Interaction of multiple disturbances: importance of disturbance interval in the effects of fire on rehabilitating mined areas

Abstract Multiple disturbance regimes are increasingly common as novel anthropogenic disturbances are added to existing natural disturbances. However, it is generally unknown whether simultaneous or sequential effects of different forms of disturbance are predictable from the independent effects of each disturbance. This study examines the short‐term effects of sequential disturbance by mineral sand‐mining followed by fire in a forest community in south‐eastern Australia. Four combinations of disturbance were sampled: unburned mined, burned mined, unburned forest (unmined) and burned forest (unmined, with between‐fire interval matching the disturbance interval between mining and fire of the burned mined treatment). All combinations were sampled approximately 12 months following fire on the burned sites. The impact of fire after mining depended on disturbance interval. Sites burned 0.5–2.4 years since mining had fewer native vascular plant species than unburned mined sites of the same mined age, whereas sites with 10–16 years or 20–26 years between mining and fire had greater native species richness than unburned mined sites of the same age. Burning 20–26 years after mining brought native species richness within the range of burned forest. For both unmined and mined sites native seedling densities increased with burning, and with longer disturbance intervals. Weed species richness and weed seedling densities were greater on mined sites than in forest, and burning mined sites elevated weed seedling densities further, particularly for short intervals. Both disturbance interval and fire intensity are likely to have contributed to these results, as intensity on mined areas increased with interval, and at 20–26 years post‐mining was equivalent to unmined forest. These results suggest that fire could be used to promote rehabilitation of these mined areas after at least 10 years, but should be excluded from earlier stages of post‐mining regeneration. However, other sources of spatial and temporal variability should be considered in addition to interval and intensity, as variation among mined areas was correlated with post‐fire weather conditions and available weed sources. Finally, the combined effects of mining and fire could not be predicted from knowledge of the disturbances operating separately, indicating that effects of multiple disturbance may be synergistic rather than additive.     

我国林火发生预测模型研究进展

通过文献回顾,总结了国内林火发生预测模型的研究现状,并从林火发生驱动因子、林火发生概率预测模型、林火发生频次预测模型和模型检验方法等方面进行归纳分析。得出以下结论: 1)气象、地形、植被、可燃物、人类活动等因素是影响林火发生及模型预测精度的主要驱动因子;2)林火发生概率模型中,地理加权逻辑斯蒂回归模型考虑了变量之间的空间相关性,Gompit回归模型适宜非对称结构的林火数据,随机森林模型不需要多重共线性检验,在避免过度拟合的同时提高了预测精度,是林火发生概率预测模型的优选方法之一;3)林火发生频次模型中,负二项回归模型更适合对过度离散数据进行模拟,零膨胀模型和栅栏模型可以处理林火数据中包含大量零值的问题;4)ROC检验、AIC检验、似然比检验和Wald检验方法是林火概率和频次模型的常用检验方法。林火发生预测模型研究仍是我国当前林火管理工作的重点,预测模型的选择需要依据不同地区林火数据特点。此外,构建林火预测模型时需要考虑更多的影响因素,以提高模型预测精度;未来,需要进一步探索其他数学模型在林火发生预测中的应用,不断提高林火发生预测模型的准确度。     

Flammability dynamics in the Australian Alps

Forests of the Australian Alps (SE Australia) are considered some of the most vulnerable to climate change in the country, with ecosystem collapse considered likely for some due to frequent fire. It is not yet known, however, whether increasing fire frequency may stabilize due to reductions in flammability related to reduced time for fuel accumulation, show no trend, or increase due to positive feedbacks related to vegetation changes. To determine what these trends have been historically, dynamics were measured for 58 years of mapped fire history. The 1.4 million ha forested area was divided into broad formations based on structure and dominant canopy trees, and dynamics were measured for each using flammability ratio, a modification of probability of ignition at a point. Crown fire likelihood was measured for each formation, based on satellite‐derived measurements of the 2003 fire effects across a large part of the area. Contrary to popular perception but consistent with mechanistic expectations, all forests exhibited pronounced positive feedbacks. The strongest response was observed in tall, wet forests dominated by Ash‐type eucalypts, where, despite a short period of low flammability following fire, post‐disturbance stands have been more than eight times as likely to burn than have mature stands. The weakest feedbacks occurred in open forest, although post‐disturbance forests were still 1.5 times as likely to burn as mature forests. Apart from low, dry open woodland where there was insufficient data to detect a trend, all forests were most likely to experience crown fire during their period of regeneration. The implications of this are significant for the Alps, as increasing fire frequency has the potential to accelerate by producing an increasingly flammable landscape. These effects may be semi‐permanent in tall, wet forest, where frequent fire promotes ecosystem collapse into either the more flammable open forest formation, or to heathland.     

