Content uploaded by Riccardo Tortini
Author content
All content in this area was uploaded by Riccardo Tortini on Jun 22, 2018
Content may be subject to copyright.
A preview of the PDF is not available
Using annual Landsat imagery to identify harvesting over a range of intensities for non-industrial family forests
Abstract and Figures
The monitoring of forested landscapes dominated by many small private forest owners is difficult or not possible without spatially explicit and up-to-date information on land cover change. Analysis of time series multispectral data from the Landsat series of satellites have the spatial and temporal characteristics required to detect sub-hectare and non-stand replacing harvest events over large areas. We identified harvests that occurred in six western upper Michigan counties from 1985 to 2011 using Landsat best available pixel (BAP) image composites and the Composite2Change (C2C) approach. We detected a total of 7,071 harvesting events with size ranging from 0.5 to 171.36 ha and average size of 6.42 ha, and analyzed their temporal trajectory. To gain confidence in our harvest mapping, we compared our findings to the overlapping decade of Global Forest Watch (GFW) data. Agreement between the datasets was high, with 94.24% of the C2C and GFW harvest pixels identified with the same change year and improving to 98.74% within ±1 year. This automated harvest detection system, which can capture small and otherwise missed harvests, is valuable to natural resource agencies responsible for monitoring and compliance with regulations over large areas, and researchers requiring estimates of harvest levels and the nature of forest cover status and trends on family forests.
Figures - uploaded by Riccardo Tortini
Author content
All figure content in this area was uploaded by Riccardo Tortini
Content may be subject to copyright.
... When comparing different harvest practices in British Columbia, Jarron et al. (2017) found that the magnitude of change in the normalized burn ratio (NBR) consecutively to harvesting was representative of the intensity of the studied harvest practices. The NBR was also used by Tortini et al. (2019) to map partial harvests and clearcut in Michigan. Although the detection of partial harvests has received considerable attention, measuring the intensity of the cut has received considerably less attention. ...
... One of the main obstacles in studies dedicated to non-stand replacing disturbances is the availability of accurate data on the actual severity of the disturbances. Calibration and validation data are therefore frequently derived from indirect sources such as aerial surveys, high-resolution photos or other remote sensing products (e.g., Tortini et al. 2019). While such sources provide useful estimates of the extent of the damage over a large spatial coverage, they often comprise estimation errors that can limit the possibility to conduct analyses at a resolution that is useful for forest management (Johnson and Ross 2008). ...
... The partial harvest model relied on the dNBR as a single predictor. This index has been previously used to study partial harvesting in different ecosystems (Jarron et al. 2017;Tortini et al. 2019) and to monitor recovery following stand-replacing disturbances in boreal forests (Pickell et al. 2016;White et al. 2018). Our results suggest this index is appropriate to measure the intensity of partial harvests in northern hardwood forests. ...
Non-stand-replacing disturbances are major drivers of northern hardwood forest dynamics, but are more challenging to characterize using satellite imagery than stand-replacing events. This study proposes a hurdle approach in which disturbance causal agents are first attributed to permanent sample plots that were either partially harvested, had sustained damage from an ice storm or remained undisturbed during the observation period, reaching an overall accuracy of 82.9%. Ordinary least square regression was then used to develop disturbance-specific models to assess the severity of partial harvests and damage from ice storms, with r-squared values of 0.57 and 0.59, respectively. The disturbance-specific models included a different set of predictors, confirming the importance of attributing a causal agent to a disturbance before assessing its severity. The sequence of models was implemented regionally to produce severity maps for two disturbance events, revealing within-stand variability in the severity that could be useful for the planning of future silvicultural actions. Although the proposed models offer acceptable performance, more research is needed to include additional disturbance agents and develop models that better capture the small variations in the spectral reflectance caused by low-severity disturbances, especially in the case of low-intensity partial harvests.
