The original source of these data is the SSPO website https://www.scottishsalmon.co.uk/ however their archives do not include the full time series of data available.

Sea lice count data - average ovigerous female lice per fish

three different reporting systems cover the data:

December 2010-2012 data was reported averaged over 6 large regions and over variable intervals (specified in sheet)
2013-2017 reporting is of monthly average lice counts over 30 "sea lice reporting areas" defined by the SSPO
2018-2020 reporting is farm level monthly average counts.
Note that from January 2014 the boundary between two Reporting Areas was modified. Farm sites located in Loch Seaforth were reclassified from the Harris Reporting Area to the Lewis East Reporting Area

Area level reporting 2010 – 2012 and 2013 – 2017
Averages represent mean lice per fish for all farms within a defined area for each month. The number of farms in each region that have reported within the month is also provided.  
From the code of good practice:
ii) Where there are more than five pens per site, five fish should be sampled from each of five pens to give a total of 25 fish.
iv) Where a site contains less than five pens all pens should be sampled to give a total of 25 fish. A similar number of fish should be selected from each pen

Raw data to obtain these averages were obtained by the SSPO from all salmon farms in Scotland
Counts are obtained as specified in the Code of Good Practice for Finfish Aquaculture
Counting of sea lice is audited by the Fish Health Inspectors of Marine Scotland Science
Making and recording regular counts are legal obligations on all salmon producers
Raw data is not regularly obtained or recorded by SSPO
Data on numbers of farms in an area was published by SSPO on a quarterly basis, hence is only identical through the quarter

Farm level reporting 2018 - 2020
The original source of these data is the SSPO website > Reports | Scottish Salmon Producers Organisation; however their archives do not include the full time series of data available as reported in Murray et al. 2021 > A preliminary assessment of indirect impacts on aquaculture species health and welfare in Scotland during COVID-19 lockdown. - ScienceDirect.
Sea lice are adult females, presented as monthly averages (per fish). The total numbers of fish on a farm is not reported and so is unknown.
Farms count sea lice every week and are sampled from a minimum of 5 fish per cage with at least 25 fish per farm with at least 5 cages. Figures are reported by calendar month. Therefore, some months include five full weeks and some include four full weeks to cover the 52 weeks of the year.
"Fallow" = the farm was fallow or empty throughout the entire reporting month.
“Farm stocked in Jan.” = the farm was fallow or empty at the start of the month but was restocked during the month.
"Farm fallowed in Jan." = the farm was stocked at the start of the month but became fallow or empty during that month.
“No count –harvesting” = fish were being harvested throughout the month. Under such circumstances it is not always possible to record the lice count.
"Broodstock fish -visual checks" -the fish on the farm are broodstock and they were checked visually without removing them from the water.
"No count -farm recently stocked" -the farm was recently stocked and it may not have been possible or appropriate to record a lice count.
"Monthly mortality (%)"= the percentage of fish lost on each farm within each month. This is reported as a percentage of the total number of fish on the farm each month. This is the measure of mortality used by the Scottish Government’s Fish Health Inspectorate when assessing the health of the fish on the farm.
"Cumulative mortality over full production cycle (%)" = the percentage of fish that have died on a farm during the entire production cycle, given as a percentage of the total number of fish that were initially stocked on the farm. It can only be reported once the entire farm has been fully harvested and fallowed. Cumulative mortality refers to mortality over the entire production cycle. Therefore, an industry-wide figure can only be calculated once all fish from a specific generation or year class have been harvested. It usually takes 2 years to rear a salmon to harvest.
"In production" = the farm is stocked and a cumulative mortality figure will be reported once all fish are harvested.
"Farm fallowed -fish not harvested" = The farm became fallow, but not because the fish were harvested. e.g. the fish may have been moved to another farm