You are currently viewing a new version of our website. To view the old version click .
MDPIMDPI
Search all articles
Forests
Download PDF
  • Review
  • Open Access

11 November 2025

Artificial Intelligence in Forest Pathology: Opportunities and Challenges

1
Département des Sciences du Bois et de la Forêt, Faculté de Foresterie, Géographie et Géomatique, Université Laval, Quebec City, QC G1V 0A6, Canada
2
Institut de Biologie Intégrative et des Systèmes (IBIS), Université Laval, Quebec City, QC G1V 0A6, Canada
3
Centre D’étude de la Forêt (CEF), Université Laval, Quebec City, QC G1V 0A6, Canada
Forests2025, 16(11), 1714;https://doi.org/10.3390/f16111714
This article belongs to the Special Issue Innovative Techniques for Monitoring and Managing Invasive Forest Pests and Pathogens
Version Notes
Order Reprints

Abstract

Forest diseases threaten tree health, biodiversity, and ecosystem services, with impacts amplified by climate change and global trade. Understanding and managing these threats is difficult due to the longevity of trees, the size and inaccessibility of forests, and the often cryptic or delayed expression of symptoms. This review first introduces the field of forest pathology and the key challenges it faces, including multifactorial declines, root and vascular diseases, and emerging invasive pathogens. We then examine how artificial intelligence (AI) can be applied to biotic, abiotic, and decline-related diseases, integrating remote sensing, imaging, genomics, and ecological data across spatial and temporal scales. Lessons from agricultural systems are discussed, highlighting potential tools and pitfalls for forestry. Finally, we outline future directions, emphasizing the need for interpretable models, incorporation of ecological context, cross-species validation, and coordinated data infrastructures to ensure AI delivers actionable, scalable solutions for complex forest ecosystems.

1. Introduction to Forest Pathology and Artificial Intelligence

Forest pathology focuses on the study of diseases affecting woody plants and is rooted at the intersection of forestry, microbiology, plant physiology and ecology. Its foundation is often attributed to the German forester Robert Hartig, who first established the connection between wood decay and microbial activity [1]. Forest pathology is concerned with diseases occurring in natural and urban forests as well as plantations and tree nurseries [2]. Diseases in forests ecosystems can be broadly categorized as biotic, caused by living organisms such as fungi, bacteria, viruses, and oomycetes; abiotic, driven by non-living factors like drought, frost, pollution, or nutrient deficiencies; and declines, which result from the combined effects of multiple stressors leading to progressive tree dieback [3]. While natural disease processes play a critical role in forest renewal by removing old or weakened trees, large-scale epidemics can cause severe damage to host populations in both natural, plantation and urban settings [4]. Examples of devastating and well-known outbreaks include chestnut blight [5], ink disease [6], Dutch elm disease [7], and sudden oak death [8] in deciduous trees, as well as white pine blister rust [9], cypress canker disease [10], and Dothistroma needle blight [11] in conifers. These events not only disrupt ecosystem dynamics but can also lead to significant economic and ecological losses [12,13].
Climate change is reshaping host–pathogen interactions, driving shifts in pathogen distribution, incidence, and severity [14,15], while simultaneously rendering locally adapted tree populations less suited to their environments [15]. These dynamics highlight the urgency of proactive surveillance and management, as early detection and prevention are consistently shown to be more cost-effective and impactful than addressing well-established epidemics [15,16,17]. Historically, forest pathology has relied on conventional approaches such as visual inspection, symptom characterization, and pathogen isolation [18]. Although these methods remain foundational, they are labor-intensive, slow, and often technically challenging. For example, most biotrophic pathogens like rusts cannot be cultured outside a living host. Molecular biology has already revolutionized forest pathology by enabling precise identification and insights into host–pathogen interactions [19,20]. Artificial intelligence (AI) offers the potential for a similar paradigm shift, complementing traditional and molecular methods by providing efficiency, scalability, and predictive power scales previously unattainable.
Artificial intelligence refers to the simulation of human intelligence in machines that are programmed to think, learn, and solve problems autonomously. AI systems can analyze large datasets, identify patterns, and make decisions or predictions, regularly outperforming traditional methods in various scientific fields [21]. In forest pathology, AI applications typically fall into three categories: machine learning, which uses algorithms to analyze data and predict outcomes, such as disease outbreaks or pathogen spread [22], computer vision, which processes images from drones or satellites to detect tree health issues or disease symptoms [23] and natural language processing, which extracts insights from scientific literature or field reports to diagnose or identify emerging threats [24]. AI is particularly well-suited for analyzing large, heterogeneous datasets, ranging from high-resolution aerial imagery to genomic sequences [25,26,27]. Unlike traditional statistical methods, which often rely on predefined linear assumptions or other parametric constraints, AI algorithms can identify complex, nonlinear relationships within data. These capabilities make AI a powerful tool for capturing intricate interactions within forest pathology datasets. Several new technologies combining AI and remote sensing are now driving the emergence of the new field of “smart forestry” [26].
While AI has been successfully applied in agricultural disease management [28,29], forests present unique complexities. Greater species diversity, long tree lifespans, and vast spatial scales make disease identification and monitoring more challenging [30]. Nevertheless, as emerging threats become more frequent and severe under the influence of climate change, the urgency for innovative, AI-driven solutions in forest pathology has never been greater [17,31,32]. In this review, we frame the key challenges in forest pathology where AI could bring transformative advances, assess its current applications through recent case studies, highlight transferable lessons from crop science, and outline the future opportunities and pitfalls that researchers will need to navigate.

2. Challenges in Forest Pathology

2.1. The Complexity of Forest Diseases

Forest ecosystems are home to a diverse array of pathogens, including oomycetes, bacteria, viruses, nematodes and the major cause of diseases, fungi [33]. These pathogens vary widely in their biology, from generalists capable of infecting a wide range of hosts (e.g., Phytophthora ramorum and its 75 confirmed host species across 48 genera, https://www.aphis.usda.gov/sites/default/files/usdaprlist.pdf (accessed on 24 September 2025)) to specialists adapted to particular tree species or genera (e.g., Cronartium ribicola causing white pine blister rust restricted to the nine five-needle pines species [34]). The diversity of hosts in forests adds another layer of complexity, although species richness is generally thought to have a dilution effect on diseases, which can protect natural forests but also make diseases more difficult to detect and predict [35,36]. To the array of biotic agents are also abiotic stressors such as drought and high temperatures, which can add up and be exacerbated by climate change, leading to large scale forest diebacks [37,38]. The relationship between host and pathogen varies spatially, across the landscape, with the interplay of host distribution, site environmental conditions, leading to the emergence of the discipline of landscape pathology [39]. The interplay between host diversity, pathogen attacks and abiotic stressors creates a complex system where diseases can range from benign background processes to catastrophic outbreaks [40]. Such complex interactions call for integrative approaches mixing climate modeling, high-throughput diagnostics and disease detection and prediction.

2.2. Scale and Accessibility Challenges

The vast spatial extent of forest ecosystems presents a major obstacle to effective disease surveillance. Unlike agricultural systems, where crops are cultivated in relatively uniform and accessible plots, forests span millions of hectares, often in remote, topographically challenging and low-access regions [30]. Monitoring these landscapes requires tremendous logistical and analytical effort. For example, ecological niche models, such as those developed for Phytophthora ramorum [41] or Dothistroma needle blight [42], can offer a powerful predictive framework if they are supported by high-resolution data and sufficient computational resources. Conventional forest health assessments still rely heavily on ground-based surveys and visual inspections, which are labor-intensive, inconsistent, and prone to observer bias [2]. Remote sensing technologies, particularly unmanned aerial vehicles (UAVs), now allow for broader coverage and finer temporal resolution. However, they also produce enormous volumes of heterogeneous data, creating new challenges in processing, interpretation, and integration into decision-making frameworks [43]. Beyond large-scale epidemics, less conspicuous issues such as wood decay also impose substantial financial burdens on the forestry sector [44]. Wood decay pathogens often become apparent only when fungal fructifications emerge, by which time the damage to timber quality and value is already severe which makes them difficult to detect [45]. Root diseases present a similar challenge, as their symptoms typically become visible only after extensive internal colonization [46]. Temporal factors can further complicate pathogen detection. Trees are long-lived organisms, and the progression of diseases can span years, making it difficult to assess their impact in real-time. Much like in animals, infected plants undergo an incubation period that delays visible symptoms, creating a diagnostic blind spot during which pathogens can spread undetected [47]. This delay can limit the window for management options, which can rapidly become limited or prohibitively expensive [48]. Furthermore, economic constraints frequently limit forest health monitoring programs, which are often underfunded compared to the agricultural sector [49,50].

