type BoundaryCondition-Coupling, but only "to Components"Could we imagine a additional Tab "Model Variables", working similar to File Tab (describe Variables and choose their Source Model Component)? Instead of "atmosphere, ocean, etc." in the From-list for BoundaryCondition-Coupling we would have "temperature_in_air_at_2_meters(atmosphere), sea_surface_temperature(ocean), etc."#501No
20CERFACSS. Bielli (marie-pierre)confusing (words)Files Tab: (add new files)"Return (no edit)" button is confusing because it name is still "Return (no edit)" even if we have clicked on "Save" button just before. Return to what?Change button name into "Return to the list of files" + a message warning in case the user have not pressed the "Save" button before returning#502No
21CERFACSS. Bielli (marie-pierre)confusing (words)Model Tab: Needed Inputs Page (text)<on the "coupling" pages at the top level components within your model (to establish default coupling), and on the "coupling" page of the simulation description>: "coupling page" does not exist + "default coupling" does not mean anything to the userreplace <"coupling" pages> by <"Resolve Inputs" button> + suppress "default couplings" or explain it. CLP: This has been fixedNoNo
22CERFACSunclear (meaning)Model Tab: Needed Inputs Page (text)<( or in the case of a boundary condition another component)>to be replaced by: <(or "bind" to another component, in the case of a boundary condition)> CLP: I can't see this problemNoNo
23CERFACSS. Bielli (marie-pierre)confusing (words)Simulation Tab: Required Inputs and Couplings Frame (title)The wording must be unique: "Required Inputs" is not the one chosenReplace the frame title by: "List of Inputs Needed by the Components of the Model that ran for this Simulation that have to be Resolved here"#503No
24CERFACSS. Bielli (marie-pierre)confusing (words)Simulation Tab: Needed Inputs Frame (text)"Input Coupling" button does not existsentence to be replaced by: <You can modify the Resolving of the "Needed Inputs" in this list by using the "Resolve Inputs" button >#504No
25CERFACSS. Bielli (marie-pierre)questionSimulation/Model? Tab: Needed Inputs Frame (list)concatenated names of Needed Inputs are prefixed with "Coupling4". What does Coupling4 mean?CLP: Coupling4 is text speak for the word Coupling for, compare to Radio K7 in French. We need to reword this form of vocabulary.#505No
26CERFACSS. Bielli (marie-pierre)confusing (words)Model Tab: Needed Inputs Page (title)<Input Requirements for component: "AtmosphericChemistry">to be replaced by <"Inputs Needed" by the AtmosphericChemistry realm component >NoNo
27CERFACSS. Bielli (marie-pierre)confusing (words)Simulation Tab: Ensemble Characteristics Frame (text)"Ensemble Characteristics" is not the button nameReplace the phrase by <(You can modify these Ensemble Characteristics using the "Describe Ensemble" button on the left hand menu.)>#506No
28CERFACSS. Bielli (marie-pierre)unclear (meaning)Simulation Tab: Conformance Frame (text)<(You can modify these using the button on the left hand menu.)>. Which button?replace by: <...using the "Conformance" button...>#507No
29CERFACSS. Bielli (marie-pierre)bug?Experiment Tab: (copy a Simulation)When copying a simulation from another, (i) the list of Numerical Requirement disappear and (ii) the list of Needed Inputs is still there, but no longer resolved.(ii) is probably normal (force the user to check for all of his inputs… anyway we could imagine only a warning <Don't forget to check your Inputs binding">). But (i) looks like a bug. CLP: Yes, this is a feature :-) #570No
30CERFACSS. Bielli (marie-pierre)bug?DiscussionExperiment Tab: (copy a Simulation)when copying a simulation, Members are lost in the copy (number of Members reinitialize to 1). Is there a reason not to copy across Members' characteristics?CLP: I guess we need to discuss this but at the moment the functinality is to not tranfer much information so that the users ensure that the information they enter is specific to the simulation in hand.#543Done
31CERFACSS. Bielli (marie-pierre)feeling (tricky structure, road path, concepts)DiscussionSimulation Tab: Conformance PageThe list of Needed (Resolved?) Inputs defined by the user appear in the "Input Bindings" Box of each Numerical Requirement. This is disappointing to the user who rather expect a list of Files (or Variables in Files).tricky concept: defining a "Needed Input" and "Resolving" it are 2 prerequisites before establishing a Conformance and establishing a Comformance consists in binging a "Numerical Requirement" to a "Resolved Input". CLP: I addressed how we should guide the users through the conformance in the Roadmap. We will link pages of this help resouce to relevant sections of the Questionnaire#513Awaiting
32CERFACSS. Bielli (marie-pierre)question + feeling (tricky structure, road path, concepts)Simulation Tab: Conformance PageThe list of Needed (Resolved?) Inputs defined by the user appear in the "Input Bindings" Box of each Numerical RequirementThis means that defining a "Needed Input" and "Resolve" it are 2 prerequisites before establishing a Conformance? CLP: YesNoAwaiting
