 Understanding the landscape requires a recognition of both the spatial characteristics of landscapes and the temporal characteristics of landscapes. As we can see in these pictures, our human populations have extensive experience with nature and with the dynamics of the environment. So understanding these processes is essential to understanding how our landscape works. One of the first concepts that is important in managing the landscape is to recognize the relationship between area and the number of species. This is a typical species area curve in which we see increasing numbers of species as the area of consideration becomes larger and larger. This is also important in terms of managing habitat types because if we only manage for very small patches, we will only protect small numbers of species. But if we manage for increasing area of habitat, we will be able to manage or protect increasing numbers of species. This is a typical terrestrial perspective, but in aquatic systems we see a slightly different relationship. Instead of a species area curve, we might want to think of it as a species distance curve in which we notice a predictable change in the number of species as we move downstream. In this example from the Willamette River, we see an increase in the number of species as we move from headwater streams downstream to the mouth of the Willamette River in Portland. We notice an increase in the number of native species as we go from the mountains down to the valley and then abruptly as we reach the valley floor streams, we see an increase in the warm water introduced species, the exotic species, resulting in an even greater increase in the number of species as we move from headwater streams to large river. The landscapes that we manage, whether we're dealing in forested landscapes or moving down from the forest into the agricultural valleys, we see an intensely fragmented landscape. We commonly think about the distribution of patches and refuges across this landscape and the implication for species diversity. But we also see not only habitat fragmentation but policy fragmentation as we move across this landscape. As an example of policy fragmentation as we move from the headwater streams in the Willamette Basin down to the lowland systems, we see a change in policy of fragmentation of policy as well as landscape. In the public forest lands, we have riparian reserves that are two site potential tree heights in width, roughly 400 feet in width on either side of the stream with no harvest and all flood plains protected. As we move down from the public forest lands to the private forest lands, we see required buffer strips. Oregon passed a Forest Practices Act in 1972 and by the late 1970s was requiring buffers and now those buffers are as much as 100 feet wide and allowing partial harvest within them. When we cross the fence from private forest land into private agricultural lands, buffers are no longer required. Voluntary farm plans are used to try to provide for some ecological functions but basically a farmer literally can plow down to the stream's edge legally. Cows can be in the middle of the creek, livestock can be in the middle of the creek and it's still be legal but it's obviously a different riparian policy than we saw on the forest lands. We can move off of the agricultural lands onto the residential lands and a private residential landowner can cut down the riparian vegetation, could plant petunias and roses right next to the stream and it would still all be legal. And then we move down even further into the industrial and urban lands and you can basically pave the stream and create a concrete stream. So all of this on the same landscape, we've got fragmented policy, everything from extensive protection on the public forest lands to essentially no protection and extreme modification on the urban lands. And this creates a fragmented landscape and it's a result of fragmented policy. As we try to create complex and rich habitats across the landscape, recognizing the importance of habitat complexity is essential. It makes a lot of sense to most people as you see a scene like this in which you have a complex habitat with many different depths and velocities, large refuges from disturbances, that it provides a great deal of ecological complexity that you would not see in a straightened and simplified channel system. We've looked at streams in the coast range and the cascades to see how fish respond to this kind of habitat complexity. Here's a graph of cutthroat trout in a coastal stream in which we added wood in different levels. Our reference condition is no wood. It had been modified by past land use practices and then low amounts of wood were just two large logs in every pool. The medium level of complexity were about three large logs and five alder tops. And then finally the high complexity was five large logs in every pool and five alder tops. And so we see that the cutthroat trout in three different years, the three different colors you see here are each of three different years, consistently showed higher abundances of trout with increasing complexity. Many people in trying to manage for habitat complexity are concerned about the dynamics of stream, these practices that we have seen here with cabling of large wood, people placing large wood into a stream to provide for natural ecological function, but being worried about it being dynamic. That really seems to be a contradiction. If you're trying to create dynamic and complex habitat but you put in the habitat features to try to keep it from being dynamic, it seems like you're working at opposite purposes. So this type of bondage of holding wood down and trying to keep it from moving seems like it might guarantee the project that you're putting in, but it's not providing the function because in a flood wood moves, it floats. We've seen in natural streams in Ogroth Forest large complexes of wood floating up a meter or more during a flood dropping right back down in place. If you bind it with cables, it can't move and the wood basically explodes during the flood as it tries to float, but it can't. So this type of trying to create a monument to our good intentions that never changes is not consistent. And even if you put this wood in the perfect place, if you've got the world's best geomorphologist and you've got the world's best hydrologist and best fish squeezer to put that wood in the perfect position for that habitat, that doesn't guarantee the stream's not