 Welcome to this third video in our series of looking at how we analyse trade-offs in land use decision-making and learning to use AHP as a tool. In this video, we're really going to try and give a practical guide to applying multi-crime decision-making using a case study that we've undertaken in New Zealand. This work was funded by the Ireland and Water National Science Challenge. So really, we have four aims within this video. We want to, first of all, as we said, show an example of how multi-criteria decision-making has been applied to land use in New Zealand. We want to briefly introduce the framework and how the tool works, and at the moment we really are just introducing it. We will go into detail on the framework in later videos. We want to highlight the sorts of results that can be derived, how in a way that it can be used. And then finally, we want to look at some potential advantages and disadvantages of the approach, as we have all tools that we may use for analysis. This has some advantages, but also perhaps there are disadvantages. So it's in the case of thinking about is this the right tool for you. But first, I really want to talk about some context about why we applied it in New Zealand to sort of highlight the sorts of questions that we might look at. So what we find is that New Zealand has had a very successful growth model in the agricultural sector based on traditional farm enterprises. If you look at our exports and earnings over the last few years, generally we've had an upward trend in the value of these. And so we've had a successful growth model really based on what we call traditional dairy beef farming systems. However, this growth has led to some challenges and the OECD and others have noted that really we are beginning to have some severe environmental pressures being placed on New Zealand. We're finding unprecedented levels of water scarcity and quality issues. We have high per capita greenhouse gas emissions and a very high proportion of New Zealand's greenhouse gas emissions come from the agricultural sector. We have threats to our biodiversity. We have many species under threat, and we also have significant erosion. So we can see we have a whole combination of environmental problems, which suggests that our current agricultural systems may not be sustainable into the future. This kind of means we can see, you know, due to things such as climate change and other factors that New Zealand is facing both external and internal challenges to its current model of primary production. And it's been argued that simply carry on like we are, or even making small changes to our systems is not going to be sufficient to enable these challenges to be addressed. So that we have, for example, lots of talk about sustainable intensification, getting more from less, but maybe more around adopting best management practices. However, these generally could be viewed as incremental changes. So what we kind of argue is that this may not be enough, and that actually we maybe need to transform into what we're calling here next generation systems. So I just want to sort of define really what we mean by next generation systems when they look at this. So these will effectively include a redevelopment or redesign of existing enterprises and production systems. They may be wholly new or novel enterprises, or it may be the adoption of new technologies. And really, what we want to see is ways that we can, you know, have value from our land, but really with a much lower environmental footprint. And that's sort of the challenge for our farming systems. But if we are going to transform our farming systems, then we need to realize that risk is really an important component in a transformation. Because adoption of new systems or technologies generally involves some risk to the business. They may be, for example, unproven in the farm situation may involve capital investment, which may mean more borrowing, more risk because you're in debt to the bank, etc. You might have to change your management practice. You may have to change your whole farm system. And so there's a process of learning here. And there's risk associated with all these areas. However, we must also realize they can be the potential to be part of risk management strategies for businesses. Because it may be these new systems actually are less risky or the technologies are less risky. They may improve your profitability. They might reduce your variability in the product that you're producing. They may enable you to meet the environmental regulations or other regulations that have been placed on your business. So we can see that risk can be positive and be negative, but it is an important component in transformation. And one of the questions that was really in our project on next generation systems was the idea whether science can help de-risk the transformation process by providing more information. Now what do we mean by next generation systems in practice? Well, if we can just think about new or different enterprises, we can see that there's a whole range of enterprises that have been discussed in New Zealand. Some of them kind of have been established, but maybe not flourished. Others relatively new. So we talk about things such as manuka honey, dairy goats, dairy sheep, cherries, kiwi fruit, obviously well established now truffles, hemp, etc. But which of these in a sense are appropriate for expansion for farmers to change their systems too? And again, we can see from earlier work that we've undertaken, there's a vast range of novel products or existing products that could expand, that could fill this gap. And generally what we're saying is that we want, you know, in a sense we're looking for systems that can add value or even higher value to our land use, but with a much lighter environmental footprint. But one of the things when we're thinking about how we adopt or whether we adopt these new systems is that we can't pick a winner as we're trying to say. It's not that everybody should go to kiwi fruit, everybody