33CERFACSS. Bielli (marie-pierre)improvement (simple)Parties Tab: (deletion)One can easily delete a Party (even if not created by him)add a warning message "are you sure you want to deleted this Party?" CLP: When I tried to recreate this the questionnaire broke and presented me with an Internal Error message"#510
34CERFACSmarie-pierreimprovement (structure)Simulation Tab: (deletion)Deletion of simulation is not implemented#511No
35CERFACSmarie-pierreimprovement (simple)Simulation Tab: Conformance Page (binding)the fulfilling of Conformance consist of binding a Numerical Requirement to (one or several) "Model Modifications" and/or to (one or several) "Input Binding". The way of registering the users's selections is not very friendly (only blue underlying, easily erasable)...could we envisage to recap the selection(s) on the right hand Notes Box? CLP: My instinct is that the non-friendly method of registering selections was chosen to save on screen real-estate. If no change is made I will make sure that this is addressed in the FAQ. A change has been made and is awaiting review.#531No
36 (supressed)CERFACSmarie-pierreconfusing (words)Simulation Tab: Ensemble Characteristics Page: Input Modifications Page (title)<Edit inputmod description>Title has changed: "Inputs Modifications" (that's OK)NoNo
37CERFACSmarie-pierreunclear (meaning)Simulation Tab: Ensemble Characteristics Page: Input Modifications Page: "Date" Parammessage: "Enter a valid date". what is the format of a valid date? What is "Date" for?The date has been removed#532No
38CERFACSmarie-pierreunclear (meaning)Simulation: Ensemble Characteristics Page: Input Modifications Page: Inputs BoxIs this box designed to select the "Reference Input" to which perturbation ("Input Modifications" are performed form an ensemble member to another?if so, an explanatory phrase would be welcome#533Ongoing
39CERFACSmarie-pierrefeeling (tricky structure, road path, concepts)DiscussionSimulation/Model? Tab: Needed Inputs; Simulation Tab: Describe Ensemble: Input Modificationsnot easy to understand the link between these concept, f.e. that an "Input Modification" are relative to a "Needed input" (and not to a "Resolve Input")I think part of the problem here is that we use the word "input" in different contexts. We use it to describe input variables, we use it to describe input modifications, we also use it to say that inputs are needed. Need to think of how we can distinugish these concepts more clearly.#558Awaiting
40CERFACSmarie-pierreunclear (meaning)Simulation Tab: Model Mods Page (title)Assign modelmod(s) to My_SimulReplace by: <Assign the"Model Modifications" (i.e. modifications compared to the basic Model) performed for the purpose of the current Simulation>#557No
41CERFACSmarie-pierrefeeling (tricky structure, road path, concepts)Simulation Tab: Model Mods; Simulation Tab: Ensemble Characteristics Page: Manage Model Modificationsnot very easy to see what is the difference of mechanism between the 2 (assignment for the former)add an explanatory phrase telling that Model Modifications assigned to the current Simulation are common to all Members (the others are member-specific)#559Ongoing
42CERFACSmarie-pierreimprovement (simple)Simulation Tab: List of Model/Inputs? Modificationsthe Modifications attached to all Simulations (and eventually to all the members of each simulation) appear in a single list. Difficult to distinguish from the sole user-defined mnemonic…add a prefix to identify the Simulation and/or Member origin of a Modification: simulX:MemonicMod; simulX:MemberY:MnemonicMod#536Ongoing
43CERFACSmarie-pierreunclear (meaning)Simulation Tab: Resolve Inputs and Couplings Page<Input resolution for the inputs to "My_Model" for simulation "My_Simul">to be replaced by: <Give details on how the Needs in term of Inputs are satisfied>#560No
44 (id.8)UREADLen Shaffrey (Gerry)unclear (meaning)DiscussionModel GenealogyWhat is the scope of 'genealogy? i.e should it refer to previous 'different' models that the current model is heavily based on, or only previous versions of the same model, e.g. HiGEM is based heavily on HadGEM1 (similar to number 8 above)The scope of the Geneology section should really be governed by the users. In the case that Len rasises, yes I would expect users to tell us about important differences between HadGEM1 and HiGEM.#537Done
45 (id.5)UREADLen Shaffrey (Gerry)improvement (simple)Model Implemented?No need for 'implemented?' at top level model pageRemove 'implemented' tick box at top model page#492No
46UREADLen Shaffrey (Gerry)Improve ease of useResponsible partiesAllow details to filter down to subcomponents - ease repetitivenessImplement a 'copy downwards' type tickbox#534No
47UREADLen Shaffrey (Gerry)unclear (meaning)Responsible partiessome confusion over specific difference between 'lead author' and 'contact'Better wording or instruction#491Done