going to change. As a matter of fact, I'll guarantee you the stream will change. It will build up, it will cut down, it will move left, it will move right, and where the wood might be in a great place today, 10 years in the future may not be a good place. And so restoration is actually more placing wood in a stream so that the stream can use it effectively to create appropriate habitats than it is to build these monuments of what we think good habitat represents. An important part of any kind of ecosystem, stream, river, lakes, or terrestrial is the vegetation and not just the large trees, but also the shrubs and herbaceous vegetation, the macrophytes that create very important habitats. In agricultural lands, these types of riparian forests have been greatly modified and converted. And quite often we only see a small string of riparian buffers left behind in that agricultural landscape. When we looked at riverine systems in the Willamette River system and compared three types of habitat, the light blue that you see there is called a single channel. That's just a straight single channel. The magenta is purple, the magenta is multiple channels that have braided or complex off-channel habitats. And then finally the green are tributary reaches of the main stem Willamette River, where a tributary junction comes in creating complex habitats of that smaller tributary and the main stem. When we looked at the richness or the number of patches of riparian forest from the headwaters up in the McKinsey, downstream along major tributary junctions till we reached the mouth in Portland, we found that in almost all cases the multiple channel reaches and the tributary junctions contained a greater number of patches than the straight single channels. And so in a river system the type of river system that we have makes a big difference about the richness of riparian forest. And many of our land use practices have resulted in the simplification of our channels, eliminating those off-channels and side channels and trying to armor and anchor the tributary junctions so that they don't move and are not dynamic. And in doing so we lower the richness of the riparian forests. These riparian forests provide more than just the complexity of the floodplain habitat and the differing depths. They also provide large wood. Large wood is an important component of aquatic habitat and we see quite a few fish species that are associated with this wood habitat in the Willamette River here. We looked at seven pairs of sites in the light green color, the unaltered forest, and then in the dark blue, riparian forests that have been altered to agricultural land, residential land, or urban land. And in the upstream reaches we found much higher volumes of wood in the unaltered or intact riparian forest than we did in the altered riparian forest. In this upriver section we have a complex braided and shallow habitat and as a result when we compare the density of wood versus the percent of the tree cover on the adjacent stream banks we see a positive relationship and being able to explain about 25% of the variation in amounts of wood simply by the nature of the riparian forest on the flood plain. So in these shallow upriver habitats the nature of the riparian forest is an important determinant of the amount of wood that's present in the aquatic habitat. In the same seven pairs of sites we found an average of four more fish species in reaches where the riparian forest was intact as compared to adjacent riparian areas that had been converted to agricultural land or to urban and residential land. And so indicating that it does make a difference whether you have intact riparian forest even in these large rivers. And a lot of that is due to the influence of the riparian forest on the habitat, on flood refuge, and on the complexity of wood habitat in those large river systems. Now, one of the goals of land management and landscape analysis is to understand how the landscape works. One of the important things that we can do in understanding the landscape is to reconstruct the historical landscape. This isn't to suggest for one second that we're trying to move back to previous landscapes. It's actually looking at those historical landscapes to understand how these ecosystems function and to design the future landscape that we want for our children and future generations. And so how do we go about looking at historical reconstruction of landscapes? One of the things we have to consider are the critical ecological processes. What are we trying to restore? And so trying to restore to the degree of understanding of things like natural disturbance events, like floods or fires, to understand processes like natural nutrient dynamics and plant succession. And so any of the evidence that we have that allows us to reconstruct those ecological processes will be extremely valuable in understanding how this landscape operates. Now, how do we do that? One of the ways we do that is through historical sources. And this is the place where communities can really step up and help because everyone has these shoeboxes of old photographs buried away, old written accounts, newspapers from the turn of the century or earlier that describe the conditions of our landscape. And so piecing together these different sources, whether it's photographs, written accounts, oral tradition, newspapers, other forms, can all be extremely valuable in piecing together this landscape. But be very, very careful. Any of these sources is biased. Anyone that went out to take a photograph wanted to capture a certain scene. So that means that they took a picture of only part of the landscape. You do that every time you go out with a camera. Everyone that went out to paint the landscape was trying to convey something, and they were biased. They wanted to paint a certain scene and convey a certain feeling. Someone who wrote a story wanted to convey some feeling or perception of that landscape in their writings. A good example of this is the confluence of the Mississippi and the Missouri River system near St. Louis. This was described by Charles Dickens and when he came over to the United States as this dark wasteland, this wet swamp land that was infested with disease, he was just coming out of Europe that had problems with disease, and so he saw this as a very negative thing. Mark Twain described that same piece of landscape as complex and wild and diverse, and he was coming at it from the perspective of the American frontier, as