Ecological thresholds at the savanna-forest boundary: how plant traits, resources and fire govern the distribution of tropical biomes

Fire shapes the distribution of savanna and forest through complex interactions involving climate, resources and species traits. Based on data from central Brazil, we propose that these interactions are governed by two critical thresholds. The fire-resistance threshold is reached when individual trees have accumulated sufficient bark to avoid stem death, whereas the fire-suppression threshold is reached when an ecosystem has sufficient canopy cover to suppress fire by excluding grasses. Surpassing either threshold is dependent upon long fire-free intervals, which are rare in mesic savanna. On high-resource sites, the thresholds are reached quickly, increasing the probability that savanna switches to forest, whereas low-resource sites are likely to remain as savanna even if fire is infrequent. Species traits influence both thresholds; saplings of savanna trees accumulate bark thickness more quickly than forest trees, and are more likely to become fire resistant during fire-free intervals. Forest trees accumulate leaf area more rapidly than savanna trees, thereby accelerating the transition to forest. Thus, multiple factors interact with fire to determine the distribution of savanna and forest by influencing the time needed to reach these thresholds. Future work should decipher multiple environmental controls over the rates of tree growth and canopy closure in savanna.     

Postfire response of North American boreal forest net primary productivity analyzed with satellite observations

Fire is a major disturbance in the boreal forest, and has been shown to release significant amounts of carbon (C) to the atmosphere through combustion. However, less is known about the effects on ecosystems following fire, which include reduced productivity and changes in decomposition in the decade immediately following the disturbance. In this study, we assessed the impact of fire on net primary productivity (NPP) in the North American boreal forest using a 17‐year record of satellite NDVI observations at 8‐ km spatial resolution together with a light‐use efficiency model. We identified 61 fire scars in the satellite observations using digitized fire burn perimeters from a database of large fires. We studied the postfire response of NPP by analyzing the most impacted pixel within each burned area. NPP decreased in the year following the fire by 60–260 g C m^?2 yr^?1 (30–80%). By comparing pre‐ and postfire observations, we estimated a mean NPP recovery period for boreal forests of about 9 years, with substantial variability among fires. We incorporated this behavior into a carbon cycle model simulation to demonstrate these effects on net ecosystem production. The disturbance resulted in a release of C to the atmosphere during the first 8 years, followed by a small, but long‐lived, sink lasting 150 years. Postfire net emissions were three times as large as from a model run without changing NPP. However, only small differences in the C cycle occurred between runs after 8 years due to the rapid recovery of NPP. We conclude by discussing the effects of fire on the long‐term continental trends in satellite NDVI observed across boreal North America during the 1980s and 1990s.     

The temporal structure of feeding behavior

Meals have long been considered relevant units of feeding behavior. Large data sets of feeding behavior of cattle, pigs, chickens, ducks, turkeys, dolphins, and rats were analyzed with the aims of 1) describing the temporal structure of feeding behavior and 2) developing appropriate methods for estimating meal criteria. Longer (between-meal) intervals were never distributed as the negative exponential assumed by traditional methods, such as log-survivorship analysis, but as a skewed Gaussian, which can be (almost) normalized by log-transformation of interval lengths. Log-transformation can also normalize frequency distributions of within-meal intervals. Meal criteria, i.e., the longest interval considered to occur within meals, can be estimated after fitting models consisting of Gaussian functions alone or of one Weibull and one or more Gaussian functions to the distribution of log-transformed interval lengths. Nonuniform data sets may require disaggregation before this can be achieved. Observations from all species were in conflict with assumptions of random behavior that underlie traditional methods for criteria estimation. Instead, the observed structure of feeding behavior is consistent with 1) a decrease in satiety associated with an increase in the probability of animals starting a meal with time since the last meal and 2) an increase in satiation associated with an increase in the probability of animals ending a meal with the amount of food already consumed. The novel methodology proposed here will avoid biased conclusions from analyses of feeding behavior associated with previous methods and, as demonstrated, can be applied across a range of species to address questions relevant to the control of food intake.     

