Anning, A. K., Dyer, J. M., & McCarthy, B. C. (2014). Tree growth response to fuel reduction treatments along a topographic moisture gradient in mixed-oak forests of Ohio, U.S.A. Canadian Journal of Forest Research, 44(5), 413–421.
Anning et al. (2014) focused on evaluating tree growth response to fuel reduction treatments along topographic moisture gradients. In order to accomplish these 648 increment bore samples were collected from 348 trees of five different species. These samples were collected from eighty .1 ha plots that varied with 4 different treatments. These treatments included a control, a thinning treatment, a prescribed burn treatment, and a thinning treatment accompanied by a prescribed burn. A GIS model estimated potential evapotranspiration, actual evapotranspiration, and moisture deficit using the water balance approach. The results indicated that the moisture gradient was most impactful on the non-managed stands. This is likely due to decreased competition, increased nutrient availability, and increased mineralization rates on managed stands. Oak tree growth in the control area were less controlled by moisture gradients than the other species studies, this is most likely due to the species ability to grow deep roots that can tap into deeper water supplies during times of limited water. This indicates that these species may do better with predicted changes in climate. This study demonstrates how GIS can be used to study fuel reduction treatments effect on forest characteristics such as tree growth.
Arima, E., Walker, R., Perz, S., & Caldas, M. (2005). Loggers and Forest Fragmentation: Behavioral Models of Road Building in the Amazon Basin. Annals of the Association of American Geographers, 95(3), 525–541.
Tropical deforestation is a common area of study, but little has been researched regarding spatial manifestations especially regarding land coverage change and fragmentation. A known cause of this fragmentation is road building by loggers. Arima et al. (2005) looks at explaining these forest road expansions using a theoretical model of economic behavior with a GIS. The model that was built was aimed at modeling two different types of roads: destination indeterminate and destination determinate roads. Destination determinate modeling used an algorithm to determine paths that would yield the most hardwood. The study did not have spatial data on tree distribution, so it was assumed that hardwood was spread evenly over the landscape. The destination determinate roads had an end point designated so it only needed to use the software to identify least expensive path to the destination, the destination indeterminate also used this software when determining the path of the road. Roads created in the software were compared to existing roads to determine accuracy of the model. Although the study had only limited success at modeling the fishbone pattern of fragmentation, they did learn from interviews that many of the roads were built by colonists or from pressure from colonists on the government. Loggers typically only extended these existing roads, which means that colonists most likely have a greater influence on road construction meriting a different approach to the model. The study was more successful at modeling destination determinate roads over destination indeterminate roads which may have been remedied if the model was changed to meet colonist criteria and/or tree distribution data was included.
Blackford, C., Heung, B., Baldwin, K., Fleming, R. L., Hazlett, P. W., Morris, D. M., Uhlig, P. W. C., & Webster, K. L. (2021). Digital soil mapping workflow for forest resource applications: a case study in the Hearst Forest, Ontario. Canadian Journal of Forest Research, 51(1), 59–77.
Soil is critically important to forests and thus critically important to forest managers. So, the ability to map soil properties effectively is very beneficial to forest managers. Conventional soil mapping combines soil pedon observations with aerial orthophotogrammetry to create soil polygons. This method can be beneficial but has some serious limitations including the fact that soil composition varies over the landscape and does not fit into polygons and the accuracy of these maps are not always reported or known. For this reason, digital soil mapping is replacing conventional soil mapping for many applications. Digital soil mapping uses statistical modeling to discover relationships between observed georeferenced soil data and environment data. The method has had limitations and limited success in large scale forestry applications because of large amounts of variation over landscapes and limited soil pedon data over these large areas. This paper aimed at testing a standardized semiautomated digital soil mapping workflow for use in forestry. The results of the case study showed promise with their workflow method and they found that misclassifications were typically with similar soil types, which they took as a good sign. Digital soil maps have their limitations as demonstrated by the study, but as a field their use is growing and with more research and technological advancements their accuracy will improve.
Çakir, G., Sıvrıkaya, F., Terzıoğlu, S., Başkent, E. Z., Sönmez, T., & Yolasiğmaz, H. A. (2007). Mapping Secondary Forest Succession with Geographic Information Systems: A Case Study from Bulanıkdere, Kırklareli, Turkey. Turkish Journal of Agriculture & Forestry, 31(1), 71–81.