2.3. Emerging and Amplifying Threats

The dynamics of forest diseases are shifting under the influence of global change. Climate change can facilitate the range expansion of both native and exotic pests and diseases (insects and pathogens) or affect tree resistance to pests by inducing physiological stress [17]. Hennon et al. (2020) [51] outlined three scenarios illustrating how biotic and abiotic factors interact to influence host–pathogen dynamics. In the first scenario, abiotic conditions create a favorable environment for pathogens to infect their host. For example, Dothistroma needle blight, a disease exacerbated by increased humidity, is expanding into regions where it was previously absent [11]. In the second, abiotic stress directly weakens the host, leading to disease either with (scenario 2a) or without (scenario 2b) the involvement of secondary agents [52]. Drought, pollution, and pest infestations can weaken tree defenses, interacting with diseases [50,53,54]. Moreover, these scenarios can overlap, particularly as climate change simultaneously affects both pathogens and hosts. In oak, combined effects of climatic extremes, insects, and pathogenic microorganisms cause a disease called acute oak decline, which can kill a mature tree in 4 to 5 years [55]. Global trade has also accelerated the introduction of invasive pathogens into new areas, with devastating consequences. Pathogens such as Cronartium ribicola, Cryphonectria parasitica, or Seiridium cardinale responsible for white pine blister rust, chestnut blight, and cypress canker disease, respectively, have been introduced in the late 19th and early 20th century into North America, where native trees lack evolved defenses [9,56,57]. Such exotic pathogen introductions often lead to explosive outbreaks with severe ecological and economic repercussions.

2.4. The High Stakes of Disease Mismanagement

The combined impact of invasive alien insect pests and pathogens and climate change can create enormous economic and environmental costs to our society, in the range of $4.3–20.2 trillion per year in ecosystem service losses [58]. Ecologically, large-scale outbreaks can reshape forest structure and function, with long-term implications for biodiversity and ecosystem services. For example, widespread mortality of keystone species like white pines can disrupt entire ecological networks, affecting wildlife and nutrient cycling [59]. Forest pathologists with multidisciplinary knowledge are needed to face the diverse threats laying on the forest and act as our first line of defense [60]. In response to the challenges previously outlined, scientists and stakeholders are increasingly turning to artificial intelligence to monitor, detect, diagnose, and even predict forest diseases. Tools such as remote sensing and image analysis play a key role in these efforts. In the following section, we will explore how these technologies are applied to both biotic and abiotic diseases, as well as decline, through a series of examples and practical applications.

3. Applications of AI in Forest Pathology

This section illustrates the applications of AI in forest pathology through selected examples, and the principal cases discussed are summarized in Table 1.
Table 1. Summary of main examples illustrating the application of AI in forest pathology, including approaches used for prediction, detection, and diagnostic of biotic, abiotic and decline diseases.

3.1. Biotic Diseases

Biotic diseases encompass a wide range of conditions, from minor leaf spots that cause esthetic damage to severe vascular diseases capable of killing a tree within a single season. As a central concern in forest pathology, these diseases are increasingly being monitored and analyzed using artificial intelligence. AI tools are now commonly employed not only for detection and surveillance but also for predicting the progression and severity of epidemics.
The integration of artificial intelligence with remote sensing is transforming forest health surveillance, offering a scalable means to detect outbreaks, quantify their spatial extent, and assess severity with unprecedented accuracy [83,84,85,86,87]. A growing body of studies illustrates how different imaging platforms, ranging from UAVs to airborne sensors and satellites, can be coupled with machine learning classification and computer vision to generate information for forest health monitoring. At local to regional scales, UAVs and multispectral sensors have achieved remarkable success in detecting foliar pathogens. For instance, aerial surveys of myrtle rust in Australia combined with machine learning models achieved prediction accuracies above 95%, enabling fine-scale mapping of infected stands [61]. Similarly, multispectral detection of poplar leaf rust in China yielded near-perfect classification of infected leaves [62]. These results demonstrate the power of coupling high-resolution imagery with AI, but they also highlight a potential limitation: such approaches can remain logistically constrained to narrow areas, when models are trained on specific or geographically limited datasets. Satellite-based detection offers broader coverage and greater potential for long-term monitoring, though often at the cost of resolution and accuracy. For instance, mapping of oak wilt disease can be achieved using variation in pigment composition caused by fungal infection and used as key targets for discriminating symptomatic oak trees from healthy or dead trees using phenological metrics [63]. Recent analyses of oak wilt in North America, using temporal and spatial satellite imagery, achieved 80%–82% accuracy in discriminating between healthy, stressed, and diseased trees [63]. While often lower than UAV-based accuracies, satellite models can enable the production of predictive maps that guide local stakeholders to target ground surveys and prioritize treatment. Such studies underscore the complementary roles of UAV and satellite platforms: one offers precision at local scales, the other operational reach for landscape-level management.
Beyond simple disease detection, hyperspectral data enable AI models not only to detect disease plants but also to classify symptoms and quantify disease severity rather than merely flagging its presence or absence. This approach has shown strong predictive capability in conifers; for instance, a recent study [64] combined inverted plant functional traits with narrow-band hyperspectral indices from UAV imagery and used random forest modeling to predict Dothistroma needle blight severity (from 0 to 58% of defoliation) with high accuracy (R2 = 0.85). The study found that variables related to key photosynthetic traits, such as chlorophyll content and degradation, were most effective in distinguishing asymptomatic trees from those with mild symptoms [64]. In another study, Bolikulov et al. (2024) [65] developed a method based on computer vision to detect poplar leaf disease using collected images. After applying a contrasting method, the algorithm was able to classify the leaf into five categories with a success of up to 95%: healthy, scab, brown spot, white-gray spot, and rust, reflecting common poplar disease of the sampled region [65]. These advancements will pave the way for real-time, field-level disease monitoring. However, one challenge of such studies is that they require a lot of data to train on, which can be unavailable or difficult to obtain depending on the area or for recently introduced pathogens. A recent study countered this pitfall by generating its own dataset to train their model for ash dieback detection at the leaf level, and obtained good precision when tested on real tree leaf footage compared to expert annotations [66].
Tree diseases vary widely in their visibility and detectability. Foliar pathogens that induce clear symptoms can be straightforward to identify, but inconspicuous diseases such as wood decay or vascular infections are far more difficult to capture remotely. Wood decay, for instance, may persist for years without detection before fruiting bodies emerge, and vascular pathogens are often only noticed once the disease is well advanced, at which point management becomes difficult or costly [63]. Broadband and narrowband reflectance-based spectrometry approaches have so far shown limited power in detecting such inconspicuous diseases. For example, Kankaanhuhta et al. (2000) [67] achieved detection rates of butt and root rot caused by Heterobasidion sp of 72%–90% for healthy trees and 94%–96% for infected spruce trees, while a more recent study reported an overall accuracy of only 65.5% in Norway spruce using hyperspectral data [68]. The difficulty arises because decay typically progresses in the heartwood rather than in living tissues, so affected trees show little outward stress, and reflectance differences across bands are insufficient for robust classification [68]. Some success has been obtained by focusing on indirect structural indicators, such as crown density and architecture, which can distinguish healthy trees from those affected by butt rot caused by Heterobasidion annosum [88]. Similarly, the timber industry has developed imaging technologies to assess decay and wood quality downstream [89], but adapting these tools for early, in situ detection of inconspicuous diseases in standing forests remains a challenge. By contrast, spectrometry based on chemical fingerprinting can offer a promising alternative. Pathogen attack alters plant metabolism, including the emission of secondary metabolites [90], which can be captured as spectral fingerprints. Fourier-transform infrared spectroscopy, for example, is a vibrational spectroscopy method that generates chemical fingerprints from solid, liquid, or gaseous samples. Using this approach, Mukrimin et al. (2019) [69] successfully classified trees into infected and non-infected categories after applying supervised machine learning to Fourier-transform infrared spectroscopy spectra. Although it was not coupled with AI, Fourier-transform infrared spectroscopy also shown great promise in detecting Dutch elm disease. Martín et al. (2005) [70] detected consistent chemical differences between resistant, susceptible, and control trees: resistant elms showed higher absorbance peaks than the other groups, suggesting increased lignin and suberin formation. These examples highlight the potential of chemical fingerprinting approaches to detect subtle physiological changes that remain invisible to broadband and narrowband reflectance analysis, their applicability to vast forested areas remains to be explored.
Beyond detection and surveillance, predicting pathogens outbreaks and distribution is a central question in forest pathology [71]. Species distribution models and ecological niche modeling have become commonly used to identify priority areas for monitoring and to determine which environmental factors most strongly shape pathogen spread. The integration of artificial intelligence in ecological niche modeling holds great promise in advancing the field because of AI’s ability to handle large datasets and extract complex patterns [91]. For example, such methods have been recently used to project suitable replacement species for sweet chestnut under future climate scenarios in southern Switzerland [72] and to model the potential distribution of the invasive pine pathogen Lecanosticta acicola across Europe, where ensemble models predicted range expansion under multiple climate projections [73]. Algorithms can also analyze environmental data to identify the most important environmental variables associated with pathogen adaptation, as shown in Dutch elm disease, Dothistroma needle blight and Swiss needle cast [74], or to uncover host responses, such as the genetic basis of resistance to ash dieback in Fraxinus excelsior [75]. Such predictive models allow not only the mapping of risk areas but also the anticipation of pathogen adaptation to changing environments. Another recent advance is the use of genomic offset, originally developed to assess climate adaptation in tree populations [92]. By quantifying the mismatch between current genomic variation and future climate conditions, genomic offset provides a means to predict pathogen maladaptation under climate change. This approach complements ecological niche modeling by integrating genomic markers directly, thereby offering finer resolution and more biologically grounded forecasts of where surveillance and intervention efforts should be focused [74].

