Traditionally a great deal of natural resources management has involved field-based surveys and plans, explains Hong S. He, Associate Professor of Forestry in the School of Natural Resources at MU. But recently these scientists and managers have come to realize that they also need to pay attention to the larger spatial configuration of natural resources. This realization has a lot of implications for wildlife conservation and biodiversity: “You can’t really consider one spot without considering the things around it,” he explains. Wildlife species require, for instance, multiple habitats, and watershed problems have shown that “if we pollute one area, it can spread over the landscape.” As an area of research, landscape ecology refers to the study of response to various natural and social factors over large spatial and temporal domains.
Consider the case of forests, which cover roughly 25% of the earth’s ground area (not counting Greenland and Antarctica). Forests are of utmost importance to the planet’s survival, functioning in so many capacities: they filter water, store carbon to fight against climate warming, serve as homes for at least 60% of plant and animal species, not to mention providing timber, fuel, food, and fodder for use by human beings. Yet humans have so far destroyed as much as 50% of forest land, and they continue to do so at alarming rates. Given how critical forests are for the present and future of all of earth’s inhabitants, and the sobering statistics about their rapid destruction, an effective ecological analysis of forests must consider the “bigger picture.” And this is exactly what He sets out to do by focusing on huge spatial areas (as much as several million hectares) and time ranges (50-100 years or longer). In order to study such large geographical areas, one needs to capture a significant amount of data (using remote sensing technology such as satellite images and air photography), to process and analyze that data (using GIS, a computer system that deals with extremely large spatial and temporal dimensions), and then to plug in known principles of landscape ecology.
Elements of ecology change across time as well as space. Typical field experiments are not reliable enough in this regard simply because the effects of today’s forestry management decisions won’t show up for many decades. Take the harvesting of trees, for example, which causes most of the destruction of the earth’s forest land. If just one area is harvested, the environmental impact will not be fully realized until 20-30 years after the fact. This means that the wrong management decisions now could lead to devastating environmental problems in the future. In lieu of an ecological crystal ball, a computer model called LANDIS is becoming a crucial tool to conduct simulations when typical field experiments are not feasible. LANDIS employs current science “to simulate the long-term effects of various management decisions,” whether assessing tree harvesting practices, the role of insects and disease, or even fire management strategies.
“Fire management?” I asked. “People tend not to think of fire as a manageable component,” He responds, “but it is.” In past decades the dominant paradigm in the U.S. was called fire suppressionâ€•that is, extinguishing fires as soon as they break out. While this strategy may seem logical to the average person, who assumes that putting out fires not only protects timber but saves trees, He explains that “actually, fire is the one important natural component that helps forest ecosystems to regenerate. . . . You actually need the fire for the health of the trees.” Ironically, the prior U.S. management policy of quickly extinguishing fires has allowed a lot of underbrush fuel to build up. “Eventually,” he warns, “if you suppress the small fires, in a lot of ecosystems you'll get larger, catastrophic fires that are unnatural.”
Fire is crucial to forests for another reason. Consider Missouri’s central hardwood area, where this supposedly destructive force used to play a key role. Frequent low-intensity fires burned the ground layer and young trees, helping to maintain the health of the large older trees and reduce problems stemming from insects and disease. “With fire suppression,” He continues, “the immediate problem is that fire and insects and disease come back.” The tragic oak decline in Missouri and Arkansas forests serves as a good reminder of this phenomenon; without periodic burning, the insects target and kill trees when they are about 70-80 years old, just when they could be harvested. Forestry and ecology experts now understand the value of using “prescribed fire” to reduce the underbrush fuel.
The value of this approach, of course, depends on the particular forest ecosystem. If ecologists can determine how frequently a fire needs to occur in order to maintain the forest's overall health, that information can then be actively put into practice. In response to this need, computer modeling has become an important tool employed by landscape ecologists when they need to answer large-scale questions and compare several management scenarios (including the future impact of climate warming on certain species of trees). “A tree is a tricky thing,” He reflects; “if you plant it now, you can’t anticipate harvesting it in the next few decades. Basically, you plant a lot of trees for your children or even for your grandchildren.” LANDIS looks into that future and predicts the results of various management policies 50-100 years from now. “If you don’t show them the result,” he cautions, the managers may not have the benefit of the latest, most helpful science, and future generations may suffer the consequences. Hopefully, by “using some active management alternatives,” our grandchildren “will end up with more desirable ecosystems.”
He’s work has immediate and important implications for students as well. It is because of their desire to be outdoors studying wildlife in forests that many students are drawn to MU’s School of Natural Resources. He explains that one of his goals is to bring students back into the classroom in order to train them to use the most advanced tools with GIS and spatial analysis techniques: “I tell them [that] when they graduate they are facing a different world. The computer information system is infiltrating many fields, including natural resources. If they do not pick up this skill, it will limit their marketability and usefulness.” Thanks to the efforts of He and others, MU now hosts successful GIS certification programs at both the undergraduate and graduate levels.
Yet the relevance of any one person’s skills is only part of the puzzle. “In my field, the individual is very limited,” He says. When the study size is as big as an entire mountain range, or time frames from 50-100 years, a lot of factors come into play, among them social development, land use change, and landowner change. Because of the complex nature of this research, He is involved in a number of collaborations with other peopleâ€•locally with MU faculty, nationally with the United States Forest Service and the United States Department of Agriculture, and internationally with scholars abroad. Comparing landscapes in North America with those in Eurasia, for instance, offers useful comparisons and cross-validates research involving large geo-ecological areas. Committed to this kind of research, He helped to organize the International Workshop of Forest Landscape Modeling (held in June 2006), in Beijing, China, where experts from around the world joined forces. Regarding such international collaborations, He mentions an interesting contrast: “We know fire ignores country boundaries. We know insects and disease cross country boundaries. However, as human beings we have to get a visa to go to another country.” He, who himself migrated across geopolitical boundaries, concludes on this note: “I find science really doesn’t have a boundary.”