should go to cherries because it's very context specific. And that was the key thing. Different farmers, different systems, there's going to be different choices that are the right choice for them. And for example, there's a whole range of things that might be going on in New Zealand, which mean their context of difference. In some areas in New Zealand, irrigation schemes are being adopted, and that kind of then opens up a whole range of new enterprises. Other areas may be near very sensitive waterways are facing very strong environmental regulation about the use of nitrogen, etc. And they have a slightly different problem they know their pressure is how do they farm within these regulations. And as well we have an expanding Maori agribusiness area. And again, and they're looking to ways that they can make add value to their traditional land systems, and so on. So the kind of thing we're arguing that was we really need to understand the motivations and perceptions of the land manager in order to be able to facilitate this adoption of next generation systems. And this leads us to a number of questions really some, you know, to what extent are these various external incentives and disincentives influencing our land use decision making. What are the key perceptions and motivations of the land manager in determining land land use. How much land managers play how much weight are they placing on these internal and external factors. And our general argument about why we undertook this analysis was that we can understand this better. We have a better chance of understanding what's needed in these what we call next generation systems to help facilitate their adoption, which will then help us tackle some of these big challenges we have in New Zealand. We won't spend too long on this now but effectively we developed a framework for thinking about this. And one of the key issues we were looking at here is how, in a sense, this line in the middle the land manager motivations and perceptions we're saying is all this information coming through all these signals are coming to farmers about change and maybe the need for change, and this new systems available compared to their old systems. And really it's about land manager motivations and perceptions related to the characteristics of their land, what can they land actually do that then, you know, farmers may trial new systems or they may adopt them or not. So the interaction of all these really will determine it. So in our analysis we're really kind of interested in this land manager motivation and perception part of this framework. And so we just highlight that we felt that multi criteria decision making or multi criteria decision analysis was an ideal tool for that and we've highlighted and you'll see in the reading. We voted to develop mental decision support tools and methodologies to address complex decision problem because clearly land use change is a hugely complex decision making this many trade offs that have to be made when deciding how to use your land. And again, we won't spend too long on this now we will look in more detail in later videos about how we identified the criteria that are important for us when we were looking at our multi criteria decision making framework. We found really that we put the sort of decision into six broad categories of what we call domains. We say farmers think about land managers think about financial market, knowledge, regulations, social well being and environmental issues when they think about changing their land use or how they use their land. And within these broad areas. There's also a range of what we call sub domains or sub criteria may be called. So for example, when we think about the financial it might be thinking about how much investment you need, what the overall return on that investment will be how much profitability per hectare you get, how long it takes to pay back, whether that profits going to be variable, whether or not it's going to diversify your income. So we think that fundamentally, you know, high level financials important at the more detailed level these individual components are more detailed. What we do is effectively identify these domains and the sub domains, and then we try and through the process of analytic hierarchical process, a HP, we get farmers to in a sense indicate the weight, the importance of the various components. So this is an example here where how we use it, just thinking about the level of the domain level. So that high level that we were talking about so financial environment knowledge base here. And what effectively analytical high ground for the process does is the pair wise comparison between these issues so here we have financial performance here against environment. And if we think the financial performance of the new system is going to be more important than the environmental footprint, we will move to the left here. As we move further to the left as we will see from these explanations here, it becomes increasingly more important. So as we go from three to five to seven to nine, then financial performance is highlighted more important environment. If we go this way to the right, then environment is seen as more important and financial, and again, we move to the right. There's too much now we will talk more about this actual process in more detail when we look at it specifically. So, for example, this is an actual process that we went through with a land manager, and we can see here for example when they're comparing financial against market. They basically highlights much more weight on the financial side and we can see their choices through this. And through this process, effectively just using mathematical formulas, we can generate an indication of how important each of these are. So in this example, we can see, if we think about the overall decision making being one or 100 or one, we can see that 0.33 of the weight of this would go to financial and knowledge base. So these were the two main areas of importance to this land manager here. And