Papers by Steven A Wernke
The 80th Annual Meeting of the Society for American Archaeology, 2015
Antiquity, 2023
Archaeological surveys conducted through the inspection of high-resolution satellite imagery prom... more Archaeological surveys conducted through the inspection of high-resolution satellite imagery promise to transform how archaeologists conduct large-scale regional and supra-regional research. However, conducting manual surveys of satellite imagery is labour- and time-intensive, and low target prevalence substantially increases the likelihood of miss-errors (false negatives). In this article, the authors compare the results of an imagery survey conducted using artificial intelligence computer vision techniques (Convolutional Neural Networks) to a survey conducted manually by a team of experts through the Geo-PACHA platform (for further details of the project, see Wernke et al. 2023). Results suggest that future surveys may benefit from a hybrid approach—combining manual and automated methods—to conduct an AI-assisted survey and improve data completeness and robustness.
Antiquity, 2023
Recent archaeological research in the Andes suggests that Indigenous herders carefully managed th... more Recent archaeological research in the Andes suggests that Indigenous herders carefully managed their environments through the modification of local hydrology and vegetation. However, the limited geographical scale of previous research makes it challenging to assess the range and prevalence of pastoralist land management in the Andes. In this article, the authors utilise large-scale, systematic imagery survey to examine the distribution and environmental contexts of corrals and pastoralist settlements in Huancavelica, Peru. Results indicate that corrals and pastoralist settlements cluster around colonial and present-day settlements and that a statistically significant relationship exists between pastoral infrastructure and perennial vegetation. This highlights the utility of remote survey for the identification of trans-regional patterns in herder-environment relationships that are otherwise difficult to detect.
Antiquity, 2023
The north coast of Peru is among the most extensively surveyed regions in the world, yet variatio... more The north coast of Peru is among the most extensively surveyed regions in the world, yet variation in research questions, sampling strategies and chronological and geospatial controls among survey projects makes comparison of disparate datasets difficult. To contextualise these issues, the authors present a systematic survey of satellite imagery focusing on hilltop fortifications in the Jequetepeque and Santa Valleys. This digital recontextualisation of pedestrian survey data demonstrates the potential of hybrid methodologies to substantially expand both the identification of archaeological sites within difficult terrain and, consequently, our understanding of the function of defensive sites.
Antiquity, 2023
Fog oases (lomas) present pockets of verdant vegetation within the arid coastal desert of Andean ... more Fog oases (lomas) present pockets of verdant vegetation within the arid coastal desert of Andean South America and archaeological excavation within some of the oases has revealed a long history of human exploitation of these landscapes. Yet lomas settlements are under-represented in archaeological datasets due to their tendency to be located in remote inter-valley areas. Here, the authors employ satellite imagery survey to map the locations of anthropogenic surface features along the central Peruvian coast. They observe two categories of archaeological features, large corrals and clustered structures, and document a concentration of settlement features within lomas landscapes that suggests a pre-Hispanic preference for both short- and long-term occupation of these verdant oases.
Antiquity, 2023
In the Andean highlands, hilltop fortifications known as pukaras are common. Dating predominantly... more In the Andean highlands, hilltop fortifications known as pukaras are common. Dating predominantly to the Late Intermediate Period (AD 1000–1450), pukaras are important to archaeological characterisations of a political landscape shaped by conflict but the distribution of these key sites is not well understood. Here, the authors employ systematic satellite imagery survey to provide a contiguous picture of pukara distribution on an inter-regional scale covering 151 103km2 in the south-central highlands of Peru. They highlight the effectiveness of such survey at identifying pukaras and capturing regional variability in size and residential occupation, and the results demonstrate that satellite surveys of high-visibility sites can tackle research questions at larger scales of analysis than have previously been possible.
Antiquity, 2023
Imagery-based survey is capable of producing archaeological datasets that complement those collec... more Imagery-based survey is capable of producing archaeological datasets that complement those collected through field-based survey methods, widening the scope of analysis beyond regions. The Geospatial Platform for Andean Culture, History and Archaeology (GeoPACHA) enables systematic registry of imagery survey data through a ‘federated’ approach. Using GeoPACHA, teams pursue problem-specific research questions through a common data schema and interface that allows for inter-project comparisons, analyses and syntheses. The authors present an overview of the platform's rationale and functionality, as well as a summary of results from the first survey campaign, which was carried out by six projects distributed across the central Andes, five of which are represented here.
