PyFIA: analyzing and visualizing forest attributes using the United States Forest Inventory and Analysis database Authors: Xinyuan Wei, Daniel Hayes, Gregory McHale, Jianheng Zhao, Aaron Weiskittel, Adam Daigneault Published: 2025 ABSTRACT AbstractBackgroundThe United States Forest Service Forest Inventory and Analysis (FIA) database is extensive and complex, requiring significant processing to link its multiple tables. In addition, integrating FIA data with external datasets, such as climate data and remote sensing products, involves substantial preprocessing and computational effort, posing challenges for users without advanced programming expertise.ResultsTo efficiently process, analyze, and visualize forest attributes using the FIA database, we developed PyFIA, an open-source Python-based tool. PyFIA provides a suite of functions, including statistical analyses, spatial mapping, and a bookkeeping model for tracking forest biomass dynamics at different scales. Additionally, it can acquire climate information for each inventory plot, enabling in-depth investigations of how climate conditions influence the spatial and temporal patterns of forest attributes.ConclusionsThis program enhances the use of FIA inventory data in forest related studies, particularly for forest carbon. It also incorporates raster datasets, providing a valuable resource for research on forest ecosystems. PyFIA is designed with a modular structure and is openly available on GitHub, enabling easy access, customization, and continuous improvement. Users can contribute to its development, ensuring long-term sustainability. In addition, its flexible architecture allows for the integration of new functions, making it highly adaptable to diverse research needs.Supplementary InformationThe online version contains supplementary material available at 10.1186/s13021-025-00364-7.Keywords: Bookkeeping, Climate, FIA data, Forest attributes, Open-source, Python, Visualization Background As the largest terrestrial ecosystem, forests cover approximately 4.06 billion hectares globally, representing 31% of the land area [1]. They play an important role in the global carbon cycle, storing an estimated 870 ± 61 Pg of carbon [2, 3]. To assess forest attributes across varying temporal and spatial scales, forest inventory data are a critical resource [4]. The Forest Inventory and Analysis (FIA) program, conducted by the United States Department of Agriculture (USDA) Forest Service, provides extensive, long-term, national-scale inventory data [5, 6]. FIA is a nationally consistent network of permanent inventory plots that are remeasured on a rotating panel every five to ten years [7]. FIA data include a wide range of measurements, such as individual tree size, species, land condition, forest coverage, and the geographic locations of inventory plots [7, 8]. It also provides the expansion factors used for estimation at plot, county, and state scales [9]. These measurements allow estimation of forest attributes such as structure, growth, aboveground biomass, and merchantable volume at various spatial scales [10]. Therefore, this comprehensive dataset is essential for quantifying forest attributes, mapping spatial patterns, and tracking temporal trends across diverse landscapes in the United States [11–13]. Given the substantial amount of forest information recorded by the FIA database and its complex data structure, numerous advanced tools and programs have been developed to enable efficient data filtering, summarizing, visualizing, and analyzing, each offering unique capabilities. For instance, EVALIDator, provided by the USDA Forest Service, is a web-based tool that allows users to generate summaries of forest attributes, such as biomass and volume, at various spatial scales [14, 15]. rFIA, an R package, enables advanced processing of FIA data, offering flexibility in filtering, summarizing, and spatial analysis, making it suitable for large-scale and regional studies [16, 17]. Similarly, tidyFIA, another R package, simplifies FIA data acquisition and assembly, streamlining workflows for researchers needing consistent data organization [18]. FIESTA, an R-based estimation tool, expands on these capabilities by providing tools for statistical analysis and integration with GIS data, supporting complex estimation procedures for forest area and volume [19]. However, there remains a gap in programs that support comprehensive and spatial studies of forest attributes. To fill this gap, we developed PyFIA, a Python-based tool that streamlines FIA data processing and expands its applications in forest ecosystem research. First, PyFIA can efficiently