so we can generate these weights of how important these criteria are when land managers are making their decision. And so we went through this approach with quite a number of case studies with farms or management land owners land managers who were considering or had already changed their land use. So there's a range of land managers ranging from small family farmers to larger family farmers to corporate and hill farmers, trustees, etc. And we went through this process with the tool that we developed to look at their motivations and perceptions through this process. So each time you go through the framework you generate this set of weights at both the domain level and that sub domain level those individual criteria that we talked about. And we can just see some sort of results that come out of this. So, on average of all those land managers that we spoke to, this is the weight that was given to those individual domains. And you can see here that of out of one, you know, 0.2 or 20% of the weight importance in the decision making was given to financial factors. And then this average just slightly more than 20% was given to factors relating to the environment. And then we start less for market regulation, etc. It gives an idea of the overall importance of these different factors in their decision making. And we break that down from that high level of domains to what we call the sub domains and see how important it was specific factors were in here. So we look down here, for example, we can see that environmental stewardship has came out particularly strongly. Now, we're not claiming this is representative of all land managers, we were looking at a particular set of land managers with the particular challenges. And so within that group that we looked at environmental stewardship was seen as particularly important. But here again, we see the ability to capture value added, the profit per hectare, etc. So it gives some idea of what the only important factors are on average for this group. But also what we're really kind of interested in is what happens individually. Now, we won't talk too much about the unweighted and unweighted at the moment, but this is just basically again looking at it, but just within each subdomain, within each domain in the subdomain, if we just look at what's important, the match we can get a fill for what's going on. So again, we'll talk about this later. So as I said, it's kind of, you know, interesting began through this process of these individual managers, we can begin to compare what it is that's important to them and this is a graph of the individual results at the domain level. I mean, it looks like spaghetti. But you know, it kind of emphasizes our point, people have very different things that are important to them when they're thinking about land use change. And if we have new systems, then they're going to have to sort of fit with the perceptions of the land managers in order for them likely to be adopted. And we can show again this act the subdomain level as well and the importance of the various components in there. So these types of outcomes, you know, the numbers that come out of using our HP in this way, but of course what's more interesting to us and more use for us is thinking, how is it actually being used. And as I said, we've worked with a number of land managers in very different situations, some have got new irrigation schemes, some under environmental regulation, some trying to work out what to do with relatively un, land is relatively unproductive at the moment can they make it more productive, and so forth. One thing we can do, for example, is you can use this approach to rate a next generation system against the values, the criteria that farmers feel are important. So in here you might, for example, identify a number of steps involved in this might identify a possible next generation system. Identify the criteria as important with the land manager and wait those criteria overall so go through that process that we've been through how important do they think pay that period is how important do they think nitrogen leaching is etc. Well, what we do so we find that was important to the land manager here, but then we can go through a process for that system and think at how well it performs within each of those criteria as well. And then we can look at its performance under each of these criteria against how much importance the land manager places on that to get a waiting of each alternative under each criteria. So in a sense, the closer the system matches the things that are important to the land manager, then the better that system will be. It's a little complicated to think about here, there is reading in the, at the end which we will describe, which highlights this in more detail, but we can show this graphically. Basically, we went through a process looking at two land managers is just to give an example, and sheet dairy dairy sheep as an alternative system. So basically, we went through the process with two land managers who were considering dairy sheep. And here we can see in this blue and these gray lines, how their weights came out this system. And then basically we just talked with just to highlight the process we talked with an expert and went through against all these criteria, how well they thought sheet dairy would actually fit against this criteria. In this system, we just use a sort of a scoring system of one to five, saying if this system fitted that criteria very well so was it quite profitable, then it was, you know, if it's highly profitable relatively profitable it score highly in this area. And so, so basically, everything was rated out of five, and we can see here, once we normalize this, this is the orange is the scoring for sheep. So we can see how well it's scored here and then against how much weight importance was placed on these by land managers. And effectively, we can then by waiting how well the system performs against the criteria by how much weight an individual puts on that criteria, we can