this is a new and rich place to be tamed. And so it depends upon your bias as to what you describe, so be very careful. Credibility. All of the sources of historical information are not equally credible. Some sources were very accurate and they paid great attention to detail. Others were very sloppy and were notoriously loose with the information. In fact, in some of the records of even surveyors, we found that surveying records were faked because the weather was bad and it was easier to sit inside a tent than it was to get out in the harsh weather and do a survey. And so you have to know your sources and their credibility. And then finally, accuracy. Take a look at what methods were being used and how accurate they could possibly be. Now, there's a tendency for us to think that everything that is modern is very accurate and everything that is old is inaccurate. Quite often that's not the case. Some of our old survey records are extremely accurate because people took out surveying equipment and actually surveyed the streams and rivers and plant communities. Some of the modern surveys for this material are very coarse because of the huge expense and the cost of people doing it these days. We take shortcuts and sometimes the resolution and accuracy on modern surveys is less than the historical ones. So anytime you take a look at historical landscapes, you should be very, very skeptical of those reconstructions. But at the same time, it provides an invaluable context for understanding the current landscape and where we might go in the future. An example here is seen in this painting by Paul Cain of the Willamette River in 1848. When I first looked at this painting, it surprised me because the historical reconstructions that had been done had described the Willamette River as a complex network of braided channels. But this work had been done in the upriver portion between Corvallis and Eugene. In this downriver portion, where the artist described it as being near the city of Oregon City, the channel is much simpler and it's in a geologic trench, a basaltic trench, and was historically more simple. But this painter was also notorious for modifying his paintings to convey what he wanted to. And so we probably see the riparian forest set back away from the river just to aid him in his illustration of the river system in this painting. So you need to be skeptical about any of these historical sources, particularly paintings that may be more artistic than accurate. I'd like to take a look at this section of the Willamette River that you just saw and the rest of the Willamette River. The Willamette River runs north from the Cascade Mountains and the coast range from Eugene north to Portland, where it enters the Columbia River and then flows on down to the Pacific Ocean at Astoria. So we're going to look at this portion of the Willamette River, this main stem portion right now that flows from Eugene north to Portland. This upriver portion was in a very flat depositional reach of the Willamette River and had very complex channels. On the far left, in 1850s, we see a very complex braided river system. And then we look at three other dates, 1895, 1932, and the present. And we see that the river becomes simpler going from a complex braided river system to a simpler and simpler river system as we closed off side channels, revetted the banks, and simplified the river system and channelized the river to create the simple pipeline that we see in today's Willamette River system. With that, the river lost tremendous off-channel habitats and complex features that made it a very productive and rich environment. Floods are nothing new. This is a picture of Portland, downtown Portland, in the floods of 1861. This was an amazing year. In the fall of 1870, it started raining, and it rained, and it rained, and it rained. And in December, the Willamette Valley was covered with floodwaters. And then as soon as those floodwaters receded right after the first of the year, it started to snow. And so the Willamette Valley was covered with a blanket of snow, and the river froze solid. And they actually used it for transportation wagons moving up and down the river. And then in the spring, it thawed and resulted in this flood that you see in this photograph. And so the settlers at that time experienced the flood of the fall. Many of them got washed out by that, made due with what they could in the snow and the cold, and then suddenly it was flooding again in the spring. The fact that they could even stay here was amazing. So we have these strong experiences that you would think would make a lasting impression, but yet our communities keep moving in tighter and tighter, trying to encroach on the river systems, use that available land, and trying to keep it from being quite as dynamic. Here we see a photograph of the Willamette River system in the 1996 flood, and the little picket fence of cottonwoods and ash that you see along the river there in the middle of the floodwaters is what we call the Willamette River Greenway, a protection program that was set up to try to conserve the existing floodplain forests of the Willamette River. And so you can see that this Greenway, a protection program, is protecting a small fraction of what was the Willamette River system and the Willamette River's forests. So we took a look at all of the area within the Willamette River system that had been inundated under historical floods. And so in the upriver portion, we see this very broad section where much of the floodplain was inundated and was several miles across during historical floods. An intermediate area bouncing between that as the river flows north, and then finally a very narrow constrained section as it flows into Portland. To study this, we wanted to follow the change in the river historically. Now just chasing river channels doesn't work at that point because the river is constantly moving and changing. So what we used was the floodplain as our axis of study. And much of the data that you will see after this is based on this analysis of the entire Willamette River floodplain from Eugene north to Portland. And we took that floodplain axis and we cut it into one kilometer slices. And we analyzed the channel characteristics and the floodplain forest characteristics on these one kilometer slices or bands along the axis of the floodplain. And so the river could change. It could flip. It could become more complex or less complex. And we were still analyzing the same slice