Geoarchaeological evidence from Angkor, Cambodia, reveals a gradual decline rather than a catastrophic 15th-century collapse.

Alternative models exist for the movement of large urban populations following the 15th-century CE abandonment of Angkor, Cambodia. One model emphasizes an urban diaspora following the implosion of state control in the capital related, in part, to hydroclimatic variability. An alternative model suggests a more complex picture and a gradual rather than catastrophic demographic movement. No decisive empirical data exist to distinguish between these two competing models. Here we show that the intensity of land use within the economic and administrative core of the city began to decline more than one century before the Ayutthayan invasion that conventionally marks the end of the Angkor Period. Using paleobotanical and stratigraphic data derived from radiometrically dated sediment cores extracted from the 12th-century walled city of Angkor Thom, we show that indicia for burning, forest disturbance, and soil erosion all decline as early as the first decades of the 14th century CE, and that the moat of Angkor Thom was no longer being maintained by the end of the 14th century. These data indicate a protracted decline in occupation within the economic and administrative core of the city, rather than an abrupt demographic collapse, suggesting the focus of power began to shift to urban centers outside of the capital during the 14th century.

Uncovering archaeological landscapes at Angkor using lidar.

Previous archaeological mapping work on the successive medieval capitals of the Khmer Empire located at Angkor, in northwest Cambodia (∼9th to 15th centuries in the Common Era, C.E.), has identified it as the largest settlement complex of the preindustrial world, and yet crucial areas have remained unmapped, in particular the ceremonial centers and their surroundings, where dense forest obscures the traces of the civilization that typically remain in evidence in surface topography. Here we describe the use of airborne laser scanning (lidar) technology to create high-precision digital elevation models of the ground surface beneath the vegetation cover. We identify an entire, previously undocumented, formally planned urban landscape into which the major temples such as Angkor Wat were integrated. Beyond these newly identified urban landscapes, the lidar data reveal anthropogenic changes to the landscape on a vast scale and lend further weight to an emerging consensus that infrastructural complexity, unsustainable modes of subsistence, and climate variation were crucial factors in the decline of the classical Khmer civilization.

Bayesian regression versus machine learning for rapid age estimation of archaeological features identified with lidar at Angkor.

Lidar (light-detection and ranging) has revolutionized archaeology. We are now able to produce high-resolution maps of archaeological surface features over vast areas, allowing us to see ancient land-use and anthropogenic landscape modification at previously un-imagined scales. In the tropics, this has enabled documentation of previously archaeologically unrecorded cities in various tropical regions, igniting scientific and popular interest in ancient tropical urbanism. An emerging challenge, however, is to add temporal depth to this torrent of new spatial data because traditional archaeological investigations are time consuming and inherently destructive. So far, we are aware of only one attempt to apply statistics and machine learning to remotely-sensed data in order to add time-depth to spatial data. Using temples at the well-known massive urban complex of Angkor in Cambodia as a case study, a predictive model was developed combining standard regression with novel machine learning methods to estimate temple foundation dates for undated Angkorian temples identified with remote sensing, including lidar. The model's predictions were used to produce an historical population curve for Angkor and study urban expansion at this important ancient tropical urban centre. The approach, however, has certain limitations. Importantly, its handling of uncertainties leaves room for improvement, and like many machine learning approaches it is opaque regarding which predictor variables are most relevant. Here we describe a new study in which we investigated an alternative Bayesian regression approach applied to the same case study. We compare the two models in terms of their inner workings, results, and interpretive utility. We also use an updated database of Angkorian temples as the training dataset, allowing us to produce the most current estimate for temple foundations and historic spatiotemporal urban growth patterns at Angkor. Our results demonstrate that, in principle, predictive statistical and machine learning methods could be used to rapidly add chronological information to large lidar datasets and a Bayesian paradigm makes it possible to incorporate important uncertainties-especially chronological-into modelled temporal estimates.

Provisioning an Early City: Spatial Equilibrium in the Agricultural Economy at Angkor, Cambodia.

