Search for: All records

The ecology of forest ecosystems depends on the composition of trees. Capturing fine-grained information on individual trees at broad scales provides a unique perspective on forest ecosystems, forest restoration, and responses to disturbance. Individual tree data at wide extents promises to increase the scale of forest analysis, biogeographic research, and ecosystem monitoring without losing details on individual species composition and abundance. Computer vision using deep neural networks can convert raw sensor data into predictions of individual canopy tree species through labeled data collected by field researchers. Using over 40,000 individual tree stems as training data, we create landscape-level species predictions for over 100 million individual trees across 24 sites in the National Ecological Observatory Network (NEON). Using hierarchical multi-temporal models fine-tuned for each geographic area, we produce open-source data available as 1 km²shapefiles with individual tree species prediction, as well as crown location, crown area, and height of 81 canopy tree species. Site-specific models had an average performance of 79% accuracy covering an average of 6 species per site, ranging from 3 to 15 species per site. All predictions are openly archived and have been uploaded to Google Earth Engine to benefit the ecology community and overlay with other remote sensing assets. We outline the potential utility and limitations of these data in ecology and computer vision research, as well as strategies for improving predictions using targeted data sampling. 

Through an examination of three cases of change in the U-2 platform, this paper compares three pathways to changeability: form changes, operational changes, and cyber changes. Each pathway can lead to change in similar properties of a system but have varying levels of performance and time to implement. For each pathway, we describe the design mechanisms necessary to implement change in that pathway. We analyze the trade-off between performance or extent of change and agility or speed of change and find that form changes offer the highest degree of changeability but take the longest time to implement. Operational changes offer the least degree of changeability but are far quicker to implement. Cyber changes lie in between these two pathways. Understanding the design choices needed and the underlying trade-off of each pathway can enable decision-makers to better select a pathway to change when the need arises. This comparative analysis is especially useful since literature has thus far examined each of these pathways in isolation, not as different paths to the same goal. 

# Individual Tree Predictions for 100 million trees in the National Ecological Observatory Network Preprint: https://www.biorxiv.org/content/10.1101/2023.10.25.563626v1 ## Manuscript Abstract The ecology of forest ecosystems depends on the composition of trees. Capturing fine-grained information on individual trees at broad scales allows an unprecedented view of forest ecosystems, forest restoration and responses to disturbance. To create detailed maps of tree species, airborne remote sensing can cover areas containing millions of trees at high spatial resolution. Individual tree data at wide extents promises to increase the scale of forest analysis, biogeographic research, and ecosystem monitoring without losing details on individual species composition and abundance. Computer vision using deep neural networks can convert raw sensor data into predictions of individual tree species using ground truthed data collected by field researchers. Using over 40,000 individual tree stems as training data, we create landscape-level species predictions for over 100 million individual trees for 24 sites in the National Ecological Observatory Network. Using hierarchical multi-temporal models fine-tuned for each geographic area, we produce open-source data available as 1km^2 shapefiles with individual tree species prediction, as well as crown location, crown area and height of 81 canopy tree species. Site-specific models had an average performance of 79% accuracy covering an average of six species per site, ranging from 3 to 15 species. All predictions were uploaded to Google Earth Engine to benefit the ecology community and overlay with other remote sensing assets. These data can be used to study forest macro-ecology, functional ecology, and responses to anthropogenic change. ## Data Summary Each NEON site is a single zip archive with tree predictions for all available data. For site abbreviations see: https://www.neonscience.org/field-sites/explore-field-sites. For each site, there is a .zip and .csv. The .zip is a set 1km .shp tiles. The .csv is all trees in a single file. ## Prediction metadata *Geometry* A four pointed bounding box location in utm coordinates. *indiv_id* A unique crown identifier that combines the year, site and geoindex of the NEON airborne tile (e.g. 732000_4707000) is the utm coordinate of the top left of the tile.  *sci_name* The full latin name of predicted species aligned with NEON's taxonomic nomenclature.  *ens_score* The confidence score of the species prediction. This score is the output of the multi-temporal model for the ensemble hierarchical model.  *bleaf_taxa* Highest predicted category for the broadleaf submodel *bleaf_score* The confidence score for the broadleaf taxa submodel  *oak_taxa* Highest predicted category for the oak model  *dead_label* A two class alive/dead classification based on the RGB data. 0=Alive/1=Dead. *dead_score* The confidence score of the Alive/Dead prediction.  *site_id* The four letter code for the NEON site. See https://www.neonscience.org/field-sites/explore-field-sites for site locations. *conif_taxa* Highest predicted category for the conifer model *conif_score* The confidence score for the conifer taxa submodel *dom_taxa* Highest predicted category for the dominant taxa mode submodel *dom_score* The confidence score for the dominant taxa submodel ## Training data The crops.zip contains pre-cropped files. 369 band hyperspectral files are numpy arrays. RGB crops are .tif files. Naming format is __, for example. "NEON.PLA.D07.GRSM.00583_2022_RGB.tif" is RGB crop of the predicted crown of NEON data from Great Smoky Mountain National Park (GRSM), flown in 2022.Along with the crops are .csv files for various train-test split experiments for the manuscript. ### Crop metadata There are 30,042 individuals in the annotations.csv file. We keep all data, but we recommend a filtering step of atleast 20 records per species to reduce chance of taxonomic or data cleaning errors. This leaves 132 species. *score* This was the DeepForest crown score for the crop. *taxonID*For letter species code, see NEON plant taxonomy for scientific name: https://data.neonscience.org/taxonomic-lists *individual*unique individual identifier for a given field record and crown crop *siteID*The four letter code for the NEON site. See https://www.neonscience.org/field-sites/explore-field-sites for site locations. *plotID* NEON plot ID within the site. For more information on NEON sampling see: https://www.neonscience.org/data-samples/data-collection/observational-sampling/site-level-sampling-design *CHM_height* The LiDAR derived height for the field sampling point. *image_path* Relative pathname for the hyperspectral array, can be read by numpy.load -> format of 369 bands * Height * Weight *tile_year*  Flight year of the sensor data *RGB_image_path* Relative pathname for the RGB array, can be read by rasterio.open() # Code repository The predictions were made using the DeepTreeAttention repo: https://github.com/weecology/DeepTreeAttentionKey files include model definition for a [single year model](https://github.com/weecology/DeepTreeAttention/blob/main/src/models/Hang2020.py) and [Data preprocessing](https://github.com/weecology/DeepTreeAttention/blob/cae13f1e4271b5386e2379068f8239de3033ec40/src/utils.py#L59). 