filter, summarize, and analyze FIA data to address questions related to forest attributes. Second, it provides various visualization functions for mapping forest attributes directly from FIA data. Third, it integrates FIA data with land cover products to generate wall-to-wall attribute maps. Fourth, PyFIA can incorporate climate data for each FIA plot, enabling analyses of climate impacts on the spatial and temporal dynamics of forest attributes. Finally, it can track forest biomass dynamics using a bookkeeping approach. Together, these features make PyFIA a comprehensive tool for advancing forest ecosystem research. Methods PyFIA structure PyFIA integrates FIA data with GIS and remote sensing products to support diverse analyses and visualizations of forest attributes (Fig. 1). The program is built on a modular design, with a main script that imports these modules. It includes six modules (i.e., Plot Statistics, Regional Statistics, Map Visualization, Map Interpolation, Climate Acquisition, and Bookkeeping Model), each comprising multiple functions designed for specific analyses and visualizations. This structure enhances maintainability by compartmentalizing functionalities, making it easier to troubleshoot and update individual components without affecting the entire program. In addition, this format promotes flexibility and scalability, allowing users to modify or expand modules independently to accommodate other research objectives, such as analyzing specific forest metrics, integrating other remote sensing products, or applying advanced statistical methods. A summary of the PyFIA program structure and the six modules (i.e., Plot Statistics, Regional Statistics, Map Visualization, Map Interpolation, Climate Acquisition, and Bookkeeping Model), key input dataset, and the outputs. Results and figures showed in this figure, along with the analysis workflow, are provided in the Jupyter Notebook PyFIA functions The plot and regional statistics modules (Fig. 1) provide various statistical functions for analyzing forest attributes at the plot, county, state, and user-defined region levels (Table 1). These functions are useful for forest attribute estimates, providing foundational data for further analyses. In addition, similar to the FIA EVALIDator, PyFIA offers enhanced filtering capabilities that allow users to refine statistics and summaries by tree size, species, forest type, and other criteria. It can also generate detailed summaries of forest attributes for all plots within a state or a specified region. These outputs support additional analyses and mapping, such as assessing how climate conditions shape the forest composition [11] and how forest management affects species diversity [20]. Table 1Summary of statistical functions provided by the plot and regional statistics modules, with demonstrations available in the Jupyter notebookFunctionForest Attributes and Potential ApplicationsModule: Plot Statistics (pyFIA_Plot_Stats.py module)plot_D_H_statsProvides statewide, plot-level tree size statistics, including mean Diameter at Breast Height (DBH), DBH standard deviation, maximum DBH, mean height, height standard deviation, and maximum height; supports monitoring tree growth and evaluating size distributions.plot_metricsProvides statewide, plot-level forest structure metrics, including the number of trees above specified DBH or height percentiles (e.g., 25%, 50%, 75%, or user-defined percentage) and the mean DBH or height within those percentiles; supports forest structure assessment and LiDAR model calibration metric selection.plot_indicesProvides statewide, plot-level structural indices, including quadratic mean diameter (QMD), canopy relief ratio (CRR), forest vertical structure score (FVSS), basal area, trees per acre (TPA), and total tree count; supports detailed stand-level structure analysis.plot_compositionProvides statewide, plot-level forest composition metrics, including Shannon diversity, evenness, species richness, dominant tree species based on tree count or basal area, dominant tree species count, dominant tree species proportion, and basal area of dominant tree species; supports biodiversity and species dominance assessments.plot_f_speciesProvides statewide, plot-level tree counts and the proportion of trees calculated by count and by basal area for a given tree species; supports species distribution and dominance assessments.plot_biomassProvides statewide, plot-level aboveground, belowground, and total biomass density; supports carbon stock estimation and biomass mapping.plot_harvestProvides statewide, plot-level harvested volume; supports monitoring removals