... Multitemporal optical data have been the most widely used remote sensing platform because they directly track changes in forest reflectance, easily detecting full canopy removals (Hansen et al. 2013;Jarron et al. 2016;Stahl et al. 2023). Despite issues with resolution, clouds, phenology, and changes in sensors, multispectral optical imagery has provided some of the most important and widely used data on forest cover and landscape configuration of forests (Banskota et al. 2014;Pasquarella et al. 2016;Senf et al. 2017;Tortini et al. 2019;White et al. 2017). Despite these successes, optical data do not directly quantify forest structure and instead rely on changes in the relative cover of photosynthetic vegetation (e.g. ...
... about forest inventories that are often poorly monitored (Phillips and Blinn 2007). For example, accurately tracking and inventorying forest activity across private lands is notoriously difficult (Cohen, Spies, and Fiorella 1995;Tortini et al. 2019;Wicker 2002), yet forests on private lands provide vast economic and environmental benefits that cannot be quantified or optimized in the absence of high-quality data (Ciuzio et al. 2013;Litvaitis et al. 2021;Phillips and Blinn 2007). ...
Effective planning for natural resource management and wildlife conservation requires detailed information on vegetation structure at landscape scales and how structure is influenced by land-use practices. In many forested landscapes, the largest impacts of land use on forest structure are driven by forest management activities, which can include invasive species control, prescribed fire, partial harvests (e.g. shelterwood harvests) or overstory removals and clearcuts. Active timber management is often used to achieve forest conservation objectives, but to be used effectively, managers require knowledge of harvest frequency and extent in adjacent forests and at landscape scales. In this paper, we develop a timber harvest mapping workflow using machine learning (XGBoost algorithm) and single campaign airborne light detection and ranging (LiDAR) surveys for the state of Pennsylvania, USA. We show that harvest type (shelterwood and overstory removals) can be mapped at high accuracy (overall accuracy = 94.9%), including broad age classes defined by the number of years since harvest. Errors of omission (false negatives) were lowest for recent (<10 yr old) overstory removal harvests (1.5%) and highest for older (10–18 yr old) shelterwood harvests (34.9%), which is consistent with the expectation that older, partial timber harvests are more difficult to distinguish from untreated forests than are recent harvests. Errors of commission (false positives) were low (<6.0%) for all timber harvest types and ages. Analysis of model results across both public and private lands in three highly forested conservation regions of Pennsylvania (the Poconos, PA Wilds, and Laurel Highlands) revealed a propensity for young overstory removals along forest edges, suggesting edge effects from inaccuracies in the underlying forest mask and mixed pixels contribute to errors of commission. Acknowledging this, overstory removal and shelterwood harvests were roughly equally common across public and private lands when expressed as a fraction of all interior forests (forests >30 m from an edge). The expectation that these harvest treatments would be rarer in private forests was not supported by the model results, which is likely due to the model’s inability to distinguish between alternative natural processes (weather damage, wildfire, pathogens, etc.) and forest treatment types (high-grading and firewood collection) that result in similar forest structures to the trained classes in the XGBoost model. This study provides a framework and validation for combining approachable machine-learning techniques with large-campaign LiDAR to accurately predict forest structure with application to a host of forestry, natural resource, and conservation-related problems. Future efforts that refine the model’s ability to better distinguish between more complex harvest classes and natural processes would be valuable.
... There is some pressure on urban planning to provide solutions in this way, where satellite-based remote sensing can help to overcome these obstacles [12]. Satellite-based remote sensing offers a long-term solution for monitoring landscape-scale land use and land cover change, as well as aid in the implementation of land management policies [13]. Through the synergy of satellite imagery (both radar and optical images), it is possible to conduct comprehensive monitoring of a territory, applying predictive measures through data analysis that help authorities apply protective measures and inhabitants improve their quality of life. ...