out of that floodplain. Here is an example of how the river system has changed. In the dark blue, you see the river in 1995. In red, you see how complex the same reach of river was in 1850. And so this very complex braided river system has become extremely simplified and narrowed into this narrow single channel. This is a projection of what this river system could look like in 2050 if we allow the channel simplification to continue. Those little scattered dots that you see along the edge of the blue channel are areas, these little fragments here and here and here, are areas that will be lost over the next 50 years if we continue to simplify channels at the rate that we've been simplifying them this century. In contrast, we can take an alternative route. We can start to reconnect existing historical side channels that have been cut off from the main channel. And in doing so, if we take existing topography and existing channels, this represents the numbers of channels that we could add or reconnect to the Willamette River system over the next 50 years. So representing a conservation scenario where we could plausibly reconnect a meandering and complex river system, we took a look at the entire river system from Eugene on the far right, downstream to Portland on the far left. What you see in dark blue is the complexity of the channel in 1850 and the light blue is the complexity of the channel in 1995. The complexity of the channel is measured as the number of or the length of channels per one kilometer band of the flood plain. And you'll notice that in 1850, in some places we had as many as 11 kilometers of channel in a one kilometer band of flood plain. And quite typically, we had anywhere from around six to eight kilometers of side channel in a one kilometer band of flood plain. And now that has been reduced quite markedly. So if we take the difference between these two curves, we see the net change in the Willamette River system between 1850 and 1995. So all of the river didn't respond the same way. Some portions that were more depositional and complex have been simplified. Other areas that were very hardened naturally and geologically remain much the same. So we have to be careful in historical reconstructions and landscape interpretation to make sure that we don't treat all the landscape identically. In addition to looking at channel complexity, we wanted to look at the extent of flood plain forest between 1850 and the present. Luckily in the general land office surveys of 1850, they had surveyed the amount of forest on the flood plain. And we were able to use these historical survey maps where people actually got out with survey lines and changed and measured the channels and measured the vegetation to reconstruct the vegetation of the Willamette Valley in 1850. And from that, we sliced out the riparian component that you see running up the middle of the Willamette Valley. And so you can see the gallery forest in the middle of that photograph or of our maps. And here is an example, a close-up of this middle section of the Willamette River Valley. And you see the green around the river, which is the gallery forest of cotton woods and Oregon ash, bigleaf maple, white alder and willows with scattered conifers throughout it. The orange that you see are the oak savannas and then the yellow are the grass prairies. So even though the prairies and the oak savannas were a major part of this valley in the 1800s, the major flood plain portion of this habitat was dominated by wooded forest. As we move toward Portland, we see that shifting to even more coniferous forest. And so the mouth of the Willamette River, as it moved into the Columbia River, was heavily forested at the time of settlement. If we take this change in forest, just as we did for Channel Complexity, we see that the pink represents the area of flood plain forest in terms of hectares of forest per one kilometer band of flood plain in 1850 compared to 1990. And we have tremendous differences between 1850 and current conditions. If we take a look at that difference along the river system, we see that throughout the river, from Eugene to Portland, throughout its length, we've lost riparian forest cover in all sections of the river. Every flood plain slice observed a decrease in the extent of flood plain forest. So that means in terms of reconstructing how this river worked, we know that it was a more complex channel, and we know that it had complex flood plain forest that has been greatly altered by land use practices such as agricultural conversion of those lands to agricultural crop lands and pasture and grasslands, conversion to residential lands, conversion to industrial and urban lands. And so now we find that a large portion of this has been changed. An example comes from this analysis where we took a 120-meter band on both sides of the river in under current conditions, under 1850 conditions, and in proposed future conditions. And so we see in this example from 1850 that if we take that 120 meters on both sides of the main channel of the Willamette River, that essentially 65% of it was in hardwoods. If we add these components of the mixed forest and the coniferous forest, we find that about 75% to 85% of the riparian area was in wooded forest. Now in 1990, we see that the agricultural lands and the urban lands, so crop lands, or grass fields, or pastures, or urban lands and residential areas occupy somewhere around 60% of the riparian area. And only about 25% or less of the riparian area is wooded, so we've gone from 85% of the riparian area being in woodlands down to only about 25% or less being in woodlands. And so we've greatly altered that riparian conditions. But what could it look like in the future? We worked with stakeholders to define plausible futures in 2050. The Willamette Basin is projected to increase its population in 2050, and by using what those stakeholders felt to be plausible scenarios of conservation measures. This would be conservation easements, programs like the CREP program for leasing the land and replanting, things like the Wetland Reserve Program and other types of conservation measures that might be possible in the future, we see that we were able to increase that wooded riparian forest land from what was about 25% up to almost 40% of the riparian area being in wooded forest in the next 50 years. We were able to contain the urban development within this area, and we were able to convert some of the agricultural lands into this wooded