3.2. Abiotic Diseases

Abiotic factors, particularly those related to climate, can be responsible for large scale forest mortality events, sometimes referred to as die-off [93]. These events can be responsible for loss of biodiversity, forest productivity, biodiversity and ecosystem services [94,95]. Climate-induced forest die-offs are becoming more frequent and severe under global change, with recent tree-killing droughts being longer and more intense, particularly in dry biomes and angiosperm-dominated forests [96]. Monitoring these extreme events at broad spatial scales is critical for assessing their ecological consequences and informing adaptive management strategies. Recent advances in tree physiology have improved our understanding of the mechanisms underlying drought-induced stress [97], while machine learning models allow for tailored monitoring and predicting of these events, as drought dieback can be site and species dependent [98,99]. For example, Yang et al. (2018) [76] used LiDAR imagery coupled with random forest modeling to assess the impact of the 2005 mega drought in the Amazon rainforest by capturing the reduction in forest canopy height. Although drought is often the primary driver [99], other abiotic stressors, such as excessive rainfall [100] or frost events [77] also contribute to dieback. Pollution, such as the one posed by heavy metals in the Sundarbans mangrove forest of Bangladesh and India, is another abiotic factor that can pose a threat to ecosystems [78]. Using extremely randomized tree models to model sediment pollution by heavy metals, Proshad et al. (2024) [78] were able to predict the accumulation of eleven heavy metals and proportion of each component with great precision, as a step in the determination of the potential sources of contamination.
Beyond monitoring forest health, a key challenge lies in identifying the specific environmental factors driving die-off events. A notable example is the work of Schiefer et al. (2024) [79], who used UAV-based remote sensing data at very fine scale through deep learning algorithms to identify the main drivers of forest dieback in Germany. Their approach enabled more accurate spatial estimates of affected areas and revealed a complex interplay of environmental factors, with hotter droughts and late frosts emerging as dominant predictors of mortality [79]. Such developments are promising for large-scale monitoring, as deep learning–based segmentation of tree crown dieback offers a cost-effective alternative to LiDAR, potentially expanding access to high-resolution forest health data [80]. The development of AI tools to predict occurrences of damaging abiotic factors is not limited to natural forests. For instance, Diniz et al. [81] used machine learning algorithms to create zoning classifications forecasting frost risk for forest plantations located in the south-central region of Rio Grande do Sul State, in Brazil. With increasing global climate extreme events, accurate risk prediction and zoning are essential for sustainable forest management and planning strategies.

3.3. Decline

Tree decline presents a particular challenge for diagnosis because it is inherently multifactorial and difficult to disentangle [101]. For decades, scientists and forest managers have been puzzled by cases in which trees appeared to die for unknown or complex reasons. Decline is best described as a syndrome rather than a single disease: long-term reductions in vigor caused by predisposing factors are often amplified by secondary stresses such as severe drought, insect outbreaks, or fungal infections, sometimes culminating in large-scale mortality events [101]. Importantly, there is no universal model of decline [102]. Each case reflects a unique combination of species, stressors, and ecological context. The involvement of multiple agents, for example, simultaneous insect and microbe attacks [55], makes decline especially suited for machine learning approaches, which can accommodate complex interactions across diverse datasets. Although AI offers exciting opportunities for studying decline, applications for diagnosis and prediction are only beginning to emerge, as illustrated by recent studies such as [82,83]. For instance, Beeson (2024) [82] used a variety of modeling techniques and bioclimatic covariates to refine the accuracy of predetermined risk maps of maple decline in the eastern United States. Although this approach is not technically machine learning, it improves map accuracy, which remains dependent on the resolution and quality of the underlying covariate layers. Phenotypic data can also be exploited, as shown in a study from Finch et al. (2021) [83], in which the authors applied unsupervised random forest models to 36 phenotypic traits of Quercus robur in England, generating quantitative indexes that distinguished acute from chronic oak decline. These indexes provide a standardized way to measure decline severity and enable comparisons across space and time [83]. Such examples illustrate how machine learning can move research beyond descriptive case studies toward robust, quantitative frameworks that support monitoring and management.

3.4. Insights from Crop Pathology and Phytoprotection

Advances in artificial intelligence, high-throughput sequencing, and remote sensing are transforming plant health monitoring in crops, and their adaptation to forest systems offers new opportunities for diagnostics, ecological insight, and management [30,32,64]. Some other tools and methods interesting to transfer to forest pathology include microbiome engineering, portative tools or diagnostic by sequencing.
High-throughput sequencing technologies coupled with AI holds transformative potential for forest health. Already widely adopted in crop systems for diagnosing fungi and oomycetes [103], these methods remain underutilized in forestry. Yet their applications extend beyond pathogen identification: sequencing can also profile the microbiome, offering insights into both diversity and functionality, and thus providing a broader measure of tree health [104,105]. This dual diagnostic and ecological perspective could be particularly valuable in long-lived hosts such as trees and to enhance the success of afforestation and reforestation practices [106].
AI further extends diagnostic and management possibilities by supporting the discovery of biological control agents. For example, Sadeghi et al. (2022) [107] applied machine learning to screen probiotic lactic acid bacteria with antimicrobial properties against the common agent of bacterial contamination in plant tissue cultures. Similar approaches helped predict the most effective strains of bacteria, individually or in synthetic communities, to control crop pathogens such as Verticillium dahliae or Fusarium oxysporum under specific environmental conditions [108,109]. By moving beyond empirical trial-and-error, AI-driven screening could optimize biocontrol strategies for forestry pathosystems, through the characterization of locally sourced and already adapted microbe communities.
Network analysis of plant–pathogen interactions can reveal the regulation of genes involved in pathogenicity and resistance, as well as protein–protein interactions and metabolic cascades in both host and pathogen [110]. For example, gene regulatory networks of the pathogenic fungus Fusarium graminearum have been constructed [111] from hundreds of transcriptomic datasets, providing insights into fungal virulence mechanisms [112]. Complementary approaches use machine learning to predict fungal effector proteins from secretomes and their subcellular localization, distinguishing effectors that remain in the apoplast from those that enter plant cells, thus facilitating functional studies [113,114]. For example, such tools for predicting how a specific tree or genotype would react to pathogen invasion could be very valuable to assure the establishment success of a plantation.
The integration of AI and machine learning with high-throughput phenotyping has opened new avenues for crop improvement, enabling the extraction of meaningful patterns between phenotypic traits and genetic factors and providing breeders with efficient tools for trait selection, reducing time and cost in variety development [111]. These approaches also empower genomic selection, as demonstrated in wheat breeding for rust resistance [115] and, more recently, in grapevine, where deep learning combined with phenomics, genomics, and transcriptomics enabled accurate identification of pest-resistant genes and prediction of pest resistance [116]. But trees and especially conifers have huge genomes [117], which makes genetic improvement slow. However, the process was already well aided by omics technologies [118], and AI could drive a similar acceleration [119].
Diagnostic tools originally developed for agriculture could be adapted for forestry, particularly portable devices suited to remote and difficult-to-access environments. Conventional pathogen detection relies on molecular technologies that, while accurate, are time-consuming and constrained to centralized laboratories. To address these limitations, Li et al. (2019) [120] developed a cost-effective, smartphone-based volatile organic compound fingerprinting platform capable of diagnosing late blight (Phytophthora infestans) non-invasively in the field. Looking forward, predictive frameworks that integrate genomic, environmental, and remote sensing data, as inspired by agricultural research, are emerging as powerful tools for anticipating forest pathogen emergence and spread [121], but validating such devices across diverse hosts, vast landscapes, and variable field conditions remains a formidable challenge in forests.
Without a coordinated effort from the scientific community toward open-access data initiatives to cross-disciplinary collaborations and shared validation platforms, forestry risks falling into a technological gap regarding AI, rich in proof-of-concept studies but poor in operational tools, as it has already been seen in other domains such as forest genomics [122]. Although tools and methods from crop science are attractive, we should resist the temptation to treat trees as agricultural plants. Their deep evolutionary divergence, larger and often more complex genomes, distinct physiologies, and long life histories demand tailored approaches rather than direct transfers from crop science. Unlike annual crops, forests cannot be rapidly replanted or replaced when disease strikes, making the maintenance of genetic diversity essential not only for resilience against disturbances but also restore its adaptive potential following environmental changes [123]. Any adaptation of AI tools to forestry must therefore preserve this long-term perspective, balancing technological innovation with the fundamental ecological and evolutionary realities of trees.

