add Hvilsom et al., 2022 reference

docs/4_Species_list/Species_list.md

@@ -22,9 +22,11 @@ Generally, we are assessing indicators at the species level. However, taxonomic
 
 ## Creating species lists 
 
+The species list could be selected specifically for monitoring genetic diversity, or be a subset of other lists that countries already use for monitoring or conservation priority-setting. [Hvilsom et al., (2022)](https://www.iucn.org/resources/publication/selecting-species-and-populations-monitoring-genetic-diversity) provide feedback on identifying the most appropriate set of species or populations for monitoring genetic diversity.
+
 Each country team should begin by making a list of **at least 100 candidate species** for which there is likely some data for the indicators. *‘Likely to have some data’* means that the species are not recently discovered, poorly known, part of a group with highly uncertain taxonomy, etc. This candidate list becomes the list of species to try to collect data for. Some proportion of these species will not have sufficient data; this is normal and expected- in a trial of indicators in nine countries, around 17% of species did not have sufficient data for at least one indicator.
 
-**It is vital to document how the list is developed in order to identify any biases (e.g. mostly common species).** *It is not necessary for all chosen species to have high quality data across their range.* While indicators would be more accurate if all species have data for all populations, complete population data may only be available rarely. *It is ok if data are available for only one of the two indicators or for only some populations of a species* (as explained under Common Issues below). Moreover, upon investigation, species initially deemed likely to have some data, may actually have insufficient data to calculate either indicator. Species should not be removed from the list after the initial list is made. We will calculate the indicators with and without various types and levels of missing data.
+**It is vital to document how the list is developed in order to identify any biases (e.g. mostly common species).** It is not necessary for all chosen species to have high quality data across their range. While indicators would be more accurate if all species have data for all populations, complete population data may only be available rarely. It is ok if data are available for only one of the two indicators or for only some populations of a species (as explained under Common Issues below). Moreover, upon investigation, species initially deemed likely to have some data, may actually have insufficient data to calculate either indicator. Species should not be removed from the list after the initial list is made. We will calculate the indicators with and without various types and levels of missing data.
 
 There are some species where it will be particularly hard or impossible to quantify the Ne 500 and Populations Maintained indicators, and they should be excluded from the species list. For example, evaluation of the Ne>500 indicator will be hard in species where natural subpopulations are typically very large and/or hard to measure, such as microcrustaceans, many insects, some fungi, some plants with deep soil seed banks (where all ‘individuals’ cannot be counted). Populations of such species can also grow in a short amount of time to very large numbers and have high standing genetic variation (Chaturvedi et al. 2021). We advise not attempting to include such species in a country’s first evaluation of these indicators due to difficulty in finding and interpreting data.