forest lands. Now for the agricultural landowner, that does represent a transformation of lands that are adjacent to the river, being flooded and in production, into wooded lands. Some of this wooded land could be cottonwood plantation that provides a commodity for the landowner, and some of it simply protecting their lands from the influences of flooding and other disturbances. It's still different than all of this going into increasing urbanization, which results in a totally different landscape for that urban and agricultural landowner. The river system has also been modified. Many landowners along rivers are often concerned about the dynamic processes of channel movement. This is a natural process in rivers. Rivers erode and deposit. That's the way they keep themselves alive and shape their habitat. Erosion is a natural process, and rivers are dynamic. Humans, as they move close to rivers, are moving into probably the most dynamic piece of the landscape. So how do we start to live next to a river and maintain channel dynamics without destroying the river? Historically, we've gone into sections like this side channel on the Willamette River system, and we put in pilings to close off that side channel. You can see in this photograph that the old side channel ran through this slot over here. Pilings were placed across it. Rock was placed across that and closed it off to route the river this way and keep it in a single, simple channel. This type of modification of the river has decreased its dynamics and complexity. So how much of a river like the Willamette River has been modified or hardened by these bank control structures? Here's a map of the Willamette River system from Eugene on the far right to Portland on the far left, and we see that we generally have somewhere between 20 to 40% of the length of the banks of these hardened structures or revetments as we go through the agricultural lands that dropping down toward 10% to 20% in the middle section of the river, and then as we get into the urban reach or the metropolitan reach near Portland, that increasing to between 60% to 80% of the length being in these bank hardened structures. So how much of the Willamette River has been revetted and rip-wrapped? Well, we did the analysis off of the maps and the field surveys. We surveyed all of the river by boat and identified every major rip-wrap project along it, and we found that on 12% of the length of the Willamette River had both banks rip-wrapped or revetted. An additional 14% had rip-wrap or revetment on one bank. And so that means that 74% of the length of the river didn't have any revetments on either bank. So that surprised us. To be honest, I didn't believe it at first. Having gone down the Willamette River and having watched the river, I thought much more of it was revetted. So we redid the analysis and re-analyzed three times and kept coming up with the same number. And then we started to think about, wait a minute, it's not just how much is revetted, it's where it's revetted. So we started to ask the question of how many of the bends where the river is dynamic and migrating, how many of those bends are revetted or rip-wrapped or controlled. And we found that approximately two-thirds of the meanders or the bends in the Willamette River have been hardened and anchored or revetted by rip-wrapping. And so that means that most of the places that are dynamic and changing are suddenly frozen and controlled by these bank-hardening processes. So we've taken a river that was dynamic and complex and changing and creating new habitats and complex habitats, and we've turned it much more into a simple pipe-like river system. Now we've done many other things in riparian areas, such as this shot from the lowland systems that have modified the quality of the habitat. This is a stream that flows through Oregon State University's property named Oak Creek, and we see that there are some remnants of the historical riparian forests that are present, but they tend to be younger and smaller now, and much of the wood, like we see in the channel, is this 2x4 that's in the lower portion of it, not really the natural large wood that we would see in this channel. There are also problems with water quality in these streams, with runoff from dairies and irrigated crop lands, runoff from residential areas that had diminished the quality of the water. So we took a look at the quality of the habitat in the streams around the Willamette Basin from the headwater forest down to the valley streams to see how they've changed from historical conditions to the present and how they're changing forward into the future. Some of these changes haven't been good, such as this dead cutthroat trout that we found in one of the reaches on OSU's property after some release of animal waste. And so many of these changes can be detrimental. This is a graph that shows you the responses of both fish and aquatic invertebrates under the different scenarios that we were talking about a little bit earlier. These are comparing these different conditions to the present. So current conditions are represented by no change or the zero on this line or the axis. If you go up, that means things are getting better. If you go down, that means things are getting worse. So under the conditions in 1850, generally the aquatic components were anywhere from 20 to as much as 90% higher in quality under the conditions of 1850. So this represents the decrease in habitat quality that we've seen from 1850 to the present. This is the habitat quality that has been lost. This is the planned trend. This is what happens if we go into the future under our current policies. We see a continued decrease, but it's not nearly as great as what we've seen since 1850. And under the high development, we also see continued decrease, but again, not as much as we've seen since 1850. But the conservation scenario showed improvement, basically a restoration of some of the habitat quality and community characteristics that we saw in 1850. It showed that generally about a third of the habitat quality that had been lost in the last 150 years could be regained in the next 50 years through plausible conservation measures defined by stakeholders themselves, farmers, foresters, community planners, environmental groups. These people came together and defined these scenarios, and this modeling exercise would show that our investment in conservation can make a difference. In addition, we looked at the upland forests and we looked