and evaluating harvest impacts on forest ecosystem.plot_typeProvides statewide forest type classification for each plot; supports tracking forest type distribution and detecting changes over time.Module: Regional Statistics (pyFIA_Regional_Stats.py)state_biomassProvides summarized forest biomass for a given state; supports state-level carbon budget development and reporting.county_biomassProvides average forest biomass for each county within a state; supports county-scale biomass assessment and comparison.state_summaryProvides summarized forest attributes, using outputs from Plot Statistics module, for a specified state; supports statewide forest condition and trend analysis.AOI_summaryProvides summarized forest attributes, using outputs from Plot Statistics module, for an area of interest defined by a shapefile; supports custom regional analyses. Summary of statistical functions provided by the plot and regional statistics modules, with demonstrations available in the Jupyter notebook The PyFIA program includes visualization functions that facilitate mapping at plot and regional scales. These maps provide spatially explicit representations of forest attribute distributions, enhancing the accessibility and interpretability of FIA data. The Map Visualization module (Fig. 1) includes three functions for mapping all inventory plots within a state, plot-level forest attributes, and county-level forest attributes, enabling flexible and customizable visualizations for diverse applications. It also generates a point shapefile of forest attributes, which can be further analyzed using GIS software such as ArcGIS. For broader spatial analyses, PyFIA supports the generation of wall-to-wall forest attribute maps through the Map Interpolation module (Fig. 1). This module integrates statewide plot-level forest attribute data, such as aboveground biomass, TPA, and basal area, with a land cover map of the state. It then applies interpolation methods, including nearest-neighbor, linear, or cubic, to generate wall-to-wall forest attribute maps. By using land cover data to constrain forested areas, the method produces more realistic and spatially accurate maps. It can also provide georeferenced TIFF-format forest attribute maps, allowing direct comparison with other spatial datasets and supporting further spatial analyses. Climate factors, including temperature, precipitation, and extreme weather events, play a critical role in shaping tree growth, forest composition, forest structure, and forest biomass [21, 22]. Consequently, understanding how climate influences forest ecosystems is essential for assessing and forecasting ecosystem dynamics and for developing sustainable forest management strategies [23, 24]. To support such analyses, the Climate Acquisition module (Fig. 1) was developed to derive climate conditions for each FIA inventory plot. The bookkeeping approach for forest biomass tracking systematically accounts for biomass gains, losses, and transfers between different biomass pools, to monitor forest biomass dynamics over time within a study area [25, 26]. Biomass gains are tracked through tree growth and afforestation, while losses result from mortality, harvesting, and disturbances. Transfers within biomass pools, such as from live trees to coarse woody debris, are accounted using empirical regression models or conversion factors [27]. The Bookkeeping Model (Fig. 1) has been developed as part of the PyFIA program to enable forest biomass tracking. This module uses FIA data to derive forest type-specific empirical stand growth models and conversion parameters. In addition, users can incorporate stand growth models from existing studies or derive them from their own forest data. To streamline the parameterization process, all parameters are stored in a CSV file, allowing users to easily modify and update parameters as needed. Additionally, the module requires key remote sensing inputs, including an initial-year forest biomass map, a forest type map, annual forest disturbance maps, and associated disturbance type information. PyFIA input data The primary input data consists of a state-level FIA forest inventory dataset, including seven key CSV-format tables, which can be downloaded directly from the FIA DataMart. Additionally, a default GIS dataset containing state and county boundaries across the United States is provided for various mapping functions (Fig. 1). These datasets serve as the foundation for basic forest attributes analysis and visualization. To generate wall-to-wall forest attribute maps, a land cover classification map is required to delineate forested areas, such as the