A dominant view in economic anthropology is that farmers must overcome decreasing marginal returns in the process of intensification. However, it is difficult to reconcile this view with the emergence of urban systems, which require substantial increases in labor productivity to support a growing non-farming population. This quandary is starkly posed by the rise of Angkor (Cambodia, 9th-fourteenth centuries CE), one of the most extensive preindustrial cities yet documented through archaeology. Here, we leverage extensive documentation of the Greater Angkor Region to illustrate how the social and spatial organization of agricultural production contributed to its food system. First, we find evidence for supra-household-level organization that generated increasing returns to farming labor. Second, we find spatial patterns which indicate that land-use choices took transportation costs to the urban core into account. These patterns suggest agricultural production at Angkor was organized in ways that are more similar to other forms of urban production than to a smallholder system. Supplementary Information: The online version contains supplementary material available at 10.1007/s10816-021-09535-5.

Diachronic modeling of the population within the medieval Greater Angkor Region settlement complex.

Angkor is one of the world's largest premodern settlement complexes (9th to 15th centuries CE), but to date, no comprehensive demographic study has been completed, and key aspects of its population and demographic history remain unknown. Here, we combine lidar, archaeological excavation data, radiocarbon dates, and machine learning algorithms to create maps that model the development of the city and its population growth through time. We conclude that the Greater Angkor Region was home to approximately 700,000 to 900,000 inhabitants at its apogee in the 13th century CE. This granular, diachronic, paleodemographic model of the Angkor complex can be applied to any ancient civilization.

Evaluating remote sensing datasets and machine learning algorithms for mapping plantations and successional forests in Phnom Kulen National Park of Cambodia.

This study develops a modelling framework by utilizing multi-sensor imagery for classifying different forest and land use types in the Phnom Kulen National Park (PKNP) in Cambodia. Three remote sensing datasets (Landsat optical data, ALOS L-band data and LiDAR derived Canopy Height Model (CHM)) were used in conjunction with three different machine learning (ML) regression techniques (Support Vector Machines (SVM), Random Forests (RF) and Artificial Neural Networks (ANN)). These ML methods were implemented on (a) Landsat spectral data, (b) Landsat spectral band & ALOS backscatter data, and (c) Landsat spectral band, ALOS backscatter data, & LiDAR CHM data. The Landsat-ALOS combination produced more accurate classification results (95% overall accuracy with SVM) compared to Landsat-only bands for all ML models. Inclusion of LiDAR CHM (which is a proxy for vertical canopy heights) improved the overall accuracy to 98%. The research establishes that majority of PKNP is dominated by cashew plantations and the nearly intact forests are concentrated in the more inaccessible parts of the park. The findings demonstrate how different RS datasets can be used in conjunction with different ML models to map forests that had undergone varying levels of degradation and plantations.

Archaeological assessment reveals Earth's early transformation through land use.

Environmentally transformative human use of land accelerated with the emergence of agriculture, but the extent, trajectory, and implications of these early changes are not well understood. An empirical global assessment of land use from 10,000 years before the present (yr B.P.) to 1850 CE reveals a planet largely transformed by hunter-gatherers, farmers, and pastoralists by 3000 years ago, considerably earlier than the dates in the land-use reconstructions commonly used by Earth scientists. Synthesis of knowledge contributed by more than 250 archaeologists highlighted gaps in archaeological expertise and data quality, which peaked for 2000 yr B.P. and in traditionally studied and wealthier regions. Archaeological reconstruction of global land-use history illuminates the deep roots of Earth's transformation and challenges the emerging Anthropocene paradigm that large-scale anthropogenic global environmental change is mostly a recent phenomenon.

Semi-supervised machine learning approaches for predicting the chronology of archaeological sites: A case study of temples from medieval Angkor, Cambodia.