48UREADLen Shaffrey (Gerry)File access rightsDiscussionFile linkWhat to do in circumstances where user doesn't have rights to distribute file link details#535Done
49 (id.18)UREADLen Shaffrey (Gerry)unclear (structure)resolve inputsUnable to bind files to input requirements, only the variables contained within. The example the user gave was when referring to the atmosphere initialisation file as a whole rather than each individual variable contained within. (as in number 18)#500
50UREADLen Shaffrey (Gerry)unclear (structure)ConformanceWhat should be done in circumstances where 100% conformance can not be attained but results are to be uploaded anyway. Real examples of this occurring were given, e.g. inability to model changing land surface conditionsThis issue has been fixed (id.66)#508Done
51UREADLen Shaffrey (Gerry)unnecessary questionResponsible partiesIs there a need for 'lead author' at the lower level model components?Should be resolved by having the 'copy responsible party details downwards' tickbox (as detailed in 46)#534
52 (cf.35)UREADLen Shaffrey (Gerry)unclear (roadmap)Model modificationsSteps needed to bind 'model modifications' to the conformance page are not detailed enoughNeed to add that once you have described the 'model modification' you need then to tick the box to bind it to the conformance page#531Ongoing
53 (id.20)UREADLen Shaffrey (Gerry)unclear (words)General'Return (no edit)' not immediately clear (similar to number 20 above).Change to 'Return (without saving)?#502No
54UREADLen Shaffrey (Gerry)unclear (words)GeneralWhat units to use when filling out component attributes/parametersChange help text to state 'remember to include units'.
55CURATORSylvia MurphyExplanation NeededSimulationsOn the main page, there is a link to add a model, add a platform or manage simulations. I kept hitting manage simulations and it would through an error...must add a platform first or must add a model first.We can prevent users from having the same click through annoyances I did by adding some explanatory text right next to the simulation button that say "Must add at least one platform and a model before adding a simulation"#517Awaiting
56CURATORSylvia MurphyMove ExplanationModelsThere is a nice explanatory text that it will take a while to create a model in the model table. This text is too far to the right and ones eyes are not drawn to it at all.Move the text closer to the button, make it clear and we may want it to same make a model template vice model. That tells the user even a blank form is going to take a bit of time to generate.#538No
57CURATORSylvia MurphyButton MissingSimulations(You can modify these using the "Input Coupling" button on the left hand menu.) list listed under the coupling section but the Input Coupling button does not exist on the left from the simulation pageAdd the button or indicate you have to go back someplace to get it. CLP: This text should read "resolve inputs" button. #504No
58MOHCCMIP5 Model Team (mark)System LogicDiscussionSimulations: Conformancecf(31) - Not one of our team understood the process of creating and resolving inputs to establish conformance, all read the help text and road map material and were completely flumoxed. Some understanding after I demonstrated it, but probably not enough to do themselves without significant hand holdingIs there anything we can do to make this easier to understand without redoing the database structure. In the internal system we use I do everything from within the simulation - define the inputs and link them to a specific model/component in one place.#513Awaiting
59MOHCCMIP5 Model Team (mark)UsabilityModel ComponentsThe information requested for the component attributes is mostly fine - however to verify that every answer is correct will involve review by a lot of people within the Hadley Centre, in the wider Met Office and in partner organisations. The current system has no effective way to review the current content without requiring everyone involved to access the questionnaire and understand how to use it. This is not going to happen.We need a mechanism to provide a per component review sheet which details the attribute, attribute description, enumerated values, and current value. Questions should be grouped as they appear on the questionnaire component page. It should be possible to cut parts of the review sheet text and paste it into an email.#488
60MOHCCMIP5 Model Team (mark)UsabilityFiles/References?In various screens there a drop down menus which list files and references in a seemingly random order. These become very hard to use when the lists get longCould the individual drop downs be sorted on something reasonable - in most situations case-insensitive alphabetical would be fine. Also on the files page could the lists be arranged by case-insensitive alphabetical vertically rather than case sensitive alphabetical horizontally.#514No