In our contemporary cities, infrastructures face a diverse range of risks, including those caused by climatic events. The availability of monitoring technologies such as remote sensing has opened up new possibilities to address or mitigate these risks. Satellite images allow the analysis of terrain over time, fostering probabilistic models to support the adoption of data-driven urban planning. This study focuses on the exploration of various satellite data sources, including nighttime land surface temperature (LST) from Landsat-8, as well as ground motion data derived from techniques such as MT-InSAR, Sentinel-1, and the proximity of urban infrastructure to water. Using information from the Local Climate Zones (LCZs) and the current land use of each building in the study area, the economic and climatic implications of any changes in the current features of the soil are evaluated. Through the construction of a Bayesian Network model, synthetic datasets are generated to identify areas and quantify risk in Barcelona. The results of this model were also compared with a Multiple Linear Regression model, concluding that the use of the Bayesian Network model provides crucial information for urban managers. It enables adopting proactive measures to reduce negative impacts on infrastructures by reducing or eliminating possible urban disparities.
... Accuracy assessment studies of the GFCC "forest loss" layer on temperate forests report underestimation of GFCC (e.g. Linke et al. 2017;Tortini et al. 2019). The underestimation in our study appears larger than in the above mentioned studies; nevertheless, in contrast with the Nevado de Toluca area, logging practices in temperate forested areas of the USA and Canada occur in the form of clear-cutting of plots mostly larger than 1 ha. ...
In 2013, the protection status of the Nevado de Toluca Volcano was changed from National Park to a less restrictive category. Much controversy has arisen surrounding this decision as the new category allows forest harvesting practices in most of the natural protected area. We assessed the forest cover loss in the Nevado de Toluca protected area and a 20 km buffer zone around it, before and after the change in protection status, and we established a map of forest types in view of assessing biodiversity vulnerability to the protection change in the forest harvesting areas. The forest type classification was carried out applying a Random Forest algorithm to Sentinel-2 imagery. The forest cover loss was assessed using the Landsat-based Global Forest Watch product during the period 2001-2019. Forest cover loss increased after the change of the Nevado de Toluca protection status; 39.7 hec-tares/year (± 16.7 ha/year) between 2001 and 2013 versus 106.0 hectares/year (± 40.2 ha/ year) between 2014 and 2019. Clear-cut deforestation occurred in 49.6 hectares during year 2018. According to our forest map, 7171 ha (54.8%) of the Abies religiosa forest and 6511 ha (32.5%) of the pine forests of the Nevado de Toluca protected area are inside the forest harvesting zonation. Also, forest cover loss around the natural protected area has increased, especially in 2018, indicating that allowing forest harvesting practices in the natural protected area has not reduced the pressure in the surrounding forests either. We ponder the consequences of the category change and give some suggestions for the future of the forests in the Nevado de Toluca protected area.
... These changes affect optical properties and their interaction within the forests, and in turn spectral reflectance measured by satellite sensors. In particular, vegetation indices (VIs), such as the normalized difference vegetation index (NDVI) [16] and the normalized burned ratio (NBR) [17][18][19][20], have been used to monitor disturbance in forest. NBR was originally proposed to detect areas burned by forest fire based on the change in the ratio between near-infrared (NIR) and shortwave infrared (SWIR) reflectance [21]. ...
The high forests in southwest Ethiopia, some of the last remaining Afromontane forests in the country, are home to significant forest coffee production. While considered as beneficial in maintaining forests, there have been growing concerns about the degradation caused by intensive management for coffee production in these forests. However, no suitable methods have been developed to map the coffee forests. In this study, we developed a tie-point approach to consistently estimate the degree of degradation caused by intensive management by combining use of Landsat imagery with in-situ canopy cover and tree survey data. Our results demonstrate a clear distinction between undisturbed natural forest and heavily managed coffee forest due to changes in forest structure and canopy cover caused by intensive management in the coffee forest. Temporal analysis of 32 years of Landsat imagery reveals a progressive and significant transition in the level of degradation in the coffee forest over this period. This is the first time to our knowledge, that this progressive intensification of coffee forest has been measured. There is a major intensification in the mid-1990s, which follows the introduction of new liberal economic policies by the Federal government established in 1991, rising coffee prices, and changes in state control over access to the forest. The question remains as to how these 20 years of intensive management in coffee forest have affected forest biodiversity and, more importantly, how canopy trees in this forest can be regenerated in the future. This study provides potential satellite-based mapping and ground-based photography and tree survey methods to help investigate the impacts of intensive management within coffee forest on biodiversity and regeneration.