develop an overall score for it. So as we said in this we were scoring out from one to five. So the nearer to five that the system scored in relation to our land manager, the better it fit. Okay, we can see here, our overall scores were 3.69 and 3.79, which highlighted that it fitted pretty well for both our land managers. As I said, there's more reading on this but we're just trying to highlight how we can use this tool. So it allows us, and again if we had a whole range of these alternative enterprises, dairy goat and nuka, we could score them all against the criteria and scores weights given by the land managers to see preliminarily which ones fitted better. We can also use it to really give us deeper insights into what's actually driving change. So in this example, really we were looking and again there's reading at the end in the, this is about an irrigation system and about understanding how land managers choose to use that water. And a big issue in New Zealand has been the move to the dairy and some of the environmental pressures with that. So we go through the process and understand in a sense what's pulling people towards dairy and what might be pushing them away. And so we went through a whole load of criteria and looked at this. And also as you go through the process and we'll talk about this more you have a discussion with the land managers, and then you can get into again it's not just a number when they get this weight. There's a reason behind why that number has been given. And what this allows us to do is to really understand what it is the motivations that are going on in that decision making. And understand what's keeping them in dairy or pushing, you know, putting them towards dairy, but you know we can then begin to understand how environment or regulation issues are beginning to change that decision making. And again, there's a reading at the end on this particular case study, which would give much more detail on it. We talked about this idea that we were kind of interested in whether science can help. And so it does give us further insight. So again, by going through this process, we can understand what is important to the decision maker. And then we can think about will new knowledge or information help support the decision maker if they were thinking of changing the system. So for example, we might find, you know, what's important decision maker, if they're in an environmentally highly regulated area, it might be the key driver is how much nitrogen is leached from that land system. Okay, if we have a new system, do we actually know how much nitrogen is being leached. So we may well be the case. Maybe science is needed to show that this new system has low leaching, therefore the farmer, the land manager could be confident in adopting it. So if we don't know the answers, then what science is needed. Okay, so in some cases, as we mentioned there, it might be about or production, you know, maybe it's a new crop, people don't know how to grow it. How suitable it might be for the farm system. As we said before, again, it might be about the environment. What are the greenhouse gas emissions? What's it leaching? It might be actually it's not a production issue, it's a supply chain issue, maybe there's no processing, maybe there's no logistics. Again, is there actually a market for this? If we have research that helps fill these gaps, then we can reduce the risk to the land manager of transforming. Maybe not remove it entirely, but if we reduce the risk, then maybe we're going to speed up this transformation to these more desirable land uses. I just want to emphasize again that, you know, a key thing about the use of the tool and the way we use it is not just those numbers, those weights that came out. But the fact there's an interactive approach using our graphical interface and discussion basically allows a detailed discussion of what's driving them. So we get a really good idea of what the land manager was thinking when they're making their decisions. So it's not just the end number, but the whole process here and then the quotes about the journey rather than the end and I think that's the key thing about it. So the other things you can do with analytical hierarchy process, as we said before, is when you go through the process, effectively it's a pair wise comparison, how important is financial against environment, how important is environment against market, etc. And because it's a mathematical process and you're waiting them, you can begin to work out whether people were consistent in their choices going through the framework. I just want to give you confidence that the results you get are kind of meaningful because they were able to consistently compare the different domains or subdomains that you were looking at. So we were kind of kind of interested to go back after we've used this tool a bit and think about how consistent were they overall, because if they weren't consistent, it might indicate that the tool isn't particularly useful may not work that well in our situations. You are able. So we're thinking about things like you know how consistent were respondents overall, did this consistency change as participants went through the process, or did they get tired because they have to make a lot of comparisons. Were some domains generally more consistent than others, and were some respondents more generally consistent or inconsistent than others. And the question is, is it the final deal breaker here? I don't want to spend too long on this in this and we will come back to how we measure it in the tool later. Basically, as we said, it allows us to get is based on this idea of transitivity. If you say financial is more important than market, and that market is important more than environment, then it should be for you as an individual that financial is more important than environment. And this is based on the process of transitivity. Yeah. And Satie, based on this process I'm a little bit more complicated than that, basically developed an index, these are