61MOHCCMIP5 Model Team (mark)UsabilityDiscussComponent TreeComponent tree doesn't expand properly with the long component names unless an unreadably small font is used - using firefox.Could a different tree component be used which supports wrap around, or just give the tree more real estate.#515
62MOHCCMIP5 Model Team (mark)Form managementReferencesThe form demands a valid hyperlink even if the document is an offline document with no doi:. I can obviously enter a default link to the metoffice site, but I would prefer to leave this link field blank. I notice that some of the 'predefined' references don't have valid links so I'm not sure what the intent is with this.Make the reference link field non-mandatory.#494No
63MOHCCMIP5 Model Team (mark)UsabilitySimulation tabIf you click the simulation tab before you have a platform defined it does nothing - but doesn't tell the user whyReport that platform is required before creating simulation#517
64MOHCCMIP5 Model Team (mark)Form managementCouplings FormLabel on description field is misleading. It says "Describe any per input details if required" - but it is a mandatory field.Change the label or (preferably) make the field non-mandatory#518No
65MOHCCMIP5 Model Team (mark)Form managementFilesOn two occasions I deleted files that I didn't want to delete - clicked the delete link by mistake - entering the file details again can be a long job. Also applies to references, but not so painful to reenterShould there be a 'confirm delete' dialogue to prevent accidental deletion.#510No
66MOHCCMIP5 Model Team (mark)UsabilityConformanceSometimes conformance is achieved by a feature inherent in the model (e.g. we calculate the mineral dust aerosol in the aerosol component - there are no inputs and no coupling to other components. Conversely, sometimes a modelling centre will run a simulation and will not be conformant with one aspect of the experiment requirements (e.g. we don't take account of CO2 emissions from volcanoes), or only partially conformant (e.g. not all species covered).There should be a mechanism to indicate that conformance is achieved through an internal component function, or that the simulation doesn't conform to a specific requirement, or is only partially compliant. The validation report needs to accept that a conformance state has been recorded (i.e. the metadata can be published), but in the case of no (or partial) conformance that this is indicated in the report.#508No
67MOHCCMIP5 Model Team (mark)UsabilityConformanceOur model does not use concentrations as an input - all of our forcing is input through emissions. It is therefore not possible to state conformance with the concentration requirements, and our metadata will not validateI understand that the validation function is going to have some logic to equate conformance with emissions or . ALternatively you could recast the requirements - combining the emissions and concentrations into a single requirement.#344
68MOHCCMIP5 Model Team (mark)UsabilityDiscussionStandardising the responseThe questionnaire response controls many aspects of the response, but it is still possible in certain areas for users to choose how to define their models and forcings. If we want to maximise the standardisation across the descriptions we need a mechanism for users to ask questions and share ideas. We will also have the situation where models with a common heritage (e.g. HadGEM2-ES and HiGEM) will be described by different groups - we need to find ways to ensure that unwanted differences are minimised in the two definitions.As a minimum a questionnaire forum is required. The metafor project will need to play an active role in responding to questions about the questionnaire operation. CLP: Expect questionnaire users of similar moddels to communicate with each other on an ad-hoc basis#520 Standardising descriptions of similar models
#542 Setting up a support forum/email is an action from Telco 57 - Thurs 3rd Dec 2009
69MOHCCMIP5 Model Team (mark)Dependent VariablesInternal Boundary ConditionThis is one example - it occurs in other places. If I select NONE for spatial regridding I don't have the option of selecting NONE for spatial regridding type or leaving it as -------Use dependent drop downs when there are group of characteristics to provide. CLP: This is no longer a problemNoNo
70MOHCCMIP5 Model Team (mark)Accurate Model DefinitionAttributesThere are some situations where I can't unenable an attribute (e.g. permafrost) or respond that we don't do permafrostI understand from the last telecon that this is related to some issue with the bundling of attributes.#521No
71MOHCCMIP5 Model Team (mark)VocabularyEnumeration ListsThere are some rogue enumeration values (e.g. cats and dogs in the precipitation types). In other lists, there appear to be some values missing e.g. land surface energy balanceReview enumeration lists before release.#522No
72MOHCCMIP5 Model Team (mark)Accurate Model DefinitionAttributesThere are some situations where I can't unenable an attribute (e.g. permafrost) or respond that we don't do permafrostI understand from the last telecon that this is related to some issue with the bundling of attributes. I don't mind if we unenable the attribute or have a "not included" option as a response.#521No
