More Like this

1. Forgotten forests: expanding potential land use in traditional Hawaiian agroecosystems, and the social-ecological implications

   https://doi.org/10.5751/ES-14656-280421  

   Lee, Tiffany; Lincoln, Noa (January 2023, Ecology and Society)

   The development of agricultural systems is a fundamental component of social-ecological transformation and a predominant factor influencing social behavior and structuring. However, oversimplification of traditional agricultural production often occurs and limits the understanding of past populations’ abilities to mitigate potential risks and enhance food security through effective land management strategies. The social-ecological traits that characterize the Hawaiian Islands provides a unique vantage to explore human ecodynamics over the longue durée and assess how these systems can be used to inform current and future land-use strategies, both locally and globally. Using the Hawaiian archipelago as a case study, digitized historical maps depicting a range of crop species and cropping systems were georeferenced to assess previous estimates of land use by early island populations and demonstrate the limitations of narratives constructed from previously modeled extents of land-use activity that rely solely on the preservation of archaeological remnants. The results of our mapped vegetation correspond well with the more intensive forms of agriculture that were included in previous models, but overall indicate that previous models do not fully represent the extent of land use by early island populations, missing vast applications of agroforestry and arboriculture. Based on our findings, we argue that the omission of cultivation systems not associated with physical infrastructure has vastly limited the comprehension of land use by early island populations and driven narratives in social-ecological dynamics that underestimate the extent of agricultural production while inferring sociopolitical outcomes based on the prevailing agricultural dichotomy. To remedy this limitation, we suggest a multimethods approach that integrates diverse data sets for an agricultural model that is more inclusive of all agricultural forms implemented by early Native Hawaiian populations and, therefore, is more representative of the extents of land use by island populations. 

   more » « less
2. Agroforestry Distribution and Contributions in Ancient Hawaiian Agriculture

   https://doi.org/10.1007/s10745-023-00471-4  

   Lincoln, Noa Kekuewa; Lee, Tiffany M.; Quintus, Seth; Haensel, Thomas P.; Chen, Qi (January 2024, Human Ecology)

   Abstract Agriculture is one of the most fundamental ways in which human societies interact with the environment. The form and function of agriculture have important socio-political implications in terms of yields, labor requirements, variability and resilience, and elite control. Hawai‘i has been used as a model system for the discussion of coupled human and natural systems, and how the uneven distribution of agricultural opportunities has manifested in the political ecology. However, consideration of agriculture has emphasized forms with physical infrastructure documented through archaeology and have not included arboricultural forms that were extensive among Pacific Islands. We leverage existing, independent data sets to build and validate spatial models of two intensities of arboriculture across the Hawaiian archipelago: Agroforestry and Novel Forest. Model validation demonstrates good accuracy that includes both expected and unexpected sources of errors. Results of the models demonstrate that arboricultural techniques accounted for ~70% of the agricultural potential by area and ~40% of the agricultural potential by yield. Unlike existing agricultural forms modeled, such as flooded wetland terrace cultivation and rainfed field production, which have strong distributional patterns based on the age of the islands, arboricultural potential is well distributed across all the islands. The extent, distribution, and characteristics of arboricultural methods provide important augmentation of the current narrative of production dynamics and distribution, and the political ecology, of pre-contact Hawai‘i. 

   more » « less
3. Resource Efficient Profiling of Spatial Variability in Performance of Regression Models

   https://doi.org/10.1109/BigData55660.2022.10020602  

   Carlson, Caleb; Warushavithana, Menuka; Mitra, Saptashwa; Barram, Kassidy; Ghosh, Sudipto; Breidt, Jay; Pallickara, Sangmi Lee; Pallickara, Shrideep (December 2022, 2022 IEEE International Conference on Big Data (Big Data))