... However, to the best of our knowledge, studies of forest parcelization using remote sensing technology are very limited. A recent study used multi-year Landsat images to identify harvesting events for private forests in Michigan from 1985 to 2011 [21]. Third, most of the existing studies focus on highly parcelized forests or compare forests separated by long spatial distance [9,22,23]. ...
Forestland parcelization (i.e., a process by which large parcels of forestland ownership are divided into many small parcels) presents an increasing challenge to sustainable forest development in the United States. In Southeastern Ohio, forests also experienced intensive forestland parcelization, where the majority of forest owners own parcels smaller than 10 acres currently. To better understand the impact of forestland parcelization on forest development, this study employed multi-source remotely sensed data and land ownership data in Hocking County, Ohio to examine the relationship between forestland parcel size and forest attributes, including forest composition and structure. Our results show that private forestland parcels are generally smaller than public forestland (the average parcel sizes are 21.5 vs. 275.0 acres). Compared with private lands, public lands have higher values in all forest attributes, including forest coverage, abundance of oak-dominant stands, canopy height and aboveground biomass. A further investigation focusing on private forestland reveals that smaller parcels tend to have smaller forest coverage, less greenness, lower height and aboveground biomass, indicating that forests in smaller parcels may experience more human disturbances than larger parcels. The results also show that logarithmic models can well quantify the non-linear relationship between forest attributes and parcel size in the study area. Our study suggests that forestland parcelization indeed has negative effects on forest development, so it is very important to take appropriate measures to protect forests in small ownership parcels.
... Furthermore, it may be difficult to determine any changes in management once they do occur, since tracking FFO timber harvests is possible only for those ownerships enrolled in voluntary tax incentive programs or using remote sensing techniques. Remote sensing methods may be the most reliable and promising means to track timber harvests over time, as demonstrated by a 2018 study that used LandSat imagery to track changes in forest cover on FFO parcels (Tortini, Mayer, Hermosilla, Coops, & Wulder, 2018) FFOs who ranked timber as an important reason for owning forestland, through the results of this survey and anecdotal evidence, appear to be entrenched with single tree selection as their method of choice, even though this method may not be best achieving their preferences. Convincing those stakeholders to consider other management options such as clearcuts, shelterwoods, or a different silvicultural method, may be challenging. ...
In the northern hardwood forests of the Western Upper Peninsula of Michigan, single-tree selection is the most commonly used silvicultural system. This system provides both a sustained yield of timber and attempts to emulate the windfall disturbance regime that determines the uneven aged structure of northern hardwood forests. However, with concerns about tree species diversity loss and a lack of early successional forests, even-aged regeneration methods are likely to become an increasingly crucial tool in the toolbox for managing northern hardwood forests of the Lake States. The forests of the Western Upper Peninsula are comprised of a mosaic of ownerships, with nearly 40% of the forested land owned by family forest owners. This study assessed family forest owners experience with, perceptions of, and interest in three different silvicultural methods recently implemented as a part of a long-term silviculture study on Michigan Technological University’s Ford Forest. A mailed survey with images and descriptions of clearcut, shelterwood, and single tree selection harvests was sent to family forest owners with at least 20 acres of forestland in the Western Upper Peninsula. The survey also included questions about landowners' use of their forestland, management experience, incentive-program enrollment, and demographics. The findings indicated that family forest owners who rank timber as important reason for owning forestland are the most likely to be accepting of all three methods. Respondents who use their land for hunting are also very likely to implement clearcut and shelterwood methods. Other findings include different preferences for management between absentee and non-absentee landowners. The results of this study suggest that targeting hunting groups and actively managing landowners may have a positive effect on the understanding and acceptance of silviculture among family forest owners.