National Land Cover Database (NLCD) from the United States Geological Survey (USGS) [28]. Furthermore, deriving plot-level climate conditions requires high-resolution raster datasets containing relevant climate variables. The program uses ClimateNA as its default climate data source, a high-resolution dataset providing gridded climate variables for North America at an approximate spatial resolution of 3 km2 [29, 30]. ClimateNA offers both historical climate normals and future projections across multiple temporal scales, including annual, seasonal, and monthly averages. The dataset includes a wide range of climate variables, such as mean annual temperature (MAT), mean annual precipitation (MAP), summer precipitation (PPT_sm), and relative humidity (RH). Other climate datasets, such as Daymet data [31], can also be used for inventory plot climate data acquisition by replacing the climate data in the corresponding folder. PyFIA will automatically detect the new dataset and derive climate conditions for each plot. To implement the bookkeeping function, remote sensing-derived products, including a baseline biomass map, a forest type map, and annual forest disturbance maps, are required. The selection of these datasets depends on the availability of data sources for the study area. To ensure proper functionality and data alignment, all input raster datasets must have a consistent geospatial projection and spatial resolution to avoid misregistration issues. PyFIA tools To promote the application of the PyFIA program, we also developed the PyFIA Tools program, which currently includes four main functions (see the Jupyter Notebook PyFIA_Tools.ipynb). First, it can recode the plot ID (PLOT) into a new field combining COUNTYCD + PLOT. In most states, the plot ID is unique statewide. However, in some states, such as Georgia, the plot ID numbering restarts at 1 for each county. This function resolves duplicate IDs by creating a unique identifier across the entire state. Second, it provides a tool to extract state-level land cover data from the national NLCD dataset. Users only need to input the two-letter state abbreviation and the Federal Information Processing Standard (FIPS) code. The tool then extracts the state-specific land cover data, which can be used directly in the interpolation function. Third, it offers a merge function for combining processed forest attribute tables from different states or subregions. This allows integration of results such as forest structure metrics from multiple states (e.g., the seven New England states) or merging climate data with biomass data to examine the influence of climate conditions on forest biomass. Finally, it includes a relationship analysis function for exploring various relationships between different variables. Results Jupyter notebook To enhance accessibility and demonstrate the functionality of PyFIA, a Jupyter Notebook (PyFIA_ME.ipynb) was developed as an interactive environment for users to explore its features. This notebook provides a user-friendly interface for executing PyFIA, enabling data processing, analysis, and visualization. By integrating code, outputs, and explanatory text, it offers a step-by-step guide to understanding the workflow for various functions. Users can modify parameters, perform analyses, and visualize results to support diverse forest ecosystem research needs. Forest attributes estimation The public Maine FIA dataset was used to demonstrate all statewide, plot-level forest attribute calculation functions and regional summary functions of the PyFIA program. The complete dataset was obtained from the FIA DataMart, and seven essential data tables were used, including TREE, COND, PLOT, COUNTY, TREE_GRM_ESTN, POP_PLOT_STRATUM_ASSGN, and POP_STRATUM. The TREE table contains individual tree attributes, including species, diameter, and biomass (Table S1). The COND table provides information on forest conditions, such as stand age. The PLOT table includes geospatial coordinates, plot-specific details, and expansion factors, which enable spatial aggregation and ensure biomass estimates are representative. The COUNTY table links plot-level data to administrative boundaries, facilitating county-level biomass aggregation. The TREE_GRM_ESTN table provides tree-level estimates of growth, removals, and mortality, which are used to quantify timber harvesting and assess stand dynamics. The POP_PLOT_STRATUM_ASSGN table assigns plots to specific inventory strata, while the POP_STRATUM table provides stratum-level area estimates, allowing for statistically robust extrapolation of plot-level data to the state level. For these estimations, global