Gyrase and Topo IV modulate chromosome domain size in vivo

In bacteria, DNA supercoil movement is restricted to subchromosomal regions or 'domains.' To elucidate the nature of domain boundaries, we analysed reaction kinetics for γδ site-specific resolution in six chromosomal intervals ranging in size from 14 to 90 kb. In stationary cultures of Salmonella typhimurium , resolution kinetics were rapid for both short and long intervals, suggesting that random stationary barriers occur with a 30% probability at approximately 80 kb intervals along DNA. To test the biochemical nature of domain barriers, a genetic screen was used to look for mutants with small domains. Rare temperature-sensitive alleles of DNA gyrase and Topo IV (the two essential type II topoisomerases) had more supercoil barriers than wild-type strains in all growth states. The most severe gyrase mutants were found to have twice as many barriers in growing cells as wild type throughout a 90 kb interval of the chromosome. We propose that knots and tangles in duplex DNA restrain supercoil diffusion in living bacteria.     

不同区域森林火灾对生态因子的响应及其概率模型

火灾是影响森林生态系统过程的重要干扰之一,其对森林生态系统内各生态因子的响应各不相同.由于植被状况及生态环境的不同,森林火灾的时空分布特征在中国不同植被气候类型内表现不同,根据植被气候类型分类系统,将中国主要森林火灾地区划分为4个区域:东北(冷温带松林)、华北(落叶阔叶林)、东南(常绿阔叶林)和西南(热带雨林),应用遥感监测数据和地面环境数据,以时空变量、生态因子(植被生长变化指数、湿度等)为可选自变量,应用半参数化Logistic回归模型,就森林火险对不同生态影响因子的响应规律进行了分析,建立了基于生态因子的着火概率模型和大火蔓延概率模型,通过模拟及实际数据散点图、火险概率图,评估了模型应用价值.结果表明,土壤湿度及植被含水量在落叶阔叶林、常绿阔叶林、热带雨林地区对着火概率影响显著.在4个植被气候区内,土壤及凋落物湿度对大火蔓延的作用较小.在冷温带松林、落叶阔叶林、常绿阔叶林地区,植被生长的年内变化对火灾发生的影响显著,在常绿阔叶林地区,年内植被生长变化对大火蔓延的作用较小.森林火险概率与各生态因子的相关关系主要呈现出非线性.不同植被气候区内,火险概率受不同生态因子组合的影响,这与不同区域的植被状况及生态环境不同有关.在不同植被气候类型,应用时空变量、生态因子建立半参数化logistic回归模型,进行着火概率和大火蔓延概率的模拟具有可行性和实际应用能力.为进一步分析森林生态系统与火灾之间的动态关系、展开生态系统火灾干扰研究提供了理论基础.     

The effect of fire on an abstract forest ecosystem: An agent based study

Our model considers a new element in forest fire modeling, namely the dynamics of a forest animal, intimately linked to the trees. We show that animals and trees react differently to different types of fire. A high probability of fire initiation results in several small fires, which do not allow for a large fuel accumulation and thus the destruction of many trees by fire, but is found to be generally devastating to the animal population at the same time. On the other hand, a low fire initiation probability allows for the accumulation of higher quantities of fuel, which in turn results in larger fires, more devastating to the trees than to the animals. Thus, we suggest that optimal fire management should take into account the relation between fire initiation and its different effects on animals and trees. Further, wildfires are often considered as prime examples for power-law-like frequency distributions, yet there is no agreement on the mechanisms responsible for the observed patterns. Our model suggests that instead of a single unified distribution, a superposition of at least two different distributions can be detected and this suggests multiform mechanisms acting on different scales. None of the discovered distributions are compatible with the power-law hypothesis.     