3.5. Future Directions and Pitfalls of AI in Forest Pathology

The effectiveness of AI ultimately depends on the quality of the data used to train it, an idea captured by the well-known expression “garbage in, garbage out”. Reinforcement learning approaches are gaining interest in forestry [124,125], and could help support managers with data-informed decision-making on forest diseases. However, many forest datasets lack the robustness, biological context and standardization required for reliable modeling. Ground-truth data from laboratory and field studies remain indispensable to calibrate and validate remotely sensed and publicly available datasets, ensuring biological relevance in model outputs [126]. Two key challenges have been identified: cross-ecological generalization, as models are often trained on narrow datasets and fail to transfer across forest types, and multisource data fusion, where mismatches in temporal and spatial scales limit accuracy [127]. Data privacy and ownership also remain unresolved, as economic and institutional interests frequently restrict data sharing, yet progress in forest AI will require coordinated, open, and large-scale databases [127].
Bias in species representation represents another pitfall. AI studies tend to prioritize economically valuable species (e.g., spruce [71,72,76], or poplar [66,69]), neglecting secondary species and mixed communities that are critical for biodiversity and resilience. Similarly, most models target single species or single diseases, whereas future approaches must integrate ecological complexity for a better understanding of forest health as a whole [64]. Models must also account for the diversity of environments in which trees grow, from urban settings to old-growth forests, which strongly shape physiology and susceptibility to infection [128].
Technical barriers also constrain progress. Remote sensing has driven many advances, yet canopy cover still obscures critical understory processes [26], and co-infections or novel pathogen variants remain difficult to detect [129]. These limitations underscore the need for new sensing approaches, better disease diagnostics, and dynamic models that can adapt to shifting pathogen life cycles.
Traditionally, forest pathologists could focus on a single disease system, as the slow development of new pathogens allowed for thorough research over many years. However, the rapid increase in new or re-emerging climate-related disease outbreaks and the introduction of alien species now necessitate a more interdisciplinary approach [60]. These issues can be tackled with the availability of advanced surveillance technologies, such as remote sensing or machine learning, however they demand new skills in technology deployment, data analysis, and computational techniques. Consequently, contemporary forest pathologists need not only traditional diagnostic skills but also the ability to integrate various data sources and work across different fields to effectively monitor and address emerging threats.
Public perceptions add another layer of complexity. While AI may be positively viewed in research contexts, forest owners and other stakeholders can express concerns regarding its maturity, potential misuse, privacy, and security [130]. The environmental cost of running AI will also be an important point to consider, as its use in research grows, just like the carbon footprint of other bioinformatic tools, good practices will be needed to reduce power consumption and carbon emission [131]. Addressing these concerns will require innovative partnerships, cost-effective tools, and sustained investments in data infrastructure, as well as careful balancing of the high costs of AI deployment against the limited funding typically available for forest health initiatives.

4. Conclusions

The application of AI in forest pathology holds great promise for detecting, surveilling, and predicting the complex interplay of pathogens and abiotic stressors that threaten forest health. By integrating remote sensing, genomic, and environmental data, AI can transform monitoring from reactive to predictive, guiding interventions with unprecedented precision. Yet the power of these models ultimately depends on the quality of the data that train them. The challenge, then, is not simply to transfer tools from agriculture into forestry but to innovate diagnostic and predictive systems tailored to long-lived species and complex ecosystems. This requires confronting whether forestry can realistically sustain the infrastructure, sustained investment, and rigorous validation that AI demands. As climate change accelerates host–pathogen interactions and intensifies forest decline, the key question is not simply whether AI can contribute to forest pathology but how the field can best harness its potential.

Funding

This research received no external funding.

Data Availability Statement

Data sources are contained within the article.

Acknowledgments

We warmly thank Louis Bernier (Université Laval, Département des sciences du bois et de la forêt, Faculté de foresterie, géographie et géomatique) for his thoughtful comments and careful proofreading of the manuscript.

Conflicts of Interest

The author declares no conflicts of interest.

References

  1. Hartig, R. Text-Book of the Diseases of Trees; Macmillan and Company: New York, NY, USA, 1894; ISBN 9780598942012. [Google Scholar]
  2. Asiegbu, F.O. Basic Concepts and Principles of Forest Pathology. In Forest Microbiology; Elsevier: Amsterdam, The Netherlands, 2022; pp. 3–15. ISBN 9780323850421. [Google Scholar]
  3. Tainter, F.H.; Baker, F.A. Principles of Forest Pathology; John Wiley & Sons: Nashville, TN, USA, 1996; ISBN 9780471129523. [Google Scholar]
  4. Ostry, M.E.; Laflamme, G. Fungi and Diseases. Botany 2009, 87, 22–25. [Google Scholar] [CrossRef]
  5. Anagnostakis, S.L. Chestnut Blight: The Classical Problem of an Introduced Pathogen. Mycologia 1987, 79, 23–37. [Google Scholar] [CrossRef]
  6. Vannini, A.; Maria, A.; Vettraino, A.M. Ink disease in chestnuts: Impact on the European chestnut. For. Snow Landsc. Res. 2001, 76, 345–350. [Google Scholar]
  7. Bernier, L. Dutch Elm Disease. In Forest Microbiology; Elsevier: Amsterdam, The Netherlands, 2022; pp. 291–309. ISBN 9780323850421. [Google Scholar]
  8. Grünwald, N.J.; LeBoldus, J.M.; Hamelin, R.C. Ecology and Evolution of the Sudden Oak Death Pathogen Phytophthora Ramorum. Annu. Rev. Phytopathol. 2019, 57, 301–321. [Google Scholar] [CrossRef] [PubMed]
  9. Kinloch, B.B. White Pine Blister Rust in North America: Past and Prognosis. Phytopathology 2003, 93, 1044–1047. [Google Scholar] [CrossRef]
  10. Scali, E.; Garbelotto, M.; Danti, R.; Della Rocca, G. The Cypress Canker Disease Pandemic. Plant Health Cases 2025, 2025, phcs20250001. [Google Scholar] [CrossRef]
  11. Welsh, C.; Lewis, K.; Woods, A. The Outbreak History of Dothistroma Needle Blight: An Emerging Forest Disease in Northwestern British Columbia, Canada. Can. J. For. Res. 2009, 39, 2505–2519. [Google Scholar] [CrossRef]
  12. Wagner, A.C.; Tomback, D.F.; Resler, L.M.; Pansing, E.R. Whitebark Pine Prevalence and Ecological Function in Treeline Communities of the Greater Yellowstone Ecosystem, U.S.A.: Potential Disruption by White Pine Blister Rust. Forests 2018, 9, 635. [Google Scholar] [CrossRef]
  13. Petucco, C.; Lobianco, A.; Caurla, S. Economic Evaluation of an Invasive Forest Pathogen at a Large Scale: The Case of Ash Dieback in France. Environ. Model. Assess. 2020, 25, 1–21. [Google Scholar] [CrossRef]
  14. Nnadi, N.E.; Carter, D.A. Climate Change and the Emergence of Fungal Pathogens. PLoS Pathog. 2021, 17, e1009503. [Google Scholar] [CrossRef]
  15. Singh, B.K.; Delgado-Baquerizo, M.; Egidi, E.; Guirado, E.; Leach, J.E.; Liu, H.; Trivedi, P. Climate Change Impacts on Plant Pathogens, Food Security and Paths Forward. Nat. Rev. Microbiol. 2023, 21, 640–656. [Google Scholar] [CrossRef] [PubMed]
  16. Alfaro, R.I.; Fady, B.; Vendramin, G.G.; Dawson, I.K.; Fleming, R.A.; Sáenz-Romero, C.; Lindig-Cisneros, R.A.; Murdock, T.; Vinceti, B.; Navarro, C.M.; et al. The Role of Forest Genetic Resources in Responding to Biotic and Abiotic Factors in the Context of Anthropogenic Climate Change. For. Ecol. Manag. 2014, 333, 76–87. [Google Scholar] [CrossRef]
  17. Sturrock, R.N.; Frankel, S.J.; Brown, A.V.; Hennon, P.E.; Kliejunas, J.T.; Lewis, K.J.; Worrall, J.J.; Woods, A.J. Climate Change and Forest Diseases: Climate Change and Forest Diseases. Plant Pathol. 2011, 60, 133–149. [Google Scholar] [CrossRef]
  18. Byrd, A.L.; Segre, J.A. Infectious Disease. Adapting Koch’s Postulates. Science 2016, 351, 224–226. [Google Scholar] [CrossRef]
  19. Hamelin, R.C.; Roe, A.D. Genomic Biosurveillance of Forest Invasive Alien Enemies: A Story Written in Code. Evol. Appl. 2020, 13, 95–115. [Google Scholar] [CrossRef]
  20. Hubbes, M. Impact of Molecular Biology on Forest Pathology: A Literature Review. Eur. J. For. Pathol. 1993, 23, 201–217. [Google Scholar] [CrossRef]
  21. Xu, Y.; Liu, X.; Cao, X.; Huang, C.; Liu, E.; Qian, S.; Liu, X.; Wu, Y.; Dong, F.; Qiu, C.-W.; et al. Artificial Intelligence: A Powerful Paradigm for Scientific Research. Innovation 2021, 2, 100179. [Google Scholar] [CrossRef]
  22. Ezzati, S.; Zenner, E.K.; Pakdaman, M.; Naseri, M.H.; Nikjoui, M.; Ahmadi, S. Spatially Explicit Modeling of Disease Surveillance in Mixed Oak-Hardwood Forests Based on Machine-Learning Algorithms. J. Environ. Manag. 2023, 337, 117714. [Google Scholar] [CrossRef] [PubMed]
  23. Roy, A.M.; Bhaduri, J. A Deep Learning Enabled Multi-Class Plant Disease Detection Model Based on Computer Vision. AI 2021, 2, 413–428. [Google Scholar] [CrossRef]
  24. Hue, Y.; Kim, J.H.; Lee, G.; Choi, B.; Sim, H.; Jeon, J.; Ahn, M.-I.; Han, Y.K.; Kim, K.-T. Artificial Intelligence Plant Doctor: Plant Disease Diagnosis Using GPT4-Vision. Sigmulbyeong Yeongu 2024, 30, 99–102. [Google Scholar] [CrossRef]
  25. Libbrecht, M.W.; Noble, W.S. Machine Learning Applications in Genetics and Genomics. Nat. Rev. Genet. 2015, 16, 321–332. [Google Scholar] [CrossRef]
  26. Buchelt, A.; Adrowitzer, A.; Kieseberg, P.; Gollob, C.; Nothdurft, A.; Eresheim, S.; Tschiatschek, S.; Stampfer, K.; Holzinger, A. Exploring Artificial Intelligence for Applications of Drones in Forest Ecology and Management. For. Ecol. Manag. 2024, 551, 121530. [Google Scholar] [CrossRef]
  27. Jafar, A.; Bibi, N.; Naqvi, R.A.; Sadeghi-Niaraki, A.; Jeong, D. Revolutionizing Agriculture with Artificial Intelligence: Plant Disease Detection Methods, Applications, and Their Limitations. Front. Plant Sci. 2024, 15, 1356260. [Google Scholar] [CrossRef]
  28. Orchi, H.; Sadik, M.; Khaldoun, M. On Using Artificial Intelligence and the Internet of Things for Crop Disease Detection: A Contemporary Survey. Agriculture 2021, 12, 9. [Google Scholar] [CrossRef]
  29. Fenu, G.; Malloci, F.M. Artificial Intelligence Technique in Crop Disease Forecasting: A Case Study on Potato Late Blight Prediction. In Intelligent Decision Technologies; IOS Press: Amsterdam, The Netherlands, 2020; pp. 79–89. ISBN 9789811559242. [Google Scholar]
  30. Hessenauer, P.; Feau, N.; Gill, U.; Schwessinger, B.; Brar, G.S.; Hamelin, R.C. Evolution and Adaptation of Forest and Crop Pathogens in the Anthropocene. Phytopathology 2021, 111, 49–67. [Google Scholar] [CrossRef] [PubMed]
  31. Pautasso, M.; Schlegel, M.; Holdenrieder, O. Forest Health in a Changing World. Microb. Ecol. 2015, 69, 826–842. [Google Scholar] [CrossRef] [PubMed]
  32. Ramsfield, T.D.; Bentz, B.J.; Faccoli, M.; Jactel, H.; Brockerhoff, E.G. Forest Health in a Changing World: Effects of Globalization and Climate Change on Forest Insect and Pathogen Impacts. Forestry 2016, 89, 245–252. [Google Scholar] [CrossRef]
  33. Prospero, S.; Botella, L.; Santini, A.; Robin, C. Biological Control of Emerging Forest Diseases: How Can We Move from Dreams to Reality? For. Ecol. Manag. 2021, 496, 119377. [Google Scholar] [CrossRef]
  34. Sniezko, R.A.; Liu, J.-J. Genetic Resistance to White Pine Blister Rust, Restoration Options, and Potential Use of Biotechnology. For. Ecol. Manag. 2022, 520, 120168. [Google Scholar] [CrossRef]
  35. Haas, S.E.; Hooten, M.B.; Rizzo, D.M.; Meentemeyer, R.K. Forest Species Diversity Reduces Disease Risk in a Generalist Plant Pathogen Invasion: Species Diversity Reduces Disease Risk. Ecol. Lett. 2011, 14, 1108–1116. [Google Scholar] [CrossRef]
  36. Jactel, H.; Bauhus, J.; Boberg, J.; Bonal, D.; Castagneyrol, B.; Gardiner, B.; Gonzalez-Olabarria, J.R.; Koricheva, J.; Meurisse, N.; Brockerhoff, E.G. Tree Diversity Drives Forest Stand Resistance to Natural Disturbances. Curr. For. Rep. 2017, 3, 223–243. [Google Scholar] [CrossRef]
  37. Deuffic, P.; Garms, M.; He, J.; Brahic, E.; Yang, H.; Mayer, M. Forest Dieback, a Tangible Proof of Climate Change? A Cross-Comparison of Forest Stakeholders’ Perceptions and Strategies in the Mountain Forests of Europe and China. Environ. Manag. 2020, 66, 858–872. [Google Scholar] [CrossRef]