at the public lands where we have the Northwest Forest Plan with extensive riparian reserves and floodplain forests versus private forest lands or private lands that include private forests that have smaller buffers and partial harvest or agricultural lands that have no required buffers. And we see that on the public lands, under all of the scenarios, historical, the conservation, the planned trend or our current policies or development into the future to 2050, there's an increase in the quality of the trout habitat. But on the private lands, there's been tremendous decreases and we continue to lose the quality of that habitat under either our current policies or high development. But only under conservation measures do we see an increase in the trout habitat quality for our private lands in the mountains around the Willamette Basin, indicating again that these measures for conservation strategies do make a difference. Now as we take a look at our river systems, we see these remnants of our previous ecosystems. We can use these kinds of visions of intact habitats present within our current landscapes, such as the Willamette Basin here, to project what might be possible in the future. So landscape analysis includes not only historical reconstruction, but future projections. And what you're going to see here is an example of some of the future scenarios that have been developed with our research team and stakeholders to show how this basin would change from present conditions that you see here to the future policies in the next 50 years to what happens if we go into high development or if we go into conservation measures. So I flip back through those from the present and our current conditions to what will happen 50 years in the future if we implement our current land use policies to what will happen if we relax those policies in favor development to what would happen if we implement plausible conservation measures on forests, agricultural, and urban lands. And all of these provide some measure of where we might go into the future, showing the value of landscape analysis and starting to create a spatial context and a temporal context to let landowners and citizens think about where they want their landscape to go in the future. We can see that we can use these same kinds of measures such as this one about the percent of floodplain area with woody vegetation as some indicator of the potential. This is what it was like in 1850. This is what it's like today. And so we see that we've lost a huge amount of the wooded forest. Now it's not to say that this is just to say that this is bad. This can be used as a context to start to plan where we could restore ecological function in this landscape. But it's not just the biological or the physical properties of our landscape that we want to look at. We also might want to look at the human characteristics such as this map of the density of people from Eugene to Portland. And so we can see that we have high densities of people within the floodplain in Eugene, in Corvallis, in Salem, in Portland. And we can start to map spatially where we start to see large numbers of people living within the floodplain. Or alternatively, we could look at density of structures. Again, we see high densities of structures in these urbanized areas and less in the agricultural areas. But they're not exactly the same. I'll toggle between them for a second and you'll notice the pattern shifts slightly based on the distribution of where people live and where they put their industrial areas. So we can use these maps of the biological properties of a river system in its floodplain, the physical properties of that floodplain, and also the social and land use properties of that floodplain to start to decide where we might want to put restoration. As an example, we could plot out these lands from this information and map out the ecological benefits. And we could also map out the demographic or economic constraints. And so a characteristic that might be very suitable for restoration would be these areas that show high potential for ecological recovery, but they have very low social obstacles in terms of development value or density of people within the floodplain. In contrast, these places where you don't get much ecological benefit, but you have high obstacles due to social factors are kind of the no-brainer. It's like, duh, you don't want to be here because anything that you put in place probably isn't going to last, and you aren't going to get much payoff from it if you do it. These areas that have high ecological value but also high obstacles or constraints are areas that you might want to consider as representative programs or some kinds of assistance programs to shift them toward lower social constraints but maintain those high ecological potentials. Now, having said that, I want to stress that you don't take this simple analysis the way I've just described and use that as your only measure. You might intentionally want to go into this section of the graph in which you have relatively low ecological potential but high social constraints. Now, why would you want to go here? I said this was a no-brainer. This is where people live. This is where people work. If you want to educate people, we do things with our landscapes where they live. They will learn more every day from going by a restoration project or a conservation project right in the places where they live and getting off into some remote location. They will see it every day and they will learn more lessons from that experience every day of their lives than they will from all the workshops that we set up for them. So we need to design some of our projects to go into the heart of where people live, whether it's in the urban areas or the residential areas or in the agricultural areas. We put those kinds of conservation projects or restoration projects right in the heart of where they exist from it and can think about what they can do with their own lands or their own opportunities. So as we go into our river systems, as I said, we have these intact pieces of our landscape, but we're not trying to take our landscapes and take them back to 1850. We're trying to take them forward to the future. We're not trying to rebuild a past landscape. We're trying to design the landscape that we want for our children and our future generations. Much of this, like the redevelopment of floodplain forests or channel complexity or hydrology, will take long periods of time. It