parameters were set to designate Maine as the target state, with the state abbreviation configured as State = ‘ME’ and the FIPS code assigned as FIPS = ‘23’. The analysis employed FIA inventory data collected between 2019 and 2023, with parameters set as inv_yr = 2023 to designate 2023 as the end year of the inventory period and n_years = 5 to specify a five-year inventory cycle (Users can also customize the analysis period to match their specific objectives.). No filters were applied to tree size or species, ensuring an assessment of all trees. Specifically, the parameters tree_size = 0, tree_spp = [], and tree_spg = [] included trees of all sizes and species (see the Jupyter Notebook PyFIA_ME.ipynb). Forest attributes visualization We used the PyFIA program to generate visualizations of all inventory plots, plot-level basal area, TPA, aboveground biomass density, and county-level average forest biomass density for Maine. The global parameters were set as State = ‘ME’ and FIPS = ‘23’, and FIA inventory data collected between 2019 and 2023 was used for both visualizations. The parameters were configured with inv_yr = 2021 to specify 2021 as the final year of the inventory period and n_years = 5 to define a five-year inventory period. No filters were applied for tree size or species, as the parameters were set to tree_size = 0, tree_spp = [], and tree_spg = [], ensuring that all trees, regardless of size or species, were included in this visualization (see the Jupyter Notebook PyFIA_ME.ipynb). These configurations enabled a visualization of forest attributes across all FIA inventory plots in the state of Maine. We applied the map interpolation module to generate the wall-to-wall biomass map for Maine. For this interpolation, the global parameters were set as State = ‘ME’ and FIPS = ‘23’, with FIA inventory data collected between 2019 and 2023. No filters were applied to tree size or species, ensuring that all trees were included in the map generation. The resolution for this interpolation was configured as resolution = 0.01 (0.01 × 0.01 degrees), with the spatial interpolation method specified as interp_method = ‘nearest’ (see the Jupyter Notebook PyFIA_ME.ipynb). In addition, the NLCD 2024 land cover map (spatial resolution: 30 × 30 m) was incorporated to constrain the forested regions, and the grid was resampled to match the resolution of the land cover map, resulting in a high-resolution biomass map with a spatial resolution of 30 × 30 m. Climate data acquisition To demonstrate the climate acquisition function in PyFIA, we used the statewide plot-level aboveground biomass dataset (ME_Plot_Biomass.csv) as the base data. The climate_data function from the climate module was applied to acquire and append climate information to each inventory plot using the default ClimateNA dataset, which is the historical climate normals for the period 1991–2020 (see the Jupyter Notebook PyFIA_ME.ipynb). This process enriched the FIA inventory data with climate variables including mean annual temperature (MAT), mean annual precipitation (MAP), climate moisture deficit (CMD), climate moisture index (CMI), and relative humidity (RH). Bookkeeping biomass tracking To illustrate the bookkeeping model in the PyFIA program, we applied the function (bookkeeping) to generate a forest aboveground biomass map for Maine in 2024 (see the Jupyter Notebook PyFIA_ME.ipynb). The model integrates various raster inputs, including a base forest aboveground biomass map for Maine in 2023 [32], a forest type map [33], and the forest disturbance map in 2024 [34] (Fig. 2). Key parameters, such as those modeling the empirical relationship between forest stand age and aboveground biomass for each forest type, were derived from FIA data. To estimate these parameters, the bookkeeping module also provides a function (stand_growth) that investigates the relationship between forest biomass and stand age. These parameters were incorporated into the bookkeeping function via the parameter file. Additionally, the file includes post-harvesting aboveground biomass flux parameters, which can be derived from empirical studies. In this study, when forests are harvested, a portion of the aboveground biomass is removed, and the model applies a harvesting efficiency parameter to estimate the remaining biomass. Following harvest, forested areas regenerate, with biomass accumulating over time based on growth patterns determined by forest type and stand age. By integrating these inputs and processes, the bookkeeping function offers a useful framework for tracking forest biomass and visualizing spatial variations in forest biomass. Workflow