Almost all forests on the planet have been altered by humans and the forests of Turkey are no different. Secondary succession is the process in which species composition or structure changes after a disturbance. Understanding this secondary succession can be very beneficial to forest mangers which is what Çakir et al. (2007) focuses on doing. The study uses GIS, remote sensing, and aerial photos to document secondary succession in the Bulan›kdere Forest planning unit. Two maps were created for this study, a forest stand type map and a secondary succession map, as well as attribute data for these maps. The forest stand type map was created using remote sensing and a field survey, and this map was used to create the secondary succession map. The determination of secondary succession stands was based on the field survey samples and a value of 1-6 based on the Clementsian theory of forest succession was used. These field sample values were then used to calculate succession values for the forest stand types which were also compared to remote sensing data. The authors noted that because few experts in the field of secondary succession are adept at GIS, remote sensing, and GPS in Turkey mapping secondary succession has had challenges, but this study is the first to extensively map secondary succession in Turkey. This information can help forest managers understand what secondary succession will look like after harvesting and what forest management prescriptions are appropriate.
Coops, N. C., & Waring, R. H. (2001). Estimating forest productivity in the eastern Siskiyou Mountains of southwestern Oregon using a satellite driven process model, 3-PGS. Canadian Journal of Forest Research, 31(1), 143.
3-PGS is a process-based forest growth model that predicts forest growth using biophysical relationships and constants. The model required data from multiple sources including climate data that was acquired using PRISM, estimations of short-wave radiation which were acquired using modeling, soil fertility using the State Soil Geographic (STATSGO) database, soil water holding capacity was calculated using the information from STATSGO, and satellite imagery was used to measure vegetation attributes. The study area was located in the Siskiyou mountains where 19 plots were located for comparison to model data. Data was collected at these 19 plots between 1965 and 1975 including air temperature, soil temperature, radiation, humidity, vapor pressure deficit, soil moisture, and soil fertility. Physiological responses were measured on two species on these plots: Douglas fir and Shasta red fir. Physiological responses measured included age, height, phenology, plant water potential, stomatal resistance, and foliar nutrition. All plots were revisited except for one because its original trees were not intact bringing the total plots to 18. 3-PGS is much less comprehensive than other ecosystem simulation models but has the advantage of using constants instead of requiring difficult calculations. This comparison demonstrates that models such as 3-PGS can provide reasonable predictions for the Pacific Northwest. Also, because of it has the potential to estimate forest characteristics under climate change scenarios.
Ducheyne, E. I., De Wulf, R. R., & De Baets, B. (2006). A spatial approach to forest‐management optimization: linking GIS and multiple objective genetic algorithms. International Journal of Geographical Information Science, 20(8), 917–928.
Spatial problems of forest management are often left out of the planning phase of projects and expected to be worked out during implementation. This can lead to misleading results and plans that are impossible to implement at all. For example, when forests are small or fragmented, the spatial component cannot be ignored or expected outcomes may be impossible. A Spatial Decision Support System (SDSS) is a tool that links spatial and non-spatial data to help decision makers in the decision-making process. The non-spatial data that this study linked with spatial data was genetic algorithms. “Genetic algorithms are ‘search and optimization’ algorithms based on natural selection and natural genetics” (Ducheyne et al., 2006). The study area was Kirkhill forest in Aberdeen, Scotland. The study aimed to develop harvesting plan using GIS and genetic algorithm with multiple objectives in mind including timber volume, abundance of edge dependent animals, and abundance of animals dependent on mature forests. The study found that using genetic algorithms was much faster than linear programming and that using a multiple objective genetic algorithm was more efficient than using a single-objective genetic algorithm for development of a harvesting plan. Overall, the study found that using genetic algorithms and GIS was useful for forest management.
Georgios Liampas, S.-A., Stamatiou, C. C., & Drosos, V. C. (2019). Forest road network planning for biomass exploitation and fire preventions: a least cost path analysis. Agricultural Engineering International: CIGR Journal, 21(4), 33–42.
