Unit of shelterVolume metric?

[ConnectedSystems]

Working on #97, I'm getting what seems to be very large values for ~Evenness~ shelterVolume, maximum value of 339.3470:

~When using default values, the metric appears to vary between 0 and 1.~

[ConnectedSystems]

I think I understand - shelter volume is in cubic meter per hectare, but the RCI function expects values between 0 and 1.

@Rosejoycrocker or @ryanheneghan would you have the time to help me hash out a good approach to deal with this case?

[KRNA01]

Please check that we are using proportional coral covers for Evenness, and not (1) absolute cover nor (2) colony numbers

Get Outlook for iOShttps://aka.ms/o0ukef

---

From: Takuya Iwanaga @.> Sent: Wednesday, February 16, 2022 10:33:03 AM To: open-AIMS/ADRIA_repo @.> Cc: Ken Anthony @.>; Assign @.> Subject: Re: [open-AIMS/ADRIA_repo] Unit of Evenness metric? (Issue #99)

Assigned #99https://github.com/open-AIMS/ADRIA_repo/issues/99 to @KRNA01https://github.com/KRNA01.

— Reply to this email directly, view it on GitHubhttps://github.com/open-AIMS/ADRIA_repo/issues/99#event-6079111344, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AVYCFW36RTAJ2VUEGLOTF3LU3LWD7ANCNFSM5OQDDYUQ. Triage notifications on the go with GitHub Mobile for iOShttps://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Androidhttps://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub. You are receiving this because you were assigned.Message ID: @.***>

[ConnectedSystems]

Hi @KRNA01 - forgot to state explicitly that my initial report incorrectly referred to evenness. This issue is actually with shelter volume.

[Rosejoycrocker]

Hi @ConnectedSystems , could we use area to make it a proportion? Say (SV_per_ha(site i)*area(site i))/sum_i(area(site i))

[Rosejoycrocker]

sorry that should be (SV_per_ha(site i)area(site i))/sum_i(SV_per_ha(site_i)area(site i))

[KRNA01]

I see. As shelter volume is in cubic meters per hectare, numbers higher than 300 m3 per ha are indeed possible. This translates to a volume of on average 3cm above the substrate.

Get Outlook for iOShttps://aka.ms/o0ukef

---

From: Takuya Iwanaga @.> Sent: Wednesday, February 16, 2022 11:53:13 AM To: open-AIMS/ADRIA_repo @.> Cc: Ken Anthony @.>; Mention @.> Subject: Re: [open-AIMS/ADRIA_repo] Unit of shelterVolume metric? (Issue #99)

Hi @KRNA01https://github.com/KRNA01 - forgot to state explicitly that my initial report incorrectly referred to evenness. This issue is actually with shelter volume.

— Reply to this email directly, view it on GitHubhttps://github.com/open-AIMS/ADRIA_repo/issues/99#issuecomment-1041010316, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AVYCFW7KL7BOSS3EQJJGFH3U3L7QTANCNFSM5OQDDYUQ. Triage notifications on the go with GitHub Mobile for iOShttps://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Androidhttps://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub. You are receiving this because you were mentioned.Message ID: @.***>

[ConnectedSystems]

@Rosejoycrocker if I understand correctly (excuse the clumsy notation, github doesn't support latex natively and I didn't want to faff about trying to work out how to get summation displaying properly - hopefully you'll get what I mean):

where:

• S = set of sites
• sv is the shelter volume for the given site
• i indicates the specific site
• a is area in whatever unit

It would give ~a study area wide value~ site specific value relative to the total possible shelter volume per site, am I understanding correctly? ~(a study area wide value is probably something we want to aggregate to later but here I'm trying to get a per site metric).~

I guess we could get total possible shelter volume by running the k values through shelterVolume() (acknowledging that this will be a blanket 70% for the Whitsunday case).

We'd also have to be careful about what unit a is in. I think Yves-Marie said his data is in km^2, whereas Moore is in m^2.