  38. Kowsari, M.; Karimi, E. A Review on Oak Decline: The Global Situation, Causative Factors, and New Research Approaches. For. Syst. 2023, 32, eR01. [Google Scholar] [CrossRef]
  39. Holdenrieder, O.; Pautasso, M.; Weisberg, P.J.; Lonsdale, D. Tree Diseases and Landscape Processes: The Challenge of Landscape Pathology. Trends Ecol. Evol. 2004, 19, 446–452. [Google Scholar] [CrossRef]
  40. Hartmann, H.; Bastos, A.; Das, A.J.; Esquivel-Muelbert, A.; Hammond, W.M.; Martínez-Vilalta, J.; McDowell, N.G.; Powers, J.S.; Pugh, T.A.M.; Ruthrof, K.X.; et al. Climate Change Risks to Global Forest Health: Emergence of Unexpected Events of Elevated Tree Mortality Worldwide. Annu. Rev. Plant Biol. 2022, 73, 673–702. [Google Scholar] [CrossRef] [PubMed]
  41. Meentemeyer, R.K.; Anacker, B.L.; Mark, W.; Rizzo, D.M. Early Detection of Emerging Forest Disease Using Dispersal Estimation and Ecological Niche Modeling. Ecol. Appl. 2008, 18, 377–390. [Google Scholar] [CrossRef] [PubMed]
  42. Herpin-Saunier, N.Y.H.; Sambaraju, K.R.; Yin, X.; Feau, N.; Zeglen, S.; Ritokova, G.; Omdal, D.; Côté, C.; Hamelin, R.C. Genetic Lineage Distribution Modeling to Predict Epidemics of a Conifer Disease. Front. For. Glob. Change 2022, 4, 756678. [Google Scholar] [CrossRef]
  43. Ecke, S.; Dempewolf, J.; Frey, J.; Schwaller, A.; Endres, E.; Klemmt, H.-J.; Tiede, D.; Seifert, T. UAV-Based Forest Health Monitoring: A Systematic Review. Remote Sens. 2022, 14, 3205. [Google Scholar] [CrossRef]
  44. Ador, M.A.H.; Ahmed, R.; Dey, B.; Haque, M.M.U. Assessment of Decay Caused by Trametes flavida and Daldinia eschscholtzii on Several Hardwood Species. For. Pathol. 2024, 54, e12870. [Google Scholar] [CrossRef]
  45. Gonthier, P.; Guglielmo, F.; Sillo, F.; Giordano, L.; Garbelotto, M. A Molecular Diagnostic Assay for the Detection and Identification of Wood Decay Fungi of Conifers. For. Pathol. 2015, 45, 89–101. [Google Scholar] [CrossRef]
  46. Liu, T.-Y.; Chen, C.-H.; Ko, Y.-C.; Wu, Z.-C.; Liao, T.-Z.; Lee, H.-H.; Tsai, I.J.; Chang, T.-T.; Wu, M.-L.; Tsai, J.-N.; et al. Development and Evaluation of Real-Time Quantitative PCR Assays for Detection of Phellinus Noxius Causing Brown Root Rot Disease. Plant Dis. 2024, 108, 3288–3299. [Google Scholar] [CrossRef]
  47. Leclerc, M.; Doré, T.; Gilligan, C.A.; Lucas, P.; Filipe, J.A.N. Estimating the Delay between Host Infection and Disease (incubation Period) and Assessing Its Significance to the Epidemiology of Plant Diseases. PLoS ONE 2014, 9, e86568. [Google Scholar] [CrossRef] [PubMed]
  48. Ahmed, D.A.; Hudgins, E.J.; Cuthbert, R.N.; Kourantidou, M.; Diagne, C.; Haubrock, P.J.; Leung, B.; Liu, C.; Leroy, B.; Petrovskii, S.; et al. Managing Biological Invasions: The Cost of Inaction. Biol. Invasions 2022, 24, 1927–1946. [Google Scholar] [CrossRef]
  49. Stone, C.; Old, K.; Kile, G.; Coopst, N. Forest Health Monitoring in Australia: National and Regional Commitments and Operational Realities. Ecosyst. Health 2001, 7, 48–57. [Google Scholar] [CrossRef]
  50. McGinley, K.A.; Guldin, R.W.; Cubbage, F.W. Forest Sector Research and Development Capacity. J. For. 2019, 117, 443–461. [Google Scholar] [CrossRef]
  51. Hennon, P.E.; Frankel, S.J.; Woods, A.J.; Worrall, J.J.; Norlander, D.; Zambino, P.J.; Warwell, M.V.; Shaw, C.G., III. A Framework to Evaluate Climate Effects on Forest Tree Diseases. For. Pathol. 2020, 50, e12649. [Google Scholar] [CrossRef]
  52. Caldeira, M.C. The Timing of Drought Coupled with Pathogens May Boost Tree Mortality. Tree Physiol. 2019, 39, 1–5. [Google Scholar] [CrossRef]
  53. Hossain, M.; Veneklaas, E.J.; Hardy, G.E.S.J.; Poot, P. Tree Host-Pathogen Interactions as Influenced by Drought Timing: Linking Physiological Performance, Biochemical Defence and Disease Severity. Tree Physiol. 2019, 39, 6–18. [Google Scholar] [CrossRef] [PubMed]
  54. Canelles, Q.; Aquilué, N.; James, P.M.A.; Lawler, J.; Brotons, L. Global Review on Interactions between Insect Pests and Other Forest Disturbances. Landsc. Ecol. 2021, 36, 945–972. [Google Scholar] [CrossRef]
  55. Brady, C.; Arnold, D.; McDonald, J.; Denman, S. Taxonomy and Identification of Bacteria Associated with Acute Oak Decline. World J. Microbiol. Biotechnol. 2017, 33, 143. [Google Scholar] [CrossRef] [PubMed]
  56. Kinloch, B.B., Jr.; Westfall, R.D.; White, E.E.; Gitzendanner, M.A.; Dupper, G.E.; Foord, B.M.; Hodgskiss, P.D. Genetics of Cronartium Ribicola. IV. Population Structure in Western North America. Can. J. Bot. 1998, 76, 91–98. [Google Scholar]
  57. Danti, R.; Della Rocca, G. Epidemiological History of Cypress Canker Disease in Source and Invasion Sites. Forests 2017, 8, 121. [Google Scholar] [CrossRef]
  58. Costanza, R.; de Groot, R.; Sutton, P.; van der Ploeg, S.; Anderson, S.J.; Kubiszewski, I.; Farber, S.; Turner, R.K. Changes in the Global Value of Ecosystem Services. Glob. Environ. Change 2014, 26, 152–158. [Google Scholar] [CrossRef]
  59. Jenkins, M.B.; Schoettle, A.W.; Wright, J.W.; Anderson, K.A.; Fortier, J.; Hoang, L.; Incashola, T., Jr.; Keane, R.E.; Krakowski, J.; LaFleur, D.M.; et al. Restoring a Forest Keystone Species: A Plan for the Restoration of Whitebark Pine (Pinus albicaulis Engelm.) in the Crown of the Continent Ecosystem. For. Ecol. Manag. 2022, 522, 120282. [Google Scholar] [CrossRef]
  60. Hadziabdic, D.; Bonello, P.; Hamelin, R.; Juzwik, J.; Moltzan, B.; Rizzo, D.; Stewart, J.; Villari, C. The Future of Forest Pathology in North America. Front. For. Glob. Change 2021, 4, 737445. [Google Scholar] [CrossRef]
  61. Sandino, J.; Pegg, G.; Gonzalez, F.; Smith, G. Aerial Mapping of Forests Affected by Pathogens Using UAVs, Hyperspectral Sensors, and Artificial Intelligence. Sensors 2018, 18, 944. [Google Scholar] [CrossRef]
  62. Jia, Z.; Wang, Y.; Tian, Y.; Xie, Z.; Wang, C. Detection of Poplar Leaf Rust Using UAV Multispectral Imaging. Res. Sq. 2024. [Google Scholar]
  63. Guzmán-Quesada, J.A.; Pinto-Ledezma, J.N.; Frantz, D.; Townsend, P.A.; Juzwik, J.; Cavender-Bares, J. Mapping Oak Wilt Disease from Space Using Land Surface Phenology. Remote Sens. Environ. 2023, 298, 113794. [Google Scholar] [CrossRef]
  64. Watt, M.S.; Poblete, T.; de Silva, D.; Estarija, H.J.C.; Hartley, R.J.L.; Leonardo, E.M.C.; Massam, P.; Buddenbaum, H.; Zarco-Tejada, P.J. Prediction of the Severity of Dothistroma Needle Blight in Radiata Pine Using Plant Based Traits and Narrow Band Indices Derived from UAV Hyperspectral Imagery. Agric. For. Meteorol. 2023, 330, 109294. [Google Scholar] [CrossRef]
  65. Bolikulov, F.; Abdusalomov, A.; Nasimov, R.; Akhmedov, F.; Cho, Y.-I. Early Poplar (Populus) Leaf-Based Disease Detection through Computer Vision, YOLOv8, and Contrast Stretching Technique. Sensors 2024, 24, 5200. [Google Scholar] [CrossRef]
  66. Bates, E.; Popović, M.; Marsh, C.; Clark, R.; Kovac, M.; Kocer, B.B. Leaf Level Ash Dieback Disease Detection and Online Severity Estimation with UAVs. IEEE Access 2025, 13, 55499–55511. [Google Scholar] [CrossRef]
  67. Kankaanhuhta, V.; Mäkisara, K.; Tomppo, E.; Piri, T.; Kaitera, J. Monitoring of Diseases Caused by Heterobasidion Annosum and Peridermiun Pini in Norway Spruce and Scots Pine Stands by Airborne Imaging Spectrometry (AISA). 2000. Available online: https://jukuri.luke.fi/items/8eef6d14-a0f1-4977-84df-9f92e25f5aed (accessed on 24 September 2025).
  68. Dalponte, M.; Kallio, A.J.I.; Ørka, H.O.; Næsset, E.; Gobakken, T. Wood Decay Detection in Norway Spruce Forests Based on Airborne Hyperspectral and ALS Data. Remote Sens. 2022, 14, 1892. [Google Scholar] [CrossRef]