The Georgios Liampas et al. (2019) study took place in the forests of Greece and involved opening up the forest to logging, research, and environmental protection by construction and maintenance of forest roads. The study uses a least cost path analysis and multi-criteria analysis using a GIS. A raster map is used where each cell is given a cost if the path crosses over it. An algorithm is then used to find the least expensive path to construct a road from forest to necessary end point. Using this a map of five proposed roads was created to open up the forest to necessary activities. The study found that GIS aided road planning was more efficient and affordable for opening up the forest. Using multiple criteria analysis allows for the decision to be made using more criteria than just the most common of slant and exploitation of land, but also include timber reserves and waste. This method can be the most affordable by reducing wasted time and allowing for analysis of many criteria in a short amount of time.
Hytönen, L. A., Leskinen, P., & Store, R. (2002). A Spatial Approach to Participatory Planning in Forestry Decision Making. Scandinavian Journal of Forest Research, 17(1), 62–71.
Public participation and sustainable development are recognized to have an important relationship in current sustainable development theory. Public participation can include surveys, decision analysis, and unstructured feedback. In Finland the use of unstructured feedback through public meetings, forest telephones, and feedback forms allows managers to collect important information on local knowledge. This local knowledge includes information about “the landscape, its social history, scenic beauty, community identity, family heritage, and spiritual values” (Hytönen et al., 2002). Hytönen et al. (2002) aimed at collecting this unstructured qualitative feedback and then turning it into quantitative data with a spatial component. This was accomplished by developing typologies for instance hunting and then grouping the feedback as arguments regarding the typology. The typologies were then linked to spatial areas either through the arguments or through proximity, for instance if the argument was old growth forest should be set aside then it was linked to old growth forests in the area. Yet, some arguments could not be included in spatial analysis as they were too abstract. This qualitative feedback that has weighted scores linked to locations and can be spatially analyzed is presented as a possible tool for use in decision making process for managers. It does have its challenges as the authors felt that feedback forms are not the best way to collect citizen information as the topics mentioned in the form guide issues and some issues are too abstract to include a spatial component.
Karlsson, J., Rönnqvist, M., & Frisk, M. (2006). RoadOpt: A decision support system for road upgrading in forestry. Scandinavian Journal of Forest Research, 21, 5–15.
Karlsson et al. (2006) covers a GIS based program called RoadOpt designed to help make decisions on forestry road upgrading. This program is designed to be used in Swedish forestry where timber companies lose money because they have to store large amounts of sawlogs when the roads are impassable causing deterioration of timber. The program has two primary parts, a Swedish road database and a mixed integer linear programming model. The program uses costs and benefits of upgrading roads based on potential harvest areas accessible using the upgraded roads. This program allows the company using it to determine if accessible roads to harvest can keep up with demand and if not, what roads can be upgraded to meet demands in variable weather periods. Forestry roads are very important for timber harvesting as well as access for firefighting efforts, maintenance and upgrade of roads can be very costly. This type of program can be very useful in forestry planning allowing managers to analyze differing forest road scenarios and assess the cost and benefit of upgrades.
Knapp, P. A., & Bishop, K. A. (1993). Use of GIS in optimizing timber thinning strategies in the eastern Sierra Nevada. Professional Geographer, 45(3), 323.
Knapp & Bishop (1993) wanted to evaluate the use of GIS in timber thinning projects in the Kyburz Planning Area in the Tahoe National Forest. A database was developed using fifteen 1:24,000 scale mylar maps which included the following characteristics: percentage slope, management area sensitivity, soil erosion rating, soil compaction limitation, percentage soil cover, deer habitat zones, sensitive plant species, survey sections, streams, vegetation habitat types, roads, timber stands, watershed boundaries, goshawk habitat, and fire hazard rating. The maps were digitized in vector format. Using this data, the researchers created two maps: one titled potential harvest area which included percentage slope, streams, vegetation habitat types, roads, timber stands, and fire hazard ratings, and the other titled vegetation types consisting of vegetation habitat types which was used as a baseline prior to thinning. In the Kyburz Planning Area the two biggest factors for determining if a thinning project is feasible is if it will improve forest health and if it will reduce fire hazards. Whether it is economically feasible plays an important part of whether the project is actually implemented. Prior to the use of GIS as was demonstrated in this study answering these two important criteria was challenging, but now with the use of GIS it is possible to spatially analyze this information. GIS has become an important tool in fuel reduction planning as well as in firefighting planning.