The objective of this study was to investigate the utility of Landsat time series data for the detection of harvested area using annual data on the main island of Kyushu. We used a change detection algorithm to extract variables for the classification of harvesting, other disturbance, stable forest, and other land cover change. The estimated annual harvested area from 1985 to 2017 was 7426.5 ha, on average, with an increasing trend in the last decade. The overall, producer’s, and user’s accuracies were 95.2％ (±0.5％), 83.1％ (±2.9％), and 93.8％ (±0.9％), respectively, based on the accuracy assessment using satellite images. When the accuracy was assessed based on ground survey data, 87.6％ of the harvested area was detected by annual Landsat time series data. When the accuracy was assessed based on statistical data, the estimated harvested area was 2.6％ larger than the recorded clearcut area in national forests and 30.5％ larger than the recorded clearcut area in private conifer forests in Miyazaki Prefecture. This study demonstrated that using annual Landsat time series data is a viable approach for the detection of harvesting in a large area across several decades.
This special issue of Landscape & Urban Planning (LAND) “Landscape dynamics of family forest owners” represents a collection of approaches to describe and understand how family forest owners around the world, in the aggregate, influence and are influenced by landscape-scale land use dynamics. Also known as smallholders, small-scale owners, communal owners, or nonindustrial private forest owners (although these terms are not strictly interchangeable; Harrison et al., 2002, Fischer et al., 2010), family forest owners are the primary land managers for forest holdings which vary in size from a few acres to hectares. In both industrial and developing countries, the decisions these landowners make are influenced by both local and large-scale economic, environmental, and social processes, such as market price fluctuations, climate change, invasive pest spread, ownership parcelization, and changes in land tenure laws. And conversely, these individual-level land use decisions impact forest patterns at a variety of spatial and temporal scales, from village to continent.
There is a pressing need to map changes in forest structure from the earliest time period possible given forest management policies and accelerated disturbances from climate change. The availability of Landsat data from over four decades helps researchers study an ecologically meaningful length of time. Forest structure is most often mapped utilizing lidar data, however these data are prohibitively expensive and cover a narrow temporal window relative to the Landsat archive. Here we describe a technique to use the entire length of the Landsat archive from Multispectral Scanner to Operational Land Imager (M2O) to produce three novel outcomes: (1) we used the M2O dataset and standard change vector analysis methods to classify annual forest structure in northwestern Montana from 1972 to 2015, (2) we improved the accuracy of each yearly forest structure classification by applying temporal continuity rules to the whole time series, with final accuracies ranging from 97% to 68% respectively for two and six-category classifications, and (3) we demonstrated the importance of pre-1984 Landsat data for long-term change studies. As the Landsat program continues to acquire Earth imagery into the foreseeable future, time series analyses that aid in classifying forest structure accurately will be key to the success of any land management changes in the future.
The use of time series satellite data allows for the temporally dense, systematic, transparent, and synoptic capture of land dynamics over time. Subsequent to the opening of the Landsat archive, several time series approaches for characterizing landscape change have been developed, often representing a particular analytical time window. The information richness and widespread utility of these time series data have created a need to maintain the currency of time series information via the addition of new data, as it becomes available. When an existing time series is temporally extended, it is critical that previously generated change information remains consistent, thereby not altering reported change statistics or science outcomes based on that change information. In this research, we investigate the impacts and implications of adding additional years to an existing 29-year annual Landsat time series for forest change. To do so, we undertook a spatially explicit comparison of the 29 overlapping years of a time series representing 1984–2012, with a time series representing 1984–2016. Surface reflectance values, and presence, year, and type of change were compared. We found that the addition of years to extend the time series had minimal effect on the annual surface reflectance composites, with slight band-specific differences (r ≥ 0.1) in the final years of the original time series being updated. The area of stand replacing disturbances and determination of change year are virtually unchanged for the overlapping period between the two time-series products. Over the overlapping temporal period (1984–2012), the total area of change differs by 0.53%, equating to an annual difference in change area of 0.019%. Overall, the spatial and temporal agreement of the changes detected by both time series was 96%. Further, our findings suggest that the entire pre-existing historic time series does not need to be re-processed during the update process. Critically, given the time series change detection and update approach followed here, science outcomes or reports representing one temporal epoch can be considered stable and will not be altered when a time series is updated with newly available data.