Fire regimes of World Heritage Kakadu National Park., Australia

Abstract LANDSAT Multi‐Spectral Scanner imagery was used to determine aspects of the fire regimes of Kakadu National Park (in the wet‐dry tropics of Australia) for the period 1980–1995. Three landscape types recognized in this Park were Plateau, Lowlands and Floodplain. Areas burned in early and late dry seasons each year were documented using a Geographical Information System. Regression analyses were used to examine time trends in the areas burned each year and the interrelationships between early and late dry season burning. The proportions of landscapes having different stand ages (years since fire), and the proportions having had different fire intervals, were compared with results expected from the simplest random model (i.e. one in which the probability of ignition at a point [PIP] burning annually was constant). Using overlays of successive stand‐age maps, PIP could be calculated as a function of stand age. The Lowlands burned extensively each year; the areas burned by late dry season fires adding to those burned in the early dry season such that around 50–60% of the total area burned annually. Early dry season fires have lower intensities than late dry season fires, on average. Using a theoretically constant PIP and the mean proportion burned per year as the only input, predictions of areas burned as a function of stand age and fire interval were reasonable when compared with the empirical data, but best for the Lowlands landscape. PIP functions for Lowlands and Floodplains had negative slopes, an unexpected result. The nature of these PIP functions may reflect heterogeneity in fire proneness of the various vegetation types within landscapes. The scale of measurement, the scale of variation in vegetation types within a landscape, and the accuracy of the determination of burned areas, are constraints on the accuracy of fire‐interval and seasonally determination perceived from an analysis of satellite data.     

Exact Confidence Bounds for Comparing Two Regression Lines with a Control Regression Line on a Fixed Interval

The problem of finding exact simultaneous confidence bounds for comparing simple linear regression lines for two treatments with a simple linear regression line for the control over a fixed interval is considered. The assumption that errors are iid normal random is considered. It is assumed that the design matrices for the two treatments are equal and the design matrix for the control has the same number of copies of each distinct row of the design matrix for the treatments. The method is based on a pivotal quantity that can be expressed as a function of four t variables. The probability point depends on the size of an angle associated with the interval. We present probability points for various sample sizes and angles. (© 2004 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)     

Post‐fire tree establishment patterns at the alpine treeline ecotone: Mount Rainier National Park,Washington, USA

Questions: Does tree establishment: (1) occur at a treeline depressed by fire, (2) cause the forest line to ascend upslope, and/or (3) alter landscape heterogeneity? (4) What abiotic and biotic local site conditions are most important in structuring establishment patterns? (5) Does the abiotic setting become more important with increasing upslope distance from the forest line? Location: Western slopes of Mount Rainier, USA. Methods: We performed classification analysis of 1970 satellite imagery and 2003 aerial photography to delineate establishment. Local site conditions were calculated from a LIDAR‐based DEM, ancillary climate data, and 1970 tree locations in a GIS. We used logistic regression on a spatially weighted landscape matrix to rank variables. Results: Considerable establishment after 1970 caused forest line elevation to increase over 150 m in specific locations. Landscape heterogeneity increased with distance from the 1970 forest line. At a broad spatial context, we found establishment was most common near existing trees (0‐50 m) and at low elevations (1250‐1350 m). Slope aspect (W, NW, N, NE, and E), slope angle (40‐60°), and other abiotic factors emerged as important predictors of establishment with increasing upslope distance from the forest line to restricted spatial extents. Conclusions: Favorable climatic conditions likely triggered widespread tree establishment. Readily available seed probably enhanced establishment rates near sexually mature trees, particularly in the less stressful environment at low elevations. The mass effect of nearly ubiquitous establishment in these areas may have obscured the importance of the abiotic setting to restricted spatial extents. Topographic variability apparently produced favorable sites that facilitated opportunistic establishment with increasing upslope distance from the forest line, thereby enabling additional trees to invade the alpine tundra.     

Error in the inference of fire history from grasstrees

Abstract A method for constructing fire histories has recently been proposed for fire‐prone southern Australia based on the pattern of coloured bands in the remnant leaf‐bases on stems of grasstrees (Xanthorrhoea species; Xanthorrhoeaceae). In the absence of alternative high‐resolution techniques extending into the premodern period, this method has been utilized to construct fire histories for forest, woodland and shrubland ecosystems, principally in south‐western Australia. However, the technique has not been validated against known fire histories spanning more than one fire interval. Here we compare fire records from 100 grasstrees with a 30‐year record (1973–2002) of fire data derived from satellite imagery in a region of sandplain shrubland vegetation near Eneabba in south‐western Australia. Fires occurred in eight of the 30 years of the satellite record, with sampled grasstrees burning between zero and four times. The grasstree and satellite records agreed in terms of the overall incidence of fires experienced over the 30‐year period, with the grasstree record matching the satellite record significantly better than chance. However, comparison of the grasstree and satellite records found substantial error in the rate of both false positives and false negatives. Grasstrees failed to identify fire in 83% of fire occurrences identified by the satellite record, down to 53% if an error of ±2 years in the attribution of year of fire was allowed. A similar proportion of grasstree fire incidents were not matched in the satellite record (false positives). The rate of false positives increased with time before present, suggesting a temporal bias in the grasstree record. It is clear that the grasstree record does reflect fire history to a degree, but that it contains at least as many false as true fire records and may tend towards over‐reporting the incidence of fire in the past.     