the index. And it's basically the consistency ratio is from this from naught, which means you're completely consistent in your choices to one where you're completely random. And he himself argues if you have a consistency of about 0.1 on that scale, which means you're pretty consistent, then that's relatively acceptable but there's quite a lot of discussion about it. I just want to look basically about the overall consistency that we achieve now point one is an indicator which is our blue here. So I think we saw generally higher levels of relatively high levels of consistency in our samples, which kind of suggests that the tool works and they are able to use it to make those changes. Now it did vary between domains, and we will talk more about this may have a some bearing on how you construct the different domains and what you put inside them and we will come back to this. Another use and it has been more widely used I think it's tall in this way, if it can help group decision making. Yeah, basically, what we did here is with groups of land managers who had control over lands it might be trustees or others. We went through the process with them individually to see how they scored the definitive criteria, and then through a process of discussion, we then were able to arrive at an overall view. So, we have here in this graph the results for four individuals that we were involved in who is trustees of particular organization and the red line is the group average. So they scored each one individually, and then discussed so what we're trying to argue is the good use of this is for group discussion making and in a later video we will talk much more about how you do this and the advantages and disadvantages of it. But it is a useful tool for thinking with people are working together over land use decision making. Are they thinking the same, you know if one has huge importance on financial up here, that's all the way from here and another has all their way on environment. What are they going to get on when it comes to making decisions, whereas in this group we can see quite a strong correlation relationship between what they were thinking very environmentally motivated and social. Yeah, but that's for this particular group. I just want to finish a brief overview of this by thinking about some advantages and sort of of the framework before considering a little bit some of the pressures. Okay. So what we can do with this is identify criteria that are important in influencing adoption of new systems, and we can draw attention to the believes things are important, we need objective information to support decision making. It can highlight whether potential gaps in our knowledge and where science can help. We can think about how well a particular system fits with land users needs. And even if they in it now will they have to change perhaps, and can think about how new technologies can shift systems. So they better meet the criteria set by land managers. We can use it potentially for decision makers at different levels, maybe council land managers, why the stakeholders to see on whether or not we can get some agreement about what they want to see from land use. There are some challenges with using the framework is a particular methodology. This idea of making pair wise comparisons is time consuming, particularly if you have a number of sub criteria you want to use for example, in our process actually respondents can do 100 comparisons like this. So that can be time consuming. Again, it's not possible to choose all criteria so choice of criteria is very important. Also, we're always arguing that you can trade one off against the other or not have to make a decision in a way, but some decisions might be more binary nature I what we call red lines, you know, it doesn't matter how well it scores on all these other things, it might be around the other criteria. If it doesn't make, for example, this return on investment, then it's not going to go through so we're always trying to argue I guess there's some sort of trade off going on here, but some people may have particular red lines it might be around the environment, it might be around financial, but it might be around supply check. And so that may actually mean that's a no go, even if it scores relatively well around all the other criteria. The process of trading off some people found difficult and as we see in the quote here, some person felt, you know, such a strong link between their social well being and their financial performance, trying to trade them off is difficult. That said, the four, you know, the process says you can stay in the middle in this case and emphasize they are both equally important to you. So there's a little challenge with thinking about making sure that the person you're talking to identifies the criteria in the same way, and how you interpret the graphs, and how you present the results back and we will talk more about this in a later video. So, really what I've just tried to do in this video is present a quick overview of how the framework was being used in the context of New Zealand. Following videos, mainly in module two, we will look more closely at the practicalities of using the framework and how it can be adapted for other uses. And so the things we've talked very briefly about here, consistency, using the tool, etc. We will give much more detail so you can be confident about how you use it. And as I said, there's a number of further reading about this. So, for example, the sheep dairy example we can find in this paper, 2.2A, Next Generation Systems. We're looking at the irrigation system we can find in this one, Central Plains Water, and a more general one thinking about how the framework is used we can find here. So this will give you some more detail and hopefully put in context what I've talked about here. And again, as I was highlighting at the beginning, this was work funded by the Arland and Water National Science Challenge, and hopefully you will enjoy the reading materials that go with this video.