Landscape ecology is a discipline that explicitly considers the influence of time and space on the environmental patterns we observe and the processes that create them. Although many of the topics studied in landscape ecology have public policy implications, three are of particular concern: climate change; land use–land cover change (LULCC); and a particular type of LULCC, urbanization. These processes are interrelated, because LULCC is driven by both human activities (e.g., agricultural expansion and urban sprawl) and climate change (e.g., desertification). Climate change, in turn, will affect the way humans use landscapes. Interactions among these drivers of ecosystem change can have destabilizing and accelerating feedback, with consequences for human societies from local to global scales. These challenges require landscape ecologists to engage policymakers and practitioners in seeking long-term solutions, informed by an understanding of opportunities to mitigate the impacts of anthropogenic drivers on ecosystems and adapt to new ecological realities.
Monitoring is a key aspect of payments for ecosystem services (PES) schemes, providing a basis for payments. PES monitoring however presents challenges, including in balancing technical accuracy with cost, local equity and legitimacy. This is particularly true in smallholder carbon PES, where managers have limited resources and capacity. Here we explore ways to improve monitoring in smallholder projects. We looked at two well-established projects in Uganda and Mexico, and appraised five monitoring methodologies: two remote sensing and three field measurement approaches. Each methodology varied in data resolution, methodological complexity and degree of local participation. We collected quantitative and qualitative information on four aspects of performance: accuracy; costs; local equity; and local legitimacy. We show that methodologies with greater data resolution and local participation performed better in all four aspects, while greater methodological complexity was not associated with significantly improved performance. We conclude that monitoring in smallholder and other types of PES may be improved through: 1) devolving analyses to the local level; 2) communicating to stakeholders a distinction between ‘applied’ and ‘scientific’ accuracy; and 3) documenting and communicating the diverse functions of monitoring, referred to here as co-benefits – a contrast to simple ‘monitor and pay’ conceptions of PES.
Disturbance is a critical ecological process in forested systems, and disturbance maps are important for understanding forest dynamics. Landsat data are a key remote sensing dataset for monitoring forest disturbance and there recently has been major growth in the development of disturbance mapping algorithms. Many of these algorithms take advantage of the high temporal data volume to mine subtle signals in Landsat time series, but as those signals become subtler, they are more likely to be mixed with noise in Landsat data. This study examines the similarity among seven different algorithms in their ability to map the full range of magnitudes of forest disturbance over six different Landsat scenes distributed across the conterminous US. The maps agreed very well in terms of the amount of undisturbed forest over time; however, for the ~30% of forest mapped as disturbed in a given year by at least one algorithm, there was little agreement about which pixels were affected. Algorithms that targeted higher-magnitude disturbances exhibited higher omission errors but lower commission errors than those targeting a broader range of disturbance magnitudes. These results suggest that a user of any given forest disturbance map should understand the map’s strengths and weaknesses (in terms of omission and commission error rates), with respect to the disturbance targets of interest.