An application of plant functional types to fire management in a conservation reserve in southeastern Australia

Abstract. The vital attribute system of Noble & Slatyer (1980) was used to classify the fire‐prone flora of Brisbane Water National Park (New South Wales, Australia) into plant functional types (PFTs), reflecting sensitivity to fire frequency (intervals between fire). A variety of information was used to assess the vital attributes of species in the predominant woodland/open‐forest vegetation within the Park. This was sufficient to allocate 54% of the species to functional types. Ca. 20% of the species belonged to PFTs defined as sensitive to either frequent or infrequent fire (e.g. obligate seeder types). Varied methods, based on the nature and quality of data were used to estimate juvenile periods and life spans among species in these types, however the estimates derived in each case were similar. On this basis, a domain of ‘acceptable’ fire intervals (7 to 30 yr) was derived for the woodland/open‐forest vegetation. Given the overall proportion of species considered, plus congruence between differing methods and sources of data, this domain was relatively robust. A landscape analysis using the domain indicated that the current trend in fire intervals, across the Park, may be adverse to floristic conservation.     

A comparative study of interspecific variation in fruit size among Australian eucalypts

We examined variation in woody fruit size among 362 Australian Eucalyptus species with respect to predictions relating fruit size to fire exposure and rainfall. Predictions for fruit size variation were established that focussed on selection for small or large seeds, given a positive allometric relationship between fruit and seed size within the genus, and on the potential for fruits to protect their valuable seed contents. Comparatively smaller fruits were found in species that continually experience frequent disturbance by fire, while both small and large fruits were found among species subjected to both short and long fire intervals. In the latter case where a broad range of fire intervals is possible, some species have adopted a strategy of producing small seeds that provide superior colonisation ability in disturbed conditions, while other species have adopted a strategy of producing large seeds which are more competitive during longer intervals between disturbance by fire. Only when taxonomic membership at the subgeneric level was accounted for in analyses across all species, did a significant relationship emerge between fruit size and rainfall independently of fire interval and plant height; comparatively larger fruits were found in species experiencing lower average annual rainfall in the subgenera Eucalyptus and Symphyomyrtus. In contrast to previous studies, larger fruits were found only in short species, while small fruits were found in both short and tall species. Many short species have adopted a strategy of protecting their seeds from high fire intensity by producing larger fruit. Since tall species can elevate their fruit far above high fire intensity, they make considerable energy savings by producing smaller fruit. It remains an open question as to why small fruit size occurs in some short species, but we suggest that these species may invest more heavily in vegetative regrowth after fire than in re-establishment by seed.     

Gradient Analysis of Fire Regimes in Montane Forests of the Southern Cascade Range,Thousand Lakes Wilderness,California, USA

Species distribution and abundance patterns in the southern Cascades are influenced by both environmental gradients and fire regimes. Little is known about fire regimes and variation in fire regimes may not be independent of environmental gradients or vegetation patterns. In this study, we analyze variation in fire regime parameters (i.e., return interval, season, size, severity, and rotation period) with respect to forest composition, elevation, and potential soil moisture in a 2042 ha area of montane forest in the southern Cascades in the Thousand Lakes Wilderness (TLW). Fire regime parameters varied with forest composition, elevation, and potential soil moisture. Median composite and point fire return intervals were shorter (4-9 yr, 14-24 yr) in low elevation and more xeric white fir (Abies concolor)-sugar pine (Pinus lambertiana) and white fir-Jeffrey pine (P. jeffreyi) and longest (20-37 yr, 20-47 yr) in mesic high elevation lodgepole pine (Pinus contorta) and red fir (Abies magnifica)-mountain hemlock (Tsuga mertensiana) forests. Values for mid-elevation red fir-white fir forests were intermediate. The pattern for fire rotation lengths across gradients was the same as for fire return intervals. The percentage of fires that occurred during the growing season was inversely related to elevation and potential soil moisture. Mean fire sizes were larger in lodgepole pine forests (405 ha) than in other forest groups (103-151 ha). In contrast to other parameters, fire severity did not vary across environmental and compositional gradients and >50% of all forests burned at high severity with most of the remainder burning at moderate severity. Since 1905, fire regimes have become similar at all gradient positions because of a policy of suppressing fire and fire regime modification will lead to shifts in landscape scale vegetation patterns.     