  69. Mukrimin, M.; Conrad, A.O.; Kovalchuk, A.; Julkunen-Tiitto, R.; Bonello, P.; Asiegbu, F.O. Fourier-Transform Infrared (FT-IR) Spectroscopy Analysis Discriminates Asymptomatic and Symptomatic Norway Spruce Trees. Plant Sci. 2019, 289, 110247. [Google Scholar] [CrossRef] [PubMed]
  70. Martín, J.A.; Solla, A.; Woodward, S.; Gil, L. Fourier Transform-Infrared Spectroscopy as a New Method for Evaluating Host Resistance in the Dutch Elm Disease Complex. Tree Physiol. 2005, 25, 1331–1338. [Google Scholar] [CrossRef] [PubMed]
  71. Václavík, T.; Kanaskie, A.; Hansen, E.M.; Ohmann, J.L.; Meentemeyer, R.K. Predicting Potential and Actual Distribution of Sudden Oak Death in Oregon: Prioritizing Landscape Contexts for Early Detection and Eradication of Disease Outbreaks. For. Ecol. Manag. 2010, 260, 1026–1035. [Google Scholar] [CrossRef]
  72. Heinz, M.; Prospero, S. A Modeling Approach to Determine Substitutive Tree Species for Sweet Chestnut in Stands Affected by Ink Disease. J. For. Res. 2025, 36, 24. [Google Scholar] [CrossRef]
  73. Ogris, N.; Drenkhan, R.; Vahalík, P.; Cech, T.; Mullett, M.; Tubby, K. The Potential Global Distribution of an Emerging Forest Pathogen, Lecanosticta Acicola, under a Changing Climate. Front. For. Glob. Change 2023, 6, 1221339. [Google Scholar] [CrossRef]
  74. Hessenauer, P.; Feau, N.; Heinzelmann, R.; Hamelin, R.C. Genomic Exploration of Climate-Driven Evolution and Evolutionary Convergence in Forest Pathogens. Genome Biol. Evol. 2025, 17, evaf069. [Google Scholar] [CrossRef]
  75. Doonan, J.M.; Budde, K.B.; Kosawang, C.; Lobo, A.; Verbylaite, R.; Brealey, J.C.; Martin, M.D.; Pliura, A.; Thomas, K.; Konrad, H.; et al. Multiple, Single Trait GWAS and Supervised Machine Learning Reveal the Genetic Architecture of Fraxinus Excelsior Tolerance to Ash Dieback in Europe. Plant Cell Environ. 2025, 48, 3793–3809. [Google Scholar] [CrossRef]
  76. Yang, Y.; Saatchi, S.S.; Xu, L.; Yu, Y.; Choi, S.; Phillips, N.; Kennedy, R.; Keller, M.; Knyazikhin, Y.; Myneni, R.B. Post-Drought Decline of the Amazon Carbon Sink. Nat. Commun. 2018, 9, 3172. [Google Scholar] [CrossRef]
  77. Olano, J.M.; García-Cervigón, A.I.; Sangüesa-Barreda, G.; Rozas, V.; Muñoz-Garachana, D.; García-Hidalgo, M.; García-Pedrero, Á. Satellite Data and Machine Learning Reveal the Incidence of Late Frost Defoliations on Iberian Beech Forests. Ecol. Appl. 2021, 31, e02288. [Google Scholar] [CrossRef] [PubMed]
  78. Proshad, R.; Rahim, M.A.; Rahman, M.; Asif, M.R.; Dey, H.C.; Khurram, D.; Al, M.A.; Islam, M.; Idris, A.M. Utilizing Machine Learning to Evaluate Heavy Metal Pollution in the World’s Largest Mangrove Forest. Sci. Total Environ. 2024, 951, 175746. [Google Scholar] [CrossRef]
  79. Schiefer, F.; Schmidtlein, S.; Hartmann, H.; Schnabel, F.; Kattenborn, T. Large-Scale Remote Sensing Reveals That Tree Mortality in Germany Appears to Be Greater than Previously Expected. Forestry 2024, 98, 535–549. [Google Scholar] [CrossRef]
  80. Allen, M.J.; Moreno-Fernández, D.; Ruiz-Benito, P.; Grieve, S.W.D.; Lines, E.R. Low-Cost Tree Crown Dieback Estimation Using Deep Learning-Based Segmentation. Environ. Data Sci. 2024, 3, e18. [Google Scholar] [CrossRef]
  81. Diniz, É.S.; Lorenzon, A.S.; de Castro, N.L.M.; Marcatti, G.E.; dos Santos, O.P.; de Deus Júnior, J.C.; Cavalcante, R.B.L.; Fernandes-Filho, E.I.; Hummeldo Amaral, C. Forecasting Frost Risk in Forest Plantations by the Combination of Spatial Data and Machine Learning Algorithms. Agric. For. Meteorol. 2021, 306, 108450. [Google Scholar] [CrossRef]
  82. Beeson, J.R. Validation and Improvement of Forest Health Risk Maps: A Case Study with Maple Decline/Dieback. Master’s Thesis, Michigan Technological University, Houghton, MI, USA, 2024; p. 31641068. [Google Scholar] [CrossRef]
  83. Finch, J.P.; Brown, N.; Beckmann, M.; Denman, S.; Draper, J. Index Measures for Oak Decline Severity Using Phenotypic Descriptors. For. Ecol. Manage. 2021, 485, 118948. [Google Scholar] [CrossRef]
  84. Hall, R.J.; Castilla, G.; White, J.C.; Cooke, B.J.; Skakun, R.S. Remote Sensing of Forest Pest Damage: A Review and Lessons Learned from a Canadian Perspective. Can. Entomol. 2016, 148, S296–S356. [Google Scholar] [CrossRef]
  85. Barmpoutis, P.; Papaioannou, P.; Dimitropoulos, K.; Grammalidis, N. A Review on Early Forest Fire Detection Systems Using Optical Remote Sensing. Sensors 2020, 20, 6442. [Google Scholar] [CrossRef]
  86. Torres, P.; Rodes-Blanco, M.; Viana-Soto, A.; Nieto, H.; García, M. The Role of Remote Sensing for the Assessment and Monitoring of Forest Health: A Systematic Evidence Synthesis. Forests 2021, 12, 1134. [Google Scholar] [CrossRef]
  87. Drechsel, J.; Forkel, M. Remote Sensing Forest Health Assessment—A Comprehensive Literature Review on a European Level. Cent. Eur. For. J. 2025, 71, 14–39. [Google Scholar] [CrossRef]
  88. Žemaitis, P.; Žemaitė, I. Does Butt Rot Affect the Crown Condition of Norway Spruce Trees? Trees 2018, 32, 489–495. [Google Scholar] [CrossRef]
  89. Vijayalakshmi, S.; Mrudhula, S.; Ashok Kumar, V.; Agastin; Varun; Latha, A.M. Artificial Intelligence-Driven Timber Wood Defect Characterization from Terahertz Images. J. Nondestruct. Eval. 2024, 43, 116. [Google Scholar] [CrossRef]
  90. Witzell, J.; Martín, J.A. Phenolic Metabolites in the Resistance of Northern Forest Trees to Pathogens—Past Experiences and Future Prospects. Can. J. For. Res. 2008, 38, 2711–2727. [Google Scholar] [CrossRef]
  91. Thuiller, W. Ecological Niche Modelling. Curr. Biol. 2024, 34, R225–R229. [Google Scholar] [CrossRef]
  92. Rellstab, C.; Dauphin, B.; Exposito-Alonso, M. Prospects and Limitations of Genomic Offset in Conservation Management. Evol. Appl. 2021, 14, 1202–1212. [Google Scholar] [CrossRef] [PubMed]
  93. Hammond, W.M.; Williams, A.P.; Abatzoglou, J.T.; Adams, H.D.; Klein, T.; López, R.; Sáenz-Romero, C.; Hartmann, H.; Breshears, D.D.; Allen, C.D. Global Field Observations of Tree Die-off Reveal Hotter-Drought Fingerprint for Earth’s Forests. Nat. Commun. 2022, 13, 1761. [Google Scholar] [CrossRef]
  94. Stursová, M.; Snajdr, J.; Cajthaml, T.; Bárta, J.; Santrůčková, H.; Baldrian, P. When the Forest Dies: The Response of Forest Soil Fungi to a Bark Beetle-Induced Tree Dieback. ISME J. 2014, 8, 1920–1931. [Google Scholar] [CrossRef]
  95. Batllori, E.; Lloret, F.; Aakala, T.; Anderegg, W.R.L.; Aynekulu, E.; Bendixsen, D.P.; Bentouati, A.; Bigler, C.; Burk, C.J.; Camarero, J.J.; et al. Forest and Woodland Replacement Patterns Following Drought-Related Mortality. Proc. Natl. Acad. Sci. USA. 2020, 117, 29720–29729. [Google Scholar] [CrossRef]
  96. Gazol, A.; Pizarro, M.; Hammond, W.M.; Allen, C.D.; Camarero, J.J. Droughts Preceding Tree Mortality Events Have Increased in Duration and Intensity, Especially in Dry Biomes. Nat. Commun. 2025, 16, 5779. [Google Scholar] [CrossRef]
  97. Colangelo, M.; Camarero, J.J.; Battipaglia, G.; Borghetti, M.; De Micco, V.; Gentilesca, T.; Ripullone, F. A Multi-Proxy Assessment of Dieback Causes in a Mediterranean Oak Species. Tree Physiol. 2017, 37, 617–631. [Google Scholar] [CrossRef]
  98. Lv, H.; Gangwisch, M.; Saha, S. Crown Die-Back of Peri-Urban Forests after Combined Heatwave and Drought Was Species-Specific, Size-Dependent, and Also Related to Tree Neighbourhood Characteristics. Sci. Total Environ. 2024, 913, 169716. [Google Scholar] [CrossRef]
  99. Camarero, J.J.; Gazol, A.; Sangüesa-Barreda, G.; Oliva, J.; Vicente-Serrano, S.M. To Die or Not to Die: Early Warnings of Tree Dieback in Response to a Severe Drought. J. Ecol. 2015, 103, 44–57. [Google Scholar] [CrossRef]
  100. Rozas, V.; García-González, I. Too Wet for Oaks? Inter-Tree Competition and Recent Persistent Wetness Predispose Oaks to Rainfall-Induced Dieback in Atlantic Rainy Forest. Glob. Planet. Change 2012, 94–95, 62–71. [Google Scholar] [CrossRef]