will take many decades of river processes or ecological processes. The same thing is true of uplands like the restoration of grasslands or forests or oak woodlands. And so we will never get to that point in the future unless we start today. And we start doing that with our communities by planning how this landscape can be restored. I want to lay out just a couple of principles of ecological restoration that may help guide you in thinking about how we restore ecosystems. First of all, one of the first principles is that ecosystem restoration is based on restoring the ability of systems to maintain natural trajectories of physical and ecological functions. Okay. We're trying to restore the ability of the system to maintain these trajectories. That implies two major factors. Number one, it's a trajectory. It is not a fixed state. These systems are always changing. They always will be changing in the future. The question is, can we move them toward a trajectory of change that is desirable? Secondly, we're not trying to build that restoration-addicted systems. We're trying to build systems that can maintain their own processes. And so we give them the components, whether it's biological or physical, that allow the system to shape itself. And so we're not building systems that require us to keep going in and establishing these monuments to our good intentions. Now, a couple of other principles that are very useful guiding principles but they're very difficult to deal with under these next two. Number one, practices that caused resource degradation must be changed to prevent continued loss of habitat, function, or species. Quite often we implement our restoration projects and yet we continue the very processes that cause the degradation in the first place. It's unlikely that we will be as successful if we do that because we're basically against the current of ecosystem degradation that's already occurring. Secondly, that these changes in management practices should precede the restoration efforts to the degree possible. So we need to implement changes in the practices that cause the degradation as early as we possibly can and hold off on our restoration efforts until the system has adjusted somewhat to that change in management. And that's the hard part. People always want to do something and so it's hard to say we're going to wait on restoration until the system has adjusted to this change in management but to the degree that we can we will have more effective restoration. Another very fundamental principle is that restoration that uses natural materials or native organisms within their natural ranges of abundance and distribution are likely to be effective over the long term. If we bring in non-native materials or materials in greater abundance than they naturally occur it is unlikely that we're going to get natural function or anything like it. So we can use this as a design criterion to help us get projects that are more likely to achieve the ecological results that we're looking for. So some of the risk to effective restoration are creating unnatural patterns using materials where they don't naturally occur in nature using materials or species at levels that are not observed in nature and using non-native species. These will tend to diminish the success of restoration. So we have to remember that ecosystems are dynamic and constantly changing and if we take a dynamic approach to conservation and restoration it is more likely that we will be successful. Restoration simply to a previous state often is impossible and quite often is ecologically undesirable. These systems are changing. The question is how do we move them forward to the conditions that we think are ecologically appropriate in the future. And along with this comes the idea of restoration. If any of us are doing a restoration or a conservation project or doing a good job of it we're trying to evaluate it. And that's a really good thing. But there is a tendency in our projects to evaluate our projects simply based on persistence. Evaluation of restoration efforts simply based on the persistence is just as static and ecologically inconsistent as attempts to erect monuments structures or maintain fixed conditions. So I've often seen both the proponents and the opponents of restoration out there with the clipboard saying is it still where we put it and if it is it's a success and if it's somewhere else it's a failure. Well that doesn't make any sense. If the stream or river has moved the material to a place that is ecologically appropriate and is using it to build functional habitats under natural processes that's a success whether it moved it from where we put it or not. And if it left it in place even though it might have been able to use it more effectively at another place but we anchored it such that it couldn't move that's a failure. And so we need to have a dynamic perspective for evaluating the success of our projects. Now we need to keep a good sense of humor about what we're trying to do. Most people that are out trying to use the landscape are trying to do a good job with the natural resources. By understanding landscape dynamics and the spatial and temporal characteristics hopefully we can design more effective conservation of existing habitats and restoration of conditions that are not really what we want them to be. So how do we keep from getting mired down or buried in our work? Well here are some signs that we may not be doing ecologically what we'd intended. Any time that we start to use cabling and hard structure to wire habitat into place we're probably not designing an ecologically dynamic and functional landscape. So we've gone toward erecting this monument that's going to stay right where we put it instead of building dynamic and healthy habitats for the future. Regularity. Now as an older person I'm starting to appreciate regularity but regularity in nature is not all that healthy. Streams are not these nice little regular stair steps of little cascades or steps in pools. And any time you start to see an even spacing of pools and riffles, pools and riffles under these little fixed structures we've implemented a geomorphic structure that probably is not natural to that stream and will actually be a degradation instead of a restoration. When we start using artificial materials such as car bodies, I mean I don't care how many fish are in the backseat of this car, that's not good habitat. And so dumping car bodies, dumping other artificial materials is