Journal of the American College of Cardiology
BACKGROUND There is marked geographic variation in cardiac rehabilitation (CR) initiation, rangin... more BACKGROUND There is marked geographic variation in cardiac rehabilitation (CR) initiation, ranging from 10% to 40% of eligible patients at the state level. The potential causes of this variation, such as patient access to CR centers, are not well studied. OBJECTIVES The authors sought to determine how access to CR centers affects CR initiation in Medicare beneficiaries. METHODS The authors used Medicare files to identify CR-eligible Medicare beneficiaries and calculate CR initiation rates at the hospital referral region (HRR) level. We used linear regression to evaluate the percent variation in CR initiation accounted for by CR access across HRRs. We then employed geospatial hotspot analysis to identify CR deserts, or counties in which patient load per CR center is disproportionately high. RESULTS A total of 1,269,147 Medicare beneficiaries were eligible for CR from 2014 to 2017, of whom 314,997 (25%) initiated CR. The West North Central Census Division had the highest adjusted CR initiation rate (37.0%) and the highest density of CR programs (5.89 per 1,000 CR-eligible Medicare beneficiaries). Density of CR programs accounted for 23.5%
ACME: An International Journal for Critical Geographies, 2023
Formal spatial modeling and analytical approaches to maroon settlement, fugitivity, and warfare i... more Formal spatial modeling and analytical approaches to maroon settlement, fugitivity, and warfare in the colonial-era Caribbean have tended to mine historical cartographic sources instrumentally to analyze the distributions and simulate processes driving marronage in St. Croix (Dunnavant 2021b; Ejstrud 2008; Norton and Espenshade, 2007). Through close-in analysis, we compare two Danish maps of St. Croix produced in 1750 and 1799 in relation to modern cartographic sources, to explore how cartographic forms and cartesian conventions (attempt to) elide blind spots in the colonial gaze. By modeling possible subject-oriented
arXiv (Cornell University), Dec 13, 2021
Archaeology has long faced fundamental issues of sampling and scalar representation. Traditionall... more Archaeology has long faced fundamental issues of sampling and scalar representation. Traditionally, the local-to-regional-scale views of settlement patterns are produced through systematic pedestrian surveys. Recently, systematic manual survey of satellite and aerial imagery has enabled continuous distributional views of archaeological phenomena at interregional scales. However, such "brute force" manual imagery survey methods are both time-and labor-intensive, as well as prone to interobserver differences in sensitivity and specificity. The development of self-supervised learning methods (e.g., contrastive learning) offers a scalable learning scheme for locating archaeological features using unlabeled satellite and historical aerial images. However, archaeological features are generally only visible in a very small proportion relative to the landscape, while the modern contrastive-supervised learning approach typically yields an inferior performance on highly imbalanced datasets. In this work, we propose a framework to address this long-tail problem. As opposed to the existing contrastive learning approaches that typically treat the labeled and unlabeled data separately, our proposed method reforms the learning paradigm under a semisupervised setting in order to fully utilize the precious annotated data (<7% in our setting). Specifically, the highly unbalanced nature of the data is employed as the prior knowledge in order to form pseudo negative pairs by ranking the similarities between unannotated image patches and annotated anchor images. In this study, we used 95,358 unlabeled images and 5,830 labeled images in order to solve the issues associated with detecting ancient buildings from a long-tailed satellite image dataset. From the results, our semi-supervised contrastive learning model achieved a promising testing balanced accuracy of 79.0%, which is a 3.8% improvement as compared to other state-of-the-art approaches.
International Journal of Remote Sensing, 2023
Archaeology has long faced fundamental issues of sampling and scalar representation. Traditionall... more Archaeology has long faced fundamental issues of sampling and scalar representation. Traditionally, the local-to-regionalscale views of settlement patterns are produced through systematic pedestrian surveys. Recently, systematic manual survey of satellite and aerial imagery has enabled continuous distributional views of archaeological phenomena at interregional scales. However, such ‘brute force’ manual imagery survey methods are both time- and labour-intensive, as well as prone to inter-observer differences in sensitivity and specificity. The development of self-supervised learning methods (e.g. contrastive learning) offers a scalable learning scheme for locating archaeological features using unlabelled satellite and historical aerial images. However, archaeological features
are generally only visible in a very small proportion relative to the landscape, while the modern contrastive-supervised learning approach typically yields an inferior performance on highly imbalanced datasets. In this work, we propose a framework to address this long-tail problem. As opposed to the existing contrastive learning approaches that typically treat the labelled and unlabelled data separately, our proposed method
reforms the learning paradigm under a semi-supervised setting in order to fully utilize the precious annotated data (<7% in our setting). Specifically, the highly unbalanced nature of the data is employed as the prior knowledge in order to form pseudo negative pairs by ranking the similarities between unannotated image patches and annotated anchor images. In this study, we used 95,358 unlabelled images and 5,830 labelled images in order to solve the issues associated with detecting ancient buildings from a long-tailed satellite image dataset. From the results, our semi-supervised contrastive learning model achieved a promising testing balanced accuracy of 79.0%, which is a 3.8%
improvement as compared to other state-of-the-art approaches.