Sustainable forest management practices allow for a range of harvest prescriptions, including clearcut, clearcut with residual, and partial or selective cutting, which are largely distinguished by the amount of canopy cover removed. The different prescriptions are aimed to emulate natural disturbance, encourage regeneration (seed trees), or offer other ecosystem services, such as the maintenance of local biodiversity or habitat features. Using remotely sensed data, stand-replacing disturbance associated with clearcutting is commonly accurately detected. Novel time series-based change detection products offer an opportunity to determine the capacity to detect and label a wider range of harvest practices. In this research, we demonstrate the capacity of time series imagery, spectral metrics, and related attributed change products, to distinguish between different harvesting practices over a study area in central British Columbia, Canada. Producer's accuracy of harvest attribution was 79%, with 93% of harvest blocks >5 ha accurately identified. In relation to the amount of canopy cover removed, clearcut harvesting was the most accurately classified (84%), followed by clearcut with residual (79%), and partial cut (64%). Applying detailed spectral metrics derived from Landsat data revealed clearcut and partial cuts to be spectrally distinct. The annual nature of the Landsat time series also offers spatial harvest information within typical, often decadal, forest inventory update cycles. The statistically significant (p < 0.05) relationship between harvest practices and Landsat spectral information indicates a capacity to add increased attribution richness to remote sensing depictions of forest harvest.
Ownership of private forestland is changing rapidly, especially in areas owned by industrial forest product companies. Following divestment by industrial owners, forested lands are increasingly likely to move from intensive timber production to subdivision of the land, development and other private uses, or alternatively to conservation under public tenure. This research follows a unique dataset of forestland properties previously owned by industrial and corporate owners in Wisconsin from 1999 to 2014. A multinomial logistic regression showed that divested lands were more likely to be publicly purchased outright if they were adjacent to water, adjacent to public land, not adjacent to roads, and had higher housing value, while in contrast the predictors of conservation easement acquisition were location in large blocks outside of zoned townships. Properties were more likely to be parcelized if they were adjacent to a paved road, adjacent to water, smaller, in a zoned township, and had fewer years remaining in their tax program enrollment. In an era of rapid industrial land divestment, these findings indicate an important role played by public policies, including preferential tax programs and funds for land and conservation easement acquisition, in shaping whether private forestland is parcelized, conserved under private ownership, or publicly acquired.
Once driven by large-scale clearings, Amazon deforestation now occurs mostly in small increments. Did this result from the emergence of a new group of agents or from a strategic adaptation in the behavior of those who led deforestation in the past? We address this question using georeferenced data on private rural properties and deforestation. We cross property-level and forest clearing data in an empirical setting designed to detect shifts toward clearing patches that were knowingly invisible to the monitoring system. We are therefore able to assess not only whether deforesters were responding strategically to stricter monitoring of deforestation, but also how this response differed across actor types. Results suggest that centralized policy efforts introduced starting in the mid-2000s inhibited medium- and large-scale deforestation, but had heterogeneous effects on small-scale deforestation. Although the relative participation of small deforestation polygons increased in both sample states, the relative participation of smallholders in total state deforestation increased in Pará, while remaining constant in Mato Grosso. We interpret these results as suggestive — albeit not causal — evidence that landholders strategically responded to the monitoring system by adapting their forest clearings practices to elude monitoring in both Mato Grosso and Pará. In the latter, however, the increase in smallholders’ share of annual deforestation suggests that their clearing practices were relatively less affected by what effectively contained deforestation in large properties. The apparent similarity in scale of deforestation across states conceals relevant baseline differences between the agents engaging in forest clearing in each locality. Tailoring policy to account for such differences could strengthen Brazilian conservation policy.
There are an estimated 10.7 million family forest ownerships across the United States who collectively control 36% or 290 million acres of the nation’s forestland. The US Department of Agriculture Forest Service National Woodland Owner Survey (NWOS) provides information on the characteristics, attitudes, and behaviors of these ownerships. Between 2011 and 2013, 8,576 randomly selected family forest ownerships with at least 10 acres of forestland participated in the NWOS. Results show: amenity values are the dominant reasons for owning; owners tend to be active on their land, but most are not engaged in traditional forestry programs; and owners are relatively old. Although the general ownership patterns and reasons for owning are the same between the 2002–2006 and current iterations of the NWOS, participation in some management activities changed (some increased and some decreased) and the percentage of female primary decisionmakers increased.