 Linked to objectives are performance measures. These are often times also called indicators, criteria, attributes, metrics. There's a lot of different names for them, but they serve the same purpose. Performance measures are created to measure the achievement of particular objectives. So they're directly linked to the objectives. They're also linked to how we want to think about designing our model output. Because performance measures are used within a decision modeling context and they're the pieces of information that we're assessing for our different options, we need to know and need information that predicts those particular performance measures. Relevant to how we think about performance measures is for a given objective you can have one performance measure or you can have multiple performance measures. For example, to assess water quality you could look at one particular pollutant that you think is key to understand eutrophication problems such as total phosphorus. What you might need to know for certain systems is you might need total phosphorus, total nitrogen, a measure of water clarity and potentially a measure of algal growth. And so having more variables allows you to more fully assess one of the key objectives that you care about. Additionally, when you think about the performance measures, one of the biggest design challenges and choices that one can make is how you represent those performance measures. So looking at our water quality example, you can think about something like total nitrogen in milligrams per liter. You can also think about it in terms of exceedances. So if you have a policy threshold, how many days did a particular water body exceed a total nitrogen threshold? Same objective, performance measure that's trying to achieve similar things but different ways of measuring it. That means that one of the key ways of thinking about performance measures is you really want to choose something that is most relevant to understanding and measuring the objective that you care about. And that means there's three different ways that you can design a performance measure or choose a performance measure. The first way is a natural measure. These are approaches or indicators that really fully capture the objective. So for example, if we wanted to maximize the population of a species, a natural indicator is abundance or the number of species that's in a particular ecosystem. The second approach that you can use to identify or design in a performance measure is by using a proxy. Now a proxy is a performance measure that's highly correlated or reasonable substitute for the objective that we care about. So for example, using the maximizing the population of a species, a proxy indicator for maximizing population of a species could be area of habitat. So it doesn't directly measure the species, but it does give us a sense of what we would anticipate the number of species that could live in that place could be. The third approach we can use to design a performance measure is using a constructed scale. A constructed scale is oftentimes performance measure specific. It can be an aggregate of lots of different ideas and it's used to translate concepts that might be more latent or more abstract into something that's more concrete that we can incorporate into our decision process. So for example, in a number of cases we're interested in the public support for a particular management action. That can be mapped on to a five point scale where what we do is we describe what level of public support is at each of those levels. So for example, a level of two might indicate you have a tremendous amount of public support. Most of the stakeholders have been fully engaged and are supportive of the particular management actions. One is taking zero might indicate folks are indifferent. Whereas a negative two might indicate that you have people who are actively working against action for that particular site. So one could construct a scale where there's no natural measure or proxy that can serve as a reasonable approach to developing a performance measure to really make sure that you're capturing that objective and measuring in a way that is meaningful.