Journal of the American College of Cardiology, 2023
BACKGROUND There is marked geographic variation in cardiac rehabilitation (CR) initiation, rangin... more BACKGROUND There is marked geographic variation in cardiac rehabilitation (CR) initiation, ranging from 10% to 40% of eligible patients at the state level. The potential causes of this variation, such as patient access to CR centers, are not well studied. OBJECTIVES The authors sought to determine how access to CR centers affects CR initiation in Medicare beneficiaries. METHODS The authors used Medicare files to identify CR-eligible Medicare beneficiaries and calculate CR initiation rates at the hospital referral region (HRR) level. We used linear regression to evaluate the percent variation in CR initiation accounted for by CR access across HRRs. We then employed geospatial hotspot analysis to identify CR deserts, or counties in which patient load per CR center is disproportionately high. RESULTS A total of 1,269,147 Medicare beneficiaries were eligible for CR from 2014 to 2017, of whom 314,997 (25%) initiated CR. The West North Central Census Division had the highest adjusted CR initiation rate (37.0%) and the highest density of CR programs (5.89 per 1,000 CR-eligible Medicare beneficiaries). Density of CR programs accounted for 23.5%
Intermediate Elites in Pre-Columbian States and Empires
Journal of the Royal Anthropological Institute
Journal of Social Computing
Societies of the late prehispanic Andes-the Inkas principal among them-have long figured as "exce... more Societies of the late prehispanic Andes-the Inkas principal among them-have long figured as "exceptions to the rule" in social evolutionary schemata, in large measure because they seemingly lacked key technological hallmarks of complex societies found in other world regions, despite their observed large scale and complex, hierarchical political and economic formations. Such presumed absences are encoded in the Seshat Global History Databank, a large global comparative diachronic database recording many dimensions of human societies. Analyses derived from the current version of the Seshat database necessarily reproduce these supposed absences, as they inhere in its data ontology, structure, and registry. Nonetheless, patterns observed in the dataset provide a means for identifying processes acting on and through Andean peoples and the complex political formations they elaborated. Specifically, this paper evaluates a proposed information processing threshold model of social evolution, which suggests that social dynamics are driven first by processes related to social scale, and then by a phase of dynamics in which further scalar increases are only possible through innovations in information processing. The Andean region appears to violate this model because the Seshat database records writing and other information processing technologies as absent in the case of the Inka empire. The author argues that the dynamics of the Andean region are actually consistent with the information threshold model, but the data as constituted do not capture the relevant variables. The Inkas elaborated sophisticated information processing on par with counterparts in other world regions, but through radically distinct forms and pathways, including the Andean khipu (knotted string registries), decimal administration, and a colossal logistical and administrative infrastructural apparatus. This interwoven bundle of technologies and institutions constituted an information revolution that surpassed the information threshold and enabled explosive Inka imperial expansion, even as it produced certain vulnerabilities and fragile sovereignty.