Lavy, B. L., & Hagelman III, R. R. (2017). Spatial and Temporal Patterns Associated with Permitted Tree Removal in Austin, Texas, 2002-2011. Professional Geographer, 69(4), 539–552.
Urban forests are an important part of the urban landscape, but large numbers of urban trees die prematurely because of damage, disease, climate change impacts, and invasive species to name a few. Lavy & Hagelman (2017) aim to spatially and temporally analyze permitted removal of trees by residents, business owners, and public entities in Austin, Texas between 2002-2011. In Austin a protected tree is a tree 19 inches DBH or larger and a heritage tree is a tree 24 inches DBH or larger. In order for anyone to modify a protected or heritage tree on public or private land a permit is required. Depending on the permit request a permit if approved can allow removal, encroachment into critical root zone, or crown thinning of thirty percent or less. Permits are granted after a site inspection based on several criteria. The GIS that was created included hydrological maps, a DEM of the study area, urban landscape characteristics (age of structure, distance to major roads, land use, and population density), socioeconomic characteristics (percentage white, percentage owner occupancy, percentage college graduates, median income, and market value) and of course tree removals (Lavy & Hagelman, 2017). The study found that development related removal peaked between 2007 and 2010 which coincides with the sub-prime mortgage crisis and non-development removals peaked 2011 which is hypothesized to be due to drought killing many trees in the area and the tree removal ordinance being in the news.
Lindemann, J. D., & Baker, W. L. (2002). Using GIS to analyse a severe forest blowdown in the Southern Rocky Mountains. International Journal of Geographical Information Science, 16(4), 377–399.
Forest disturbances have been studied extensively because of their impacts on forest structure and function, but little research has been done on why they occur where they do. Understanding spatial factors which influence the location of forest blowdown can help us to understand disturbances and help guide management decisions. In 1997 the largest blowdown known to occur in the Rockies happened impacting 10,000 hectares of forest. Aerial photos of the blowdown taken a few days after the blowdown were used for this study along with pre-blowdown images to create polygons with blowdown percentages. A DEM of the area provided by the Forest Service was used to determine elevation, slope angle, and aspect. The EXPOS program was used to determine wind exposure as well as vegetation and habitat data and soil data were collected. The two model that were created in this study found that relatively few variables contribute to blowdown. These few variables were topographic features not vegetation or soil features as might have been expected.
Platt, R., Veblen, T., & Sherriff, R. L. (2006). Are Wildfire Mitigation and Restoration of Historic Forest Structure Compatible? A Spatial Modeling Assessment. Annals of the Association of American Geographers, 96(3), 455–470.
Ponderosa Pine dominated forests historically experience frequent low intensity ground fires every 2-20 years. Fire suppression in the recent century has changed the complexity of stands allowing for higher intensity crown fires. It is logical to think that forest thinning projects could reduce fuel loads while returning forest structure to historic conditions. Platt et al. (2006) focuses on the use of GIS to evaluate the restoration of historic forest structure and fire mitigation through thinning projects. The study area is a ponderosa pine dominated forest in Boulder County, Colorado. Two spatial models were overlayed for this analysis, one that models potential fire line intensity and one that models historic fire frequency. This overlay allows for the assessment of whether an area needs to be thinned for fire intensity or both thinned for fire intensity and thinned to return to a historical structure. If the fire intensity is low it is assumed that the fire can be suppressed, but if it is high then it is assumed that fire mitigation thinning is necessary. Areas that are not characterized by low fire line intensity and high fire frequency may not be characterized as historical stand structure. The study found that only 27 percent of the study area land needed both outcomes, 27 percent needed thinning for fire intensity and the remainder needed neither outcome. It also found that most the area that needed treatment was non-Forest Service land, but the Forest Service is the agency that receives the most funding for forest thinning projects. A study such as this can teach forest managers and policymakers a lot, for example the need for funding for thinning projects on non-Forest Service land.
Robinson, D. T., & Brown, D. G. (2009). Evaluating the effects of land-use development policies on ex-urban forest cover: An integrated agent-based GIS approach. International Journal of Geographical Information Science, 23(9), 1211–1232.