   Scientists design models to understand phenomena, make predictions, and/or inform decision-making. This study targets models that encapsulate spatially evolving phenomena. Given a model, our objective is to identify the accuracy of the model across all geospatial extents. A scientist may expect these validations to occur at varying spatial resolutions (e.g., states, counties, towns, and census tracts). Assessing a model with all available ground-truth data is infeasible due to the data volumes involved. We propose a framework to assess the performance of models at scale over diverse spatial data collections. Our methodology ensures orchestration of validation workloads while reducing memory strain, alleviating contention, enabling concurrency, and ensuring high throughput. We introduce the notion of a validation budget that represents an upper-bound on the total number of observations that are used to assess the performance of models across spatial extents. The validation budget attempts to capture the distribution characteristics of observations and is informed by multiple sampling strategies. Our design allows us to decouple the validation from the underlying model-fitting libraries to interoperate with models constructed using different libraries and analytical engines; our advanced research prototype currently supports Scikit-learn, PyTorch, and TensorFlow. 

   more » « less
4. Modeling Hawaiian Agroecology: Depicting traditional adaptation to the world's most diverse environment

   https://doi.org/10.3389/fsufs.2023.1116929  

   Lincoln, Noa Kekuewa; Haensel, Thomas P.; Lee, Tiffany M. (March 2023, Frontiers in Sustainable Food Systems)

   Most neglected and underutilized crops were developed and utilized within indigenous agroecological cropping systems. While crop suitability must consider the constraints of the environment, the conditions of agroecological systems and the role of crops within those systems should be considered. Such consideration may guide the implementation of appropriate farming systems specific to different ecosystems and microhabitats. Using the Hawaiian archipelago as a model system of socioecological dynamics, we consider the distribution of agroecological systems and their associated crops to explore how agroecological suitability changes across climate, topography, and soils. We conduct spatial modeling of the potential nature and extent of seven agroecological archetypes based on historical records. The seven spatial models of pre-colonial agroecological systems produced extents distributed across much of the Hawaiian archipelago, with clear adaptive patterns within and across the islands. The distribution of cropping system further affects the appropriateness of crop species application. We argue that the consideration of agroecological niches and associated cropping systems is critical for realizing the potential of underutilized crops and improving the efficiency of contemporary agricultural systems. 

   more » « less
5. geodl: An R package for geospatial deep learning semantic segmentation using torch and terra

   https://doi.org/10.1371/journal.pone.0315127  

   Maxwell, Aaron E; Farhadpour, Sarah; Das, Srinjoy; Yang, Yalin (December 2024, PLOS ONE)

   Sun, Xiaoyong (Ed.)

   Convolutional neural network (CNN)-based deep learning (DL) methods have transformed the analysis of geospatial, Earth observation, and geophysical data due to their ability to model spatial context information at multiple scales. Such methods are especially applicable to pixel-level classification or semantic segmentation tasks. A variety of R packages have been developed for processing and analyzing geospatial data. However, there are currently no packages available for implementing geospatial DL in the R language and data science environment. This paper introduces the geodl R package, which supports pixel-level classification applied to a wide range of geospatial or Earth science data that can be represented as multidimensional arrays where each channel or band holds a predictor variable. geodl is built on the torch package, which supports the implementation of DL using the R and C++ languages without the need for installing a Python/PyTorch environment. This greatly simplifies the software environment needed to implement DL in R. Using geodl, geospatial raster-based data with varying numbers of bands, spatial resolutions, and coordinate reference systems are read and processed using the terra package, which makes use of C++ and allows for processing raster grids that are too large to fit into memory. Training loops are implemented with the luz package. The geodl package provides utility functions for creating raster masks or labels from vector-based geospatial data and image chips and associated masks from larger files and extents. It also defines a torch dataset subclass for geospatial data for use with torch dataloaders. UNet-based models are provided with a variety of optional ancillary modules or modifications. Common assessment metrics (i.e., overall accuracy, class-level recalls or producer’s accuracies, class-level precisions or user’s accuracies, and class-level F1-scores) are implemented along with a modified version of the unified focal loss framework, which allows for defining a variety of loss metrics using one consistent implementation and set of hyperparameters. Users can assess models using standard geospatial and remote sensing metrics and methods and use trained models to predict to large spatial extents. This paper introduces the geodl workflow, design philosophy, and goals for future development. 

   more » « less