a very poor substitute for ecological function and restoring ecological function. The only advantage is it's very easy to know the date of your project based on the year of the car but that's about the only good that we're going to get out of this type of restoration. Now many people criticize restoration but there's a question of does it work? Well typically we get out during benign conditions in the middle of summer or nice calm conditions and evaluate our restoration. But we have to stop and think about disturbance. Now I want to ask you a question. Let's imagine you're all in the room here and how many of you drove to this session? And then the second question is how many of you wore seatbelt? And then the third question is of those of you who drove to this session and wore seatbelt, how many of you were killed? Well, you'd probably sit there and say well none of us were because none of us were in a wreck. And that's just the point. Part of our evaluation of restoration is how do these systems respond in the face of a disturbance? And so just evaluating them under benign conditions does not necessarily indicate their full ecological value. And here's an example of cutthroat trout abundance that we saw in a project in the Cascades and the restoration project is this histogram or this bar that's in orange. Now we went in in the late 1980s and put in large pieces of wood to restore the function and over five years it reformed its structure of large wood and the full composition of large wood that we would expect to see in an old growth forest. We had a flood that came through in 1996. Upstream of that restoration reach where there was no wood we saw a decrease in the trout numbers by almost 40%. In that same flood in the restoration reach we saw an increase by more than 40%. And so the trout not only survived but they increased their density within the restored reach where we had abundant large wood habitat. And then outside that restoration reach just downstream of it that was buffered by the large wood we saw no change in the trout populations. So this project showed that it did provide a very positive function in that disturbance event in providing critical habitat that allowed the fish to survive the disturbance. So we need to evaluate our restoration efforts not just during benign conditions but also in the face of disturbances. The other thing is that sometimes the restoration efforts do work. Here's an example from one of our experiments that shows that cutthroat trout abundances do increase as we increase the complexity of large wood. And so we can find evidence that our efforts do work but at the same time experiments have also shown no change in trout abundance with increase in large wood in areas that have large amounts of boulders. And so we need to be careful to be as clear and accurate and honest with our results of restoration and conservation as we can. Sometimes it works, sometimes it doesn't and we need to keep learning. So one of the first things we need to dedicate ourselves to in any conservation project or any restoration project is learning. It is shocking how many projects we put in on the ground in which we do not do it in an organized fashion so that we can learn from it. And so if we dedicated even a small fraction of our collective resources, money, time, people and natural resources to rigorous experimentation at appropriate scales, the right size stream, the right length of stream, the right amount of time, the right size of forest, the right size of grassland, dedicating our resources to experimentation that these appropriate scales would advance our knowledge and perhaps increase the effectiveness of our future efforts to restore aquatic and terrestrial ecosystems. And so I beg you, if you have the opportunity, plan learning as part of your conservation and restoration project so we learn more about how the landscape works. Now, not all restoration, not all conservation works. Some of it does work very effectively, some of it doesn't work. We're not trying to sell snake oil here. We're not trying to mislead anyone and we're not trying to confuse the issue. We need to be as straight as we can with the public about what we're trying to do in conservation and restoration of the landscape and how long it's going to take. This isn't a quick fix. Some of my friends have laughed and said that quite often even immediate gratification is not fast enough for most of us and that tends to be the way it is with conservation. We want our results fast. Well, some of the ecological restoration that we're trying to do is simply going to take decades, even centuries, but it does start today. And so some of the restoration that we're doing will work and we need to learn from it. We need to learn what works and what doesn't. And we need to understand the spatial and temporal characteristics of our landscape so that we can make better decisions about tomorrow. This isn't about going back in time. This is about going forward. So how do we learn about those key relationships between ecological processes and structure and the kinds of things we try to do in the name of conservation and restoration? We don't know all the answers, but we can create a good, sound, rigorous context for planning conservation and restoration by using these principles of landscapes, the spatial characteristics, the temporal characteristics of the landscapes. And so we can develop good concepts and try to implement them as honestly and straightforwardly as we can and make good observations and then, based on those good observations, make good choices about our future actions to conserve or restore natural resources. We need to establish relationships between the spatial and the temporal characteristics of our landscapes and the desired management outcomes and the ecological functions of our landscape. But in this, we need to acknowledge uncertainty. We do not have all the answers. We will never have all the answers. And so how do we make good decisions and good choices when we don't have all the answers? One of the ways that we do that is by building the solid frameworks of concepts and making good observations and then learning from what we do. And so by developing a dynamic view of landscapes that recognizes past conditions and functional landscapes, current conditions and how they've been modified and where we want to head these ecosystems in the future, we can start to create spatial and temporal context for understanding the landscape.