The relationship between nutrient cycling and water quality in mixed-use ecosystems is driven by interactions among biotic and abiotic processes. However, the underlying processes cannot always be directly observed or modeled at broad spatial scales. Numerous empirical studies have employed land use patterns, variations in watershed physiography or disturbance regimes to characterize nutrient export from mixed-use watersheds, but simultaneously disentangling the effects of such factors remains challenging and few models directly incorporate vegetation biochemistry. Here we use structural equation models (SEMs) to assess the relative influence of foliar chemical traits (derived from imaging spectroscopy), watershed physiography, and human land use on the water quality (summer baseflow nitrate-N and soluble reactive phosphorus concentration) in watersheds across the Upper Midwestern United States. We use an SEM to link water quality (stream nitrate-nitrogen and dissolved phosphorus) to foliar retention (AVIRIS-Classic derived foliar traits related to recalcitrance), watershed retention (wetland proportion, MODIS Tasseled Cap Wetness), runoff (agricultural and urban land use), and watershed leakiness (AVIRIS-Classic foliar nitrogen, nitrogen deposition). The SEMs confirmed that variables associated with foliar retention derived from imaging spectroscopy are negatively related to watershed leakiness (standardized path coefficient = −0.892) and positively to watershed retention (standardized path coefficient = 0.705), with features related to watershed retention and runoff exerting the strongest controls on water quality (standardized path coefficients of −0.270 and 0.331 respectively). Comparing forested and agricultural watersheds, we found significantly increased importance of foliar retention to watershed leakiness in forests compared to agriculture (standardized coefficients of −1.004 and −0.764 respectively), with measures of watershed retention more important to runoff and water quality in agricultural watersheds. The results illustrate the capacity of imaging spectroscopy to provide measures of foliar traits that influence nutrient cycling in watersheds. Ultimately, the results may help focus development and restoration policies towards building more resilient landscapes that take into consideration associations among functional traits of vegetation, physiography and climate. 

Complex engineered systems with long life cycles can expect to face operational uncertainty. Two common approaches to maintain system performance in an uncertain operating environment are flexibility, where the system is designed to change easily in response to a change in the operating environment and robustness, where the system is designed to sustain performance despite change. Prior work has examined how to design systems to be either flexible or robust, but so far this work largely assumes that these strategies are implemented during the design phase and that designers know the possible changes that the system will face. However, in practice, many systems face unforeseeable needs and must be modified to sustain value post-production. Through an inductive case study, this paper examines that process: documenting how aircraft were modified post-production to gain new capabilities for close air support in Operation Desert Storm. Consistent with prior studies, it finds that new capabilities can be gained through both changes to form and changes to tactics. Extending this line of work, this study examines the conditions under which each type of change is effective. Additionally, it highlights an important interaction between form and tactical changes that has not been well defined in existing literature.