Robinson & Brown (2009) used a GIS agency-based model (ABM) called dynamic ecological exurban development (DEED) to evaluate hypothetical scenarios of lot-size zoning and municipal land acquisition strategies on forest coverage. The study area was the Scio Township in Michigan, USA. The first baseline scenario had unrestricted development and lacked policies on minimum lot size zoning or land acquisition strategies. This was compared to 20 different land use policies that all resulted in more forest cover levels than the unrestricted scenario. Only one policy that acquired land for conservation on the most forested lands led to increases in aggregate forest cover and reduced area in residential developments. All the policies which acquired land for forest conservation increased the amount of forest cover in the township as expected. The results of this study verified other literature that large lot sizing zoning policies lead to greater sprawl and that large lot zoning strategies can influence forest cover, but the effect was small relative to the effect of municipal land acquisition (Robinson & Brown, 2009). The study overall demonstrated that a GIS-based ABM using hypothetical scenarios and real-world data can produce results that describe the effects of lot zone sizing and land acquisition strategies on forest cover (Robinson & Brown, 2009).
Rongxia Li, Bettinger, P., Danskin, S., & Hayashi, R. (2007). A Historical Perspective on the Use of GIS and Remote Sensing in Natural Resource Management, as Viewed through Papers Published in North American Forestry Journals from 1976 to 2005. Cartographica, 42(2), 165–178.
This article assessed the evolution of the use of GIS in natural resource management by reviewing North American forestry journals from 1976 to 2005. The article reviewed articles from seven major forestry journals that are intended to be read by field practitioners. The seven journals reviewed were “Forestry Chronicle, Canadian Journal of Forest Research, Northern Journal of Applied Forestry, Southern Journal of Applied Forestry, Western Journal of Applied Forestry, Forest Science, and Journal of Forestry” (Rongxia et al., 2007). The article found that the use of GIS has increased since the 1970’s and is predominantly used for local site-specific applications, but the use of GIS in landscape scale applications is gaining popularity. In the early years that were reviewed in this study only a couple of articles each year included GIS, but by 2005, 38 papers published in these journals included GIS. The paper predicts that as technology and education evolve spatial analysis will transition from being done only by GIS analysts to being done by entry level managers especially as college level courses on the topic become more prevalent.
Szabó, P., Suchánková, S., Křížová, L., Kotačka, M., Kvardová, M., Macek, M., Müllerová, J., & Brázdil, R. (2018). More than trees: The challenges of creating a geodatabase to capture the complexity of forest history. Historical Methods, 51(3), 175–189.
The Szabó et al. (2018) paper discussed the challenges of creating a geodatabase called the LONGWOOD database, the database is made up of historical forest data of forests in the Czech Republic. The database included information from the 11th century up to the 20th century from 3,567 townships. Information included in this data was information on forest history area, tree species composition, and forest management. The database was created in Microsoft Access 2007 and updated in Microsoft Access 2010. The data stored in Access can be spatially analyzed using GIS. One aspect the project felt was lacking was spatial forest boundaries and they felt this would be a good focus for further research. Finding common resolution from all sources was also a challenge as the depth of some of the archival data was almost too deep. A spatial database that collects historical information such as this one can give a lot of insight into the history of the forest. It can also help managers extensively if the goal of management is to return historical forest structure. There are challenges with this as was stated with the depth of information and possibly with the accuracy of information collected so long ago.
Yool, S. R., Eckhardt, D. W., Estes, J. E., & Cosentino, M. J. (1985). Describing the Brushfire Hazard in Southern California. Annals of the Association of American Geographers, 75(3), 417–430.
Brushfires are very common in the forests and foothills of southern California and have a history of causing a great deal of damage. Developing models and a collection of data that can predict and potentially help mitigate fire severity is very beneficial in these high brushfire danger areas. Yool et al. (1985) discusses the use of remote sensing data and landscape data to create a GIS that can help predict brushfire hazards. The brushfire GIS application developed for Yool et al. (1985) study contains fire history, rainfall, topographic, vegetative data, LANDSAT, and digital topographic data. The two test sites were located in the Santa Ynez mountains near Santa Barbara and the Condor Peak test site in the Angeles National Forest. The study concluded that the use of remote sensing data and landscape data in a GIS could be useful for modeling rate and direction of fire spread.