  101. Manion, P. Tree Disease Concepts; Manion, P.D., Ed.; Prentice-Hall: Englewood Cliffs, NJ, USA, 1981. [Google Scholar]
  102. Houston, D.R. Forest Tree Declines of Past and Present: Current Understanding. Can. J. Plant Pathol. 1987, 9, 349–360. [Google Scholar] [CrossRef]
  103. Aragona, M.; Haegi, A.; Valente, M.T.; Riccioni, L.; Orzali, L.; Vitale, S.; Luongo, L.; Infantino, A. New-Generation Sequencing Technology in Diagnosis of Fungal Plant Pathogens: A Dream Comes True? J. Fungi 2022, 8, 737. [Google Scholar] [CrossRef] [PubMed]
  104. Koskella, B.; Meaden, S.; Crowther, W.J.; Leimu, R.; Metcalf, C.J.E. A Signature of Tree Health? Shifts in the Microbiome and the Ecological Drivers of Horse Chestnut Bleeding Canker Disease. New Phytol. 2017, 215, 737–746. [Google Scholar] [CrossRef]
  105. Pinho, D.; Barroso, C.; Froufe, H.; Brown, N.; Vanguelova, E.; Egas, C.; Denman, S. Linking Tree Health, Rhizosphere Physicochemical Properties, and Microbiome in Acute Oak Decline. Forests 2020, 11, 1153. [Google Scholar] [CrossRef]
  106. Busby, P.E.; Newcombe, G.; Neat, A.S.; Averill, C. Facilitating Reforestation through the Plant Microbiome: Perspectives from the Phyllosphere. Annu. Rev. Phytopathol. 2022, 60, 337–356. [Google Scholar] [CrossRef]
  107. Sadeghi, M.; Panahi, B.; Mazlumi, A.; Hejazi, M.A.; Komi, D.E.A.; Nami, Y. Screening of Potential Probiotic Lactic Acid Bacteria with Antimicrobial Properties and Selection of Superior Bacteria for Application as Biocontrol Using Machine Learning Models. Lebenson. Wiss. Technol. 2022, 162, 113471. [Google Scholar] [CrossRef]
  108. Gómez-Lama Cabanás, C.; Legarda, G.; Ruano-Rosa, D.; Pizarro-Tobías, P.; Valverde-Corredor, A.; Niqui, J.L.; Triviño, J.C.; Roca, A.; Mercado-Blanco, J. Indigenous pseudomonas Spp. Strains from the Olive (Olea europaea L.) Rhizosphere as Effective Biocontrol Agents against Verticillium dahliae: From the Host Roots to the Bacterial Genomes. Front. Microbiol. 2018, 9, 277. [Google Scholar] [CrossRef]
  109. Prigigallo, M.I.; Gómez-Lama Cabanás, C.; Mercado-Blanco, J.; Bubici, G. Designing a Synthetic Microbial Community Devoted to Biological Control: The Case Study of Fusarium Wilt of Banana. Front. Microbiol. 2022, 13, 967885. [Google Scholar] [CrossRef]
  110. Botero, D.; Alvarado, C.; Bernal, A.; Danies, G.; Restrepo, S. Network Analyses in Plant Pathogens. Front. Microbiol. 2018, 9, 35. [Google Scholar] [CrossRef]
  111. Sheikh, M.; Iqra, F.; Ambreen, H.; Pravin, K.A.; Ikra, M.; Chung, Y.S. Integrating Artificial Intelligence and High-Throughput Phenotyping for Crop Improvement. J. Integr. Agric. 2024, 23, 1787–1802. [Google Scholar] [CrossRef]
  112. Guo, L.; Zhao, G.; Xu, J.-R.; Kistler, H.C.; Gao, L.; Ma, L.-J. Compartmentalized Gene Regulatory Network of the Pathogenic Fungus Fusarium Graminearum. New Phytol. 2016, 211, 527–541. [Google Scholar] [CrossRef] [PubMed]
  113. Sperschneider, J.; Gardiner, D.M.; Dodds, P.N.; Tini, F.; Covarelli, L.; Singh, K.B.; Manners, J.M.; Taylor, J.M. EffectorP: Predicting Fungal Effector Proteins from Secretomes Using Machine Learning. New Phytol. 2016, 210, 743–761. [Google Scholar] [CrossRef] [PubMed]
  114. Sperschneider, J.; Dodds, P.N.; Singh, K.B.; Taylor, J.M. ApoplastP: Prediction of Effectors and Plant Proteins in the Apoplast Using Machine Learning. New Phytol. 2018, 217, 1764–1778. [Google Scholar] [CrossRef] [PubMed]
  115. Ornella, L.; González-Camacho, J.M.; Dreisigacker, S.; Crossa, J. Applications of Genomic Selection in Breeding Wheat for Rust Resistance. Methods Mol. Biol. 2017, 1659, 173–182. [Google Scholar]
  116. Gan, Y.; Liu, Z.; Zhang, F.; Xu, Q.; Wang, X.; Xue, H.; Su, X.; Ma, W.; Long, Q.; Ma, A.; et al. Deep Learning Empowers Genomic Selection of Pest-Resistant Grapevine. Hortic. Res. 2025, 12, uhaf128. [Google Scholar] [CrossRef]
  117. Ahuja, M.R.; Neale, D.B. Evolution of Genome Size in Conifers. Silvae Genet. 2005, 54, 126–137. [Google Scholar] [CrossRef]
  118. Lebedev, V.G.; Lebedeva, T.N.; Chernodubov, A.I.; Shestibratov, K.A. Genomic Selection for Forest Tree Improvement: Methods, Achievements and Perspectives. Forests 2020, 11, 1190. [Google Scholar] [CrossRef]
  119. Bayer, P.E.; Petereit, J.; Danilevicz, M.F.; Anderson, R.; Batley, J.; Edwards, D. The Application of Pangenomics and Machine Learning in Genomic Selection in Plants. Plant Genome 2021, 14, e20112. [Google Scholar] [CrossRef]
  120. Li, Z.; Paul, R.; Ba Tis, T.; Saville, A.C.; Hansel, J.C.; Yu, T.; Ristaino, J.B.; Wei, Q. Non-Invasive Plant Disease Diagnostics Enabled by Smartphone-Based Fingerprinting of Leaf Volatiles. Nat. Plants 2019, 5, 856–866. [Google Scholar] [CrossRef] [PubMed]
  121. Garrett, K.A.; Bebber, D.P.; Etherton, B.A.; Gold, K.M.; Plex Sulá, A.I.; Selvaraj, M.G. Climate Change Effects on Pathogen Emergence: Artificial Intelligence to Translate Big Data for Mitigation. Annu. Rev. Phytopathol. 2022, 60, 357–378. [Google Scholar] [CrossRef] [PubMed]
  122. Heuertz, M.; Carvalho, S.B.; Galindo, J.; Rinkevich, B.; Robakowski, P.; Aavik, T.; Altinok, I.; Barth, J.M.I.; Cotrim, H.; Goessen, R.; et al. The Application Gap: Genomics for Biodiversity and Ecosystem Service Management. Biol. Conserv. 2023, 278, 109883. [Google Scholar] [CrossRef]
  123. Kremer, A.; Chen, J.; Lascoux, M. “Chimes of Resilience”: What Makes Forest Trees Genetically Resilient? New Phytol. 2025, 246, 1934–1951. [Google Scholar] [CrossRef]
  124. Malo, P.; Tahvonen, O.; Suominen, A.; Back, P.; Viitasaari, L. Reinforcement Learning in Optimizing Forest Management. Can. J. For. Res. 2021, 51, 1393–1409. [Google Scholar] [CrossRef]
  125. Zhao, J.; Wang, J.; Yin, J.; Chen, Y.; Wu, B. Optimization of the Stand Structure in Secondary Forests of Pinus Yunnanensis Based on Deep Reinforcement Learning. Forests 2024, 15, 2181. [Google Scholar] [CrossRef]
  126. Estrada, J.S.; Fuentes, A.; Reszka, P.; Auat Cheein, F. Machine Learning Assisted Remote Forestry Health Assessment: A Comprehensive State of the Art Review. Front. Plant Sci. 2023, 14, 1139232. [Google Scholar] [CrossRef]
  127. Xu, Z.; Jiang, D. AI-Powered Plant Science: Transforming Forestry Monitoring, Disease Prediction, and Climate Adaptation. Plants 2025, 14, 1626. [Google Scholar] [CrossRef]
  128. Quigley, M.F. Street Trees and Rural Conspecifics: Will Long-Lived Trees Reach Full Size in Urban Conditions? Urban Ecosyst. 2004, 7, 29–39. [Google Scholar] [CrossRef]
  129. Sarkar, C.; Gupta, D.; Gupta, U.; Hazarika, B.B. Leaf Disease Detection Using Machine Learning and Deep Learning: Review and Challenges. Appl. Soft Comput. 2023, 145, 110534. [Google Scholar] [CrossRef]
  130. Luca Liehner, G.; Hick, A.; Biermann, H.; Brauner, P.; Ziefle, M. Perceptions, Attitudes and Trust toward Artificial Intelligence—An Assessment of the Public Opinion. In Proceedings of the AHFE International, San Francisco, CA, USA, 20–24 July 2023; Volume 72. [Google Scholar]
  131. Grealey, J.; Lannelongue, L.; Saw, W.-Y.; Marten, J.; Méric, G.; Ruiz-Carmona, S.; Inouye, M. The Carbon Footprint of Bioinformatics. Mol. Biol. Evol. 2022, 39, msac034. [Google Scholar] [CrossRef] [PubMed]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Article Metrics

Citations

Article Access Statistics

Journal Statistics
Multiple requests from the same IP address are counted as one view.
Download PDF
Urban Greening Strategies and Ecosystem Services: The Differential Impact of Street-Level Greening Structures on Housing Prices
Previous
Trees, Deadwood and Tree-Related Microhabitats Explain Patterns of Alpha and Beta Saproxylic Beetle Diversity in Silver Fir-Beech Forests in Central Italy
Next