Granulometric analysis of spots in DNA microarray images

As the topological properties of each spot in DNA microarray images may vary from one another, we employed granulometries to understand the shape-size content contributed due to a significant intensity value within a spot. Analysis was performed on the microarray image that consisted of 240 spots by using concepts from mathematical morphology. In order to find out indices for each spot and to further classify them, we adopted morphological multiscale openings, which provided microarrays at multiple scales. Successive opened microarrays were subtracted to identify the protrusions that were smaller than the size of structuring element. Spot-wise details, in terms of probability of these observed protrusions,were computed by placing a regularly spaced grid on microarray such that each spot was centered in each grid. Based on the probability of size distribution functions of these protrusions isolated at each level, we estimated the mean size and texture index for each spot. With these characteristics, we classified the spots in a microarray image into bright and dull categories through pattern spectrum and shape-size complexity measures. These segregated spots can be compared with those of hybridization levels.     

What is the Value of a Good Map? An Example Using High Spatial Resolution Imagery to Aid Riparian Restoration

Riparian areas contain structurally diverse habitats that are challenging to monitor routinely and accurately over broad areas. As the structural variability within riparian areas is often indiscernible using moderate-scale satellite imagery, new mapping techniques are needed. We used high spatial resolution satellite imagery from the QuickBird satellite to map harvested and intact forests in coastal British Columbia, Canada. We distinguished forest structural classes used in riparian restoration planning, each with different restoration costs. To assess the accuracy of high spatial resolution imagery relative to coarser imagery, we coarsened the pixel resolution of the image, repeated the classifications, and compared results. Accuracy assessments produced individual class accuracies ranging from 70 to 90% for most classes; whilst accuracies obtained using coarser scale imagery were lower. We also examined the implications of map error on riparian restoration budgets derived from our classified maps. To do so, we modified the confusion matrix to create a cost error matrix quantifying costs associated with misclassification. High spatial resolution satellite imagery can be useful for riparian mapping; however, errors in restoration budgets attributable to misclassification error can be significant, even when using highly accurate maps. As the spatial resolution of imagery increases, it will be used more routinely in ecosystem ecology. Thus, our ability to evaluate map accuracy in practical, meaningful ways must develop further. The cost error matrix is one method that can be adapted for conservation and planning decisions in many ecosystems.     

Landscape Dynamics in Northwestern Amazonia: An Assessment of Pastures,Fire and Illicit Crops as Drivers of Tropical Deforestation

Many studies have identified drivers of deforestation throughout the tropics and, in most cases, have recognised differences in the level of threat. However, only a few have also looked at the temporal and spatial dynamics by which those drivers act, which is critical for assessing the conservation of biodiversity as well as for landscape planning. In this study, we analyse land cover change between 2000 and 2009 in north-western Colombian Amazonia to identify the interactions between the use of fire, cultivation of illicit crops and establishment of pastures, and their impacts on the loss of forest in the region. Yearly analyses were undertaken at randomly selected sample areas to quantify the average areas of transition of land cover types under different landscape compositions: forest-dominated mosaics, pasture mosaics, fire mosaics, and illicit crop mosaics. Our results indicate that despite the fact that forest areas were well-preserved, deforestation occurred at a low annual rate (0.06%). Conversion to pasture was the main factor responsible for forest loss (the area of pastures tripled within forest mosaics over 8 years), and this process was independent of the landscape matrix in which the forests were located. In fire mosaics, burning is a common tool for forest clearing and conversion to pasture. Thus, forests in fire mosaics were highly disturbed and frequently transformed from primary to secondary forests. The use of fire for illicit cropping was not detected, partly due to the small size of common illicit crops. Forest regeneration from pastures and secondary vegetation was observed in areas with large amounts of natural forest. Overall, assuming the continuation of the observed pasture conversion trend and the use of forest fire, we suggest that our results should be incorporated into a spatially explicit and integrated decision support tool to target and focus land-planning activities and policies.