of the bookkeeping function for tracking aboveground forest biomass in Maine from 2023 to 2024 Discussion PyFIA provides an integrated workflow to process, analyze, and visualize FIA data at plot level, county level, and state level, with optional linkage to climate and remote sensing products. The modular design allows users to run only the steps they need, to add custom functions, and to adapt the workflow to different forest research questions. In addition to summary statistics and mapping, PyFIA includes a bookkeeping approach to track forest biomass dynamics, which is useful for evaluating forest carbon patterns through time and exploring management or disturbance impacts. The Jupyter Notebook offers an accessible entry point for users who prefer an interactive environment. The PyFIA program is designed to analyze and visualize forest attributes using FIA data across various geographic scales, ranging from one or multiple counties within a state to a single state or user-defined regions with a boundary data. For users needing to process FIA data across multiple states or counties in different states, the program allows data to be processed separately for each state or through an iterative approach. In addition, we applied PyFIA for various states and demonstrated its applicability for the FIA of different states (see the Jupyter Notebook PyFIA_ds.ipynb). Our applications confirm its applicability to different states. Several established tools support FIA data use, each with distinct strengths. EVALIDator enables rapid, predefined summaries through a web interface [35], which is convenient for common queries but less flexible for custom pipelines or integration with external datasets. rFIA and tidyFIA provide powerful data access, filtering, and estimation in R [17], and FIESTA offers robust estimation and design-based inference with connections to GIS [36]. PyFIA complements these efforts by offering a Python-based pathway that joins FIA data with climate variables and remote sensing inputs within a single workflow. For users already working in Python for data processing and geospatial analysis, PyFIA simplifies and accelerates movement between FIA tables, climate data, and spatial outputs in a unified environment. Despite these advantages, several constraints remain. First, the modular script design favors transparency and easy modification, but users must manage dependencies and environments, and any change in FIA file structures or naming conventions can require updates to both scripts and documentation. Second, computation and memory demands can be substantial for analyses spanning multiple states or producing high-resolution outputs, particularly for raster operations and interpolation, which may require extended processing times or access to high-performance computing resources. Third, public FIA plot coordinates are intentionally perturbed to protect landowner privacy, with locations randomly displaced by up to approximately 0.5 miles (0.8 km). This intentional perturbation reduces the precision of fine-scale spatial analyses. In addition, measured plot coordinates may contain positional errors in GPS acquisition or field recording, introducing further uncertainty when aligning plot data with high-resolution spatial layers. These factors limit the reliability of fine-scale spatial analyses and add uncertainty when linking to high-resolution raster data; therefore, plot-level spatial patterns should be interpreted with caution. Fourth, the current interpolation options are simple and fast but do not quantify spatial uncertainty and may fail to capture complex spatial structure. Finally, the bookkeeping module relies on empirical parameters for growth, removals, and transfers among pools, so results should be viewed as scenario-based estimates rather than definitive forecasts, and they depend on the representativeness of the fitted relationships. Successful runs also depend on consistent projections, extents, and resolutions across input raster datasets. Conclusions Unlike a Python package, which requires extensive maintenance to keep up with changes in Python versions and FIA data structures, the modular script-based design of PyFIA ensures long-term functionality and adaptability. The program is hosted on GitHub, allowing users to create branches, develop new functions, and expand its applications within the forest ecosystem research community. More importantly, this approach enables PyFIA to remain flexible as Python libraries evolve and FIA file structures or naming conventions change, a common challenge in FIA data processing programs. Additionally, the open-source