Archaeologists often need to date and group artifact types to discern typologies, chronologies, and classifications. For over a century, statisticians have been using classification and clustering techniques to infer patterns in data that can be defined by algorithms. In the case of archaeology, linear regression algorithms are often used to chronologically date features and sites, and pattern recognition is used to develop typologies and classifications. However, archaeological data is often expensive to collect, and analyses are often limited by poor sample sizes and datasets. Here we show that recent advances in computation allow archaeologists to use machine learning based on much of the same statistical theory to address more complex problems using increased computing power and larger and incomplete datasets. This paper approaches the problem of predicting the chronology of archaeological sites through a case study of medieval temples in Angkor, Cambodia. For this study, we have a large dataset of temples with known architectural elements and artifacts; however, less than ten percent of the sample of temples have known dates, and much of the attribute data is incomplete. Our results suggest that the algorithms can predict dates for temples from 821-1150 CE with a 49-66-year average absolute error. We find that this method surpasses traditional supervised and unsupervised statistical approaches for under-specified portions of the dataset and is a promising new method for anthropological inquiry.

The demise of Angkor: Systemic vulnerability of urban infrastructure to climatic variations.

Complex infrastructural networks provide critical services to cities but can be vulnerable to external stresses, including climatic variability. This vulnerability has also challenged past urban settlements, but its role in cases of historic urban demise has not been precisely documented. We transform archeological data from the medieval Cambodian city of Angkor into a numerical model that allows us to quantify topological damage to critical urban infrastructure resulting from climatic variability. Our model reveals unstable behavior in which extensive and cascading damage to infrastructure occurs in response to flooding within Angkor's urban water management system. The likelihood and extent of the cascading failure abruptly grow with the magnitude of flooding relative to normal flows in the system. Our results support the hypothesis that systemic infrastructural vulnerability, coupled with abrupt climatic variation, contributed to the demise of the city. The factors behind Angkor's demise are analogous to challenges faced by modern urban communities struggling with complex critical infrastructure.

Evaluating the ability of community-protected forests in Cambodia to prevent deforestation and degradation using temporal remote sensing data.

Community forests are known to play an important role in preserving forests in Cambodia, a country that has seen rapid deforestation in recent decades. The detailed evaluation of the ability of community-protected forests to retain forest cover and prevent degradation in Cambodia will help to guide future conservation management. In this study, a combination of remotely sensing data was used to compare the temporal variation in forest structure for six different community forests located in the Phnom Kulen National Park (PKNP) in Cambodia and to assess how these dynamics vary between community-protected forests and a wider study area. Medium-resolution Landsat, ALOS PALSAR data, and high-resolution LiDAR data were used to study the spatial distribution of forest degradation patterns and their impacts on above-ground biomass (AGB) changes. Analysis of the remotely sensing data acquired at different spatial resolutions revealed that between 2012 and 2015, the community forests had higher forest cover persistence and lower rates of forest cover loss compared to the entire study area. Furthermore, they faced lower encroachment from cashew plantations compared to the wider landscape. Four of the six community forests showed a recovery in canopy gap fractions and subsequently, an increase in the AGB stock. The levels of degradation decreased in forests that had an increase in AGB values. However, all community forests experienced an increase in understory damage as a result of selective tree removal, and the community forests with the sharpest increase in understory damage experienced AGB losses. This is the first time multitemporal high-resolution LiDAR data have been used to analyze the impact of human-induced forest degradation on forest structure and AGB. The findings of this work indicate that while community-protected forests can improve conservation outcomes to some extent, more interventions are needed to curb the illegal selective logging of valuable timber trees.

The deep human prehistory of global tropical forests and its relevance for modern conservation.

Significant human impacts on tropical forests have been considered the preserve of recent societies, linked to large-scale deforestation, extensive and intensive agriculture, resource mining, livestock grazing and urban settlement. Cumulative archaeological evidence now demonstrates, however, that Homo sapiens has actively manipulated tropical forest ecologies for at least 45,000 years. It is clear that these millennia of impacts need to be taken into account when studying and conserving tropical forest ecosystems today. Nevertheless, archaeology has so far provided only limited practical insight into contemporary human-tropical forest interactions. Here, we review significant archