EuropaBON WP3.1

You're not logged. You can see and visualize the entered data but you can't modify or create it.
If you want to suggest changes or enter missing initiatives, you can contact the WP3.1 coordinators and you will receive your login details.

Data Stream - modify

Tab created: 2021-10-25

Project

1) Data origin - from where is data coming from

Origin integration node

2) Data process - how is the data processed (if)? And how does it flow?

Data type data_stream_datatype_desc

Ebv ess type data_stream_ebv_ess_type_desc

Format data_process_format_desc

Method data_process_method_desc

Frequency data_process_frequency_desc

3) Data set - which dataset/integrated product is processed and how?

Name data_set_name_desc

Description

The basic observational units, the minke whale tracks, from the two platforms A and B were compared for
matching by an automatic routine (duplicate identification rule) that has as its criterion difference in timing,
bearing and radial distances (Schweder et al. 1997, Skaug et al. 2004, Bøthun & Skaug 2009). Before the
matching, missing values of radial distance and and/or angle are imputed by interpolation between adjacent
surfacings and taking into consideration the movement of the vessel. An initial sighting has three possible
outcomes which are: seen by either platform or by both simultaneously. If one platform detects the whale before
the other platform, it sets up Bernoulli trials where the outcome is seen or not seen by the other platform.
A hazard probability model is developed as described in Skaug et al. (2004), where parameters are estimated by
maximizing the likelihood based on the observed data. The simulation part of the earlier estimation processes
(Skaug et al. 2004, Bøthun et al. 2009) which took care of bias correction for measurement errors, duplicate
identification errors, clustering and other factors have not been performed on the present data. Instead other
approaches have been chosen. After investigating measurement errors from the experiments during the 2008-
2013 survey cycle (Solvang et al. 2015) we decided to use the recorded data uncorrected. Clustering has been
taken care of through the variance estimation process (Skaug & Solvang 2015).
For the chosen covariate model, the parameter estimates were used for calculating the effective strip half
widths wA and wB . These are in turn used to obtain an abundance estimate (by survey block)
2( )
A B
A B
n n N Area
w wL
+ = +
where A n and B n are the total number of sighted whales from platforms A and B, L is the realized transect
length, and Area is the area of the survey block. 
3
The quantities wA and wB are obtained from the fitted hazard probability model, which is parameterized using
a GLM approach as follows (see Skaug et al. 2004 for details): The radial distance at which the hazard
probability has dropped by 50% is exp(ηr), where ηr is a linear predictor. The linear predictor consists of the
intercept βr and covariate effects. Similarly, there is a linear predictor associated with the effect of sighting angle
(inctercept βθ). The hazard probability at the origin (r = 0) is parameterized as μ = [1+exp(-βμ)]-1
.
The hazard probability model involves one additional parameter, the surfacing rate intensity α, which is
determined from external data. For that purpose we used dive time data collected by radio-tagging of 20 minke
whales (Øien et al. 2009) over the period 2001-2008. The mean surfacing rate α was estimated from those data,
and where sea state information was available, truncated for Beaufort > 4. The estimate is 45.78 blows/hour,
which gives the parameter α = 0.0127.
The variance of the abundance estimates have been calculated using a new approach (Skaug & Solvang 2015).
The inter-annual variation in spatial distribution (additional variance) of the minke whales has been included as
in earlier estimates (Skaug et al. 2004, Skaug & Solvang 2015).

data_set_description_desc

Target biodiversity data bio_data_target_bio_data_desc

Taxonomy

Target taxa data_set_target_taxa_desc

data_stream_taxa_subset_desc

Target species bio_data_target_species_desc

Spatial scale

Spatial scale data_set_spatial_scale_desc

Resolution data_set_resolution_desc

Resolution units data_set_resolution_units_desc

4) Data destination - where does the data/product flow to?

Two options here:

a) The data flows into an upper-level integration node b) The data is a generated Product

Select the destination integration node

Dest integration node

Is the product generated used by policy - e.g. EBV, EESV, indicator?

Used by policy data_stream_user_by_policy_desc

Note here that some indicators might currently lack connection to policy but we want to capture them here too because of their potential to inform policy in the future.