framework encourages collaborative maintenance, ensuring its long-term usability and continuous improvement. Supplementary Information Supplementary Material 1 Abbreviations Forest inventory and analysis Trees per acre Federal Information Processing Standard Author contributions X.W. and D.H. conceptualized this program. X.W. and J.Z. synthesized the FIA forest inventory data and various remote sensing products. X.W. developed the PyFIA program and G.M. developed the Jupyter Notebook. X.W., D.H., and A.W. managed the project. X.W. wrote the original draft. X.W., D.H., G.M., J.Z., A.W., and A.D. contributed to review and edit this paper. Funding This work was supported by NASA Carbon Monitoring System grants (80NSSC21K0966 and 80NSSC25K7219), USDA Forest Service Northeastern States Research Cooperative grants (20-DG-11242307-025 and 5410861), USDA NIFA Sustainable Agricultural Systems grant (2023-68012-38992), and Maine Space Grant Consortium grant (SG-24-30). In addition, this work was supported by the McIntire-Stennis project award no. ME-041907, from the U.S. Department of Agriculture’s National Institute of Food and Agriculture. Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the authors and should not be construed to represent any official USDA or U.S. Government determination or policy. Data availability The PyFIA program is available at [https://github.com/xinyuanwylb19/PyFIA](https://github.com/xinyuanwylb19/PyFIA). The Maine demonstration Jupyter Notebook is available at [https://github.com/xinyuanwylb19/PyFIA/blob/main/PyFIA_ME.ipynb](https://github.com/xinyuanwylb19/PyFIA/blob/main/PyFIA_ME.ipynb). The Oregon demonstration Jupyter Notebook is available at [https://github.com/xinyuanwylb19/PyFIA/blob/main/PyFIA_OR.ipynb](https://github.com/xinyuanwylb19/PyFIA/blob/main/PyFIA_OR.ipynb). The tools demonstration Jupyter Notebook is available at [https://github.com/xinyuanwylb19/PyFIA/blob/main/PyFIA_Tools.ipynb](https://github.com/xinyuanwylb19/PyFIA/blob/main/PyFIA_Tools.ipynb). The raster data used for the bookkeeping model are available at [https://doi.org/10.7910/DVN/RVIRPA). The FIA data can be downloaded from [https://research.fs.usda.gov/products/dataandtools/fia-datamart](https://research.fs.usda.gov/products/dataandtools/fia-datamart). A potential baseline biomass map that can be used for the bookkeeping model is available from the USDA Forest Service at [https://data.fs.usda.gov/geodata/rastergateway/biomass/index.php](https://data.fs.usda.gov/geodata/rastergateway/biomass/index.php). Details on FIA metadata, including temporal availability and sampling frequency, are available at [https://research.fs.usda.gov/understory/forest-inventory-and-analysis-database-user-guide-nfi](https:/doi.org/https:/research.fs.usda.gov/understory/forest-inventory-and-analysis-database-user-guide-nfi). Other ClimateNA data can be found at [https://climatena.ca/spatialData] (https://climatena.ca/spatialData). Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1 Data Availability Statement The PyFIA program is available at [https://github.com/xinyuanwylb19/PyFIA](https://github.com/xinyuanwylb19/PyFIA). The Maine demonstration Jupyter Notebook is available at [https://github.com/xinyuanwylb19/PyFIA/blob/main/PyFIA_ME.ipynb](https://github.com/xinyuanwylb19/PyFIA/blob/main/PyFIA_ME.ipynb). The Oregon demonstration Jupyter Notebook is available at [https://github.com/xinyuanwylb19/PyFIA/blob/main/PyFIA_OR.ipynb](https://github.com/xinyuanwylb19/PyFIA/blob/main/PyFIA_OR.ipynb). The tools demonstration Jupyter Notebook is available at [https://github.com/xinyuanwylb19/PyFIA/blob/main/PyFIA_Tools.ipynb](https://github.com/xinyuanwylb19/PyFIA/blob/main/PyFIA_Tools.ipynb). The raster data used for the bookkeeping model are available at [https://doi.org/10.7910/DVN/RVIRPA). The FIA data can be downloaded from [https://research.fs.usda.gov/products/dataandtools/fia-datamart](https://research.fs.usda.gov/products/dataandtools/fia-datamart). A potential baseline biomass map that can be used for the bookkeeping model is available from the USDA Forest Service at [https://data.fs.usda.gov/geodata/rastergateway/biomass/index.php](https://data.fs.usda.gov/geodata/rastergateway/biomass/index.php). Details on FIA metadata, including temporal availability and sampling frequency, are available at [https://research.fs.usda.gov/understory/forest-inventory-and-analysis-database-user-guide-nfi](https:/doi.org/https:/research.fs.usda.gov/understory/forest-inventory-and-analysis-database-user-guide-nfi). Other ClimateNA data can be found at [https://climatena.ca/spatialData] (https://climatena.ca/spatialData).