73MOHCCMIP5 Model Team (mark)VocabularyAerosolThe enumerated values for aerosols mix both type (e.g. urban) and species (e.g. soot) - we use species in our model, so we would include all the species, but this shouldn't be taken to mean that we don't do urban aerosols for example - we do but via species. How will the scientists reading the metadata understand this.If we include a mixed list like this are we sure that it will be interpreted correctly by the end user. CLP: The lists of Aerosols and Green House Gasses that are associated with the Atmosphere Radiation Scheme is used to describe a coupling. Coupling is covered separately so these lists should be removed#530No
74MOHCCMIP5 Model Team (mark)VocabularyDiscussLand Use TypesThese are defined as biomes. Our model and many others use vegetation type. Its not easy to map between the two as the model dynamically changes the fractions as part of the simulation so the biomes change over time.Our scientists believe that vegetation type is the more common approach to defining land use in GCMs. CLP: Could this issue simply be a missuderstanding about what is meant by "Biome" and "Vegetation type" in the questionnaire?#528Ongoing
75MOHCCMIP5 Model Team (mark)Missing functionalityexportXMLNot dealing with all objects correctly e.g. participants#523No
76MOHCCMIP5 Model Team (mark)Defining InputsFilesOur dump files used for initialisation contain too many variables to define in the file forms.What approach should we use to defining dump variables#500
77MOHCCMIP5 Model Team (mark)UsabilityTabsIf you haven't selected a model or simulation - clicking on these tabs does nothing and confuses users.One approach would be to have the tab link to the summary page if a simulation or model haven't been selected - then when they select a model or simulation it will work as it does now, but there won't be a situation when clicking the tab does nothing.#517
78MOHCCMIP5 Model Team (mark)Error messageSimulationI reduced the number of ensemble members after originally saving it with a higher number. It resulted in the following error message.Request URL:
Exception Type: AssertionError?
Exception Value: Negative indexing is not supported
79MOHCCMIP5 Model Team (mark)Unrecognised informationEnsemble CharacteristicsWhen I selected the drop down for the "member 1 uses" field I was presented with what I took to be initialisation conditions - none of which I had entered. I assume these are from other model entries, or are these some CMIP5 standards that we are not aware of.Either limit the options to ones we define or make it clear that these are common project options#529No
80M. KolasinskiClimpactImprovement(meaning)Summary tab - Introduction partThere should be an image describing the relationships between the main objects in the questionnaire: see Gerard Devine's presentation called CMIP5QuestionnaireRoadmapAlpha7.ppt. This presentation shows that the filling of the questionnaire is not really linear. Then there should be a guide specifying the first things to do (the order might be wrong): create the model and platforms first ; then, filling parties and references ; then, filling components and subcomponents ; then couplings ; then experiments...This is covered by the beta1 roadmap#376Awaiting
81M. KolasinskiClimpactImprovement(meaning)Model tab: Top level componentMeaning of "component attributes" (or "general attributes") for top level components is not clear for me (see also "tickets" 6 and 13 on this wiki pageGive an example of what could be a general component attribute of an oceanic/atmospheric model#497Done
82M. KolasinskiClimpactQuestionDiscussionModel tab: AtmosHorizontalDomain subcomponent for instanceSince it appears that gridspec files won't exist before the filling of the questionnaire, should we add some new attributes ?Add a reference (url) to the files which describe weird grids (ORCA (NEMO) or equal area grids such as WRF) ? Add a reference to software packages which read easily those weird grids (SAXO,NCL for ORCA, NCL for WRF) ?#540No
83M. KolasinskiClimpactQuestionExperiments tabEven though this questionnaire is intended for CMIP5 simulations, will it be possible to describe other types of experiments such as seasonal forecats, weather forecasts ? It might help broaden the community to meteorologists so that they describe their own models (GFS (NCEP), EPS (ECMWF)...)Yes the questionnaire could be configured to support these communities.NoNo
84M. KolasinskiClimpactImprovement(visualisation)Separators for multiple choices for a given fieldThe separator "|" is not really clear (might be taken as a 1 sometimes)"|" could be replaced by " | " (add spaces around it)#539Ongoing
85DKRZTesterContact (model)model: contact and fundingboth contact and funding could be 'UNKNOWN', at least contact MUST be a party with a valid mail||mail of 'contact' should be mandatoryOngoing