As an aside we should pre-process everything that goes into ADRIA into expected units...

[Rosejoycrocker]

Hey @ConnectedSystems , I can't view the equation for some reason. Maybe an attached screen shot?

[ConnectedSystems]

Well, it was working for a few mins, and then it stopped working ☹️

[Rosejoycrocker]

ooh wait its working now, maybe my internet?

[Rosejoycrocker]

Yes the equation is basically what I'm suggesting

[ConnectedSystems]

No, I replaced it with a screenshot like you suggested. Seems GitHub mangles the URL or something for the latex rendering service I used.

[Rosejoycrocker]

oh right, weird

[ConnectedSystems]

Could we make it agnostic to spatial scale somehow, or do we have to pre-process site data to give area in hectares only (or meters only?)

[Rosejoycrocker]

I guess we could do a straight SV_i/(sum_i(SV_i))

[ConnectedSystems]

That would be misleading though if all sites were doing poorly, no?

[Rosejoycrocker]

Yeah you're probably right... @KRNA01 or @ryanheneghan would probably know more about the significance of the spatial scale

[ConnectedSystems]

I think you're right, we'll need to bring in area (carefully)

[Rosejoycrocker]

I guess so long as SV and area have the same area units (e.g. ha and per ha) you could generalise, but I'm not sure if that's a reasonable assumption

[KRNA01]

Hi Both, Apologies for being offline. Is this sorted or do you need input?

[ConnectedSystems]

Hi @KRNA01 I think we'll need to discuss what the best course of action is.

It seems we do need to bring in area, but

1. we need to be careful the spatial units remain consistent throughout, which probably means preprocessing to make everything into m^2 or km^2 or hectares
2. the current process for calculating metrics does not bring in site/reef data so that needs to be adjusted too as we'll need area

Unless there is a straightforward way of doing this without considering area? (e.g., SV is taken to range between 0 to > 300, where anything >= 300 is mapped to 1.0, and we do some interpolation of values to fit it between 0 and 1?)

[KRNA01]

Agree, a failure of mine by not having done this properly previously. I think for shelter volume, I assumed sites had one hectare of coral real estate. We have now gone more sophisticated with IPMF areas. Suggest we set up a new branch and I go through all steps to include site area properly. This includes recruitment. Do you agree?

Get Outlook for iOShttps://aka.ms/o0ukef

---

From: Takuya Iwanaga @.> Sent: Wednesday, February 16, 2022 2:59:55 PM To: open-AIMS/ADRIA_repo @.> Cc: Ken Anthony @.>; Mention @.> Subject: Re: [open-AIMS/ADRIA_repo] Unit of shelterVolume metric? (Issue #99)

Hi @KRNA01https://github.com/KRNA01 I think we'll need to discuss what the best course of action is.

It seems we do need to bring in area, but

1. we need to be careful the spatial units remain consistent throughout, which probably means preprocessing to make everything into m^2 or km^2 or hectares
2. the current process for calculating metrics does not bring in site/reef data so that needs to be adjusted too as we'll need area

Unless there is a straightforward way of doing this without considering area? (e.g., SV is taken to range between 0 to > 300, where > 300 is mapped to 1.0, and we do some interpolation of values to fit it between 0 and 1?)

— Reply to this email directly, view it on GitHubhttps://github.com/open-AIMS/ADRIA_repo/issues/99#issuecomment-1041109435, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AVYCFWZEOZTWHJGQXJAEP23U3MVMXANCNFSM5OQDDYUQ. Triage notifications on the go with GitHub Mobile for iOShttps://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Androidhttps://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub. You are receiving this because you were mentioned.Message ID: @.***>

[ConnectedSystems]

Okay - hopefully we can squeeze this in tomorrow and wrap up in time for me to set up some runs before end of day 😄

[KRNA01]

Good idea - I'll make it my early morning gig Which branch should I build a new branch from?

[ConnectedSystems]

Just pushed my work-in-progress to summary-metric-stats, please use branch off that one 😺