AGU Fall Meeting Abstracts, Dec 1, 2015
Journal of Social Computing, 2022
Societies of the late prehispanic Andes-the Inkas principal among them-have long figured as "exce... more Societies of the late prehispanic Andes-the Inkas principal among them-have long figured as "exceptions to the rule" in social evolutionary schemata, in large measure because they seemingly lacked key technological hallmarks of complex societies found in other world regions, despite their observed large scale and complex, hierarchical political and economic formations. Such presumed absences are encoded in the Seshat Global History Databank, a large global comparative diachronic database recording many dimensions of human societies. Analyses derived from the current version of the Seshat database necessarily reproduce these supposed absences, as they inhere in its data ontology, structure, and registry. Nonetheless, patterns observed in the dataset provide a means for identifying processes acting on and through Andean peoples and the complex political formations they elaborated. Specifically, this paper evaluates a proposed information processing threshold model of social evolution, which suggests that social dynamics are driven first by processes related to social scale, and then by a phase of dynamics in which further scalar increases are only possible through innovations in information processing. The Andean region appears to violate this model because the Seshat database records writing and other information processing technologies as absent in the case of the Inka empire. The author argues that the dynamics of the Andean region are actually consistent with the information threshold model, but the data as constituted do not capture the relevant variables. The Inkas elaborated sophisticated information processing on par with counterparts in other world regions, but through radically distinct forms and pathways, including the Andean khipu (knotted string registries), decimal administration, and a colossal logistical and administrative infrastructural apparatus. This interwoven bundle of technologies and institutions constituted an information revolution that surpassed the information threshold and enabled explosive Inka imperial expansion, even as it produced certain vulnerabilities and fragile sovereignty.
Remote. Sens., 2021
This paper analyzes remotely sensed data sources to evaluate land-use history within the Peruvian... more This paper analyzes remotely sensed data sources to evaluate land-use history within the Peruvian department of Amazonas and demonstrates the utility of comparing present and past land-use patterns using continuous datasets, as a complement to the often dispersed and discrete data produced by archaeological and paleoecological field studies. We characterize the distribution of ancient (ca. AD 1–1550) terracing based on data drawn from high-resolution satellite imagery and compare it to patterns of deforestation between 2001 and 2019, based on time-series Landsat data. We find that the patterns reflected in these two datasets are statistically different, indicating a distinctive shift in land-use, which we link to the history of Inka and Spanish colonialism and Indigenous depopulation in the 15th through 17th centuries AD as well as the growth of road infrastructure and economic change in the recent past. While there is a statistically significant relationship between areas of ancien...
Uploads
Papers by Steven A Wernke
are generally only visible in a very small proportion relative to the landscape, while the modern contrastive-supervised learning approach typically yields an inferior performance on highly imbalanced datasets. In this work, we propose a framework to address this long-tail problem. As opposed to the existing contrastive learning approaches that typically treat the labelled and unlabelled data separately, our proposed method
reforms the learning paradigm under a semi-supervised setting in order to fully utilize the precious annotated data (<7% in our setting). Specifically, the highly unbalanced nature of the data is employed as the prior knowledge in order to form pseudo negative pairs by ranking the similarities between unannotated image patches and annotated anchor images. In this study, we used 95,358 unlabelled images and 5,830 labelled images in order to solve the issues associated with detecting ancient buildings from a long-tailed satellite image dataset. From the results, our semi-supervised contrastive learning model achieved a promising testing balanced accuracy of 79.0%, which is a 3.8%
improvement as compared to other state-of-the-art approaches.
are generally only visible in a very small proportion relative to the landscape, while the modern contrastive-supervised learning approach typically yields an inferior performance on highly imbalanced datasets. In this work, we propose a framework to address this long-tail problem. As opposed to the existing contrastive learning approaches that typically treat the labelled and unlabelled data separately, our proposed method
reforms the learning paradigm under a semi-supervised setting in order to fully utilize the precious annotated data (<7% in our setting). Specifically, the highly unbalanced nature of the data is employed as the prior knowledge in order to form pseudo negative pairs by ranking the similarities between unannotated image patches and annotated anchor images. In this study, we used 95,358 unlabelled images and 5,830 labelled images in order to solve the issues associated with detecting ancient buildings from a long-tailed satellite image dataset. From the results, our semi-supervised contrastive learning model achieved a promising testing balanced accuracy of 79.0%, which is a 3.8%
improvement as compared to other state-of-the-art approaches.
Los resultados indican que Mawchu Llacta fue establecida sobre un asentamiento de la época inkaica, confirmándose una secuencia ocupacional hasta mediados del siglo XIX. Sin embargo, y a pesar de los desplazamientos efectuados por la reducción, se reutilizaron espacios y elementos arquitectónicos del asentamiento inkaico como una plaza trapezoidal y piedras talladas de estilo cuzqueño en la iglesia y parroquia adyacente. De esta manera, la reducción habría sido diseñada como centro ritual y ceremonial de la localidad, debido a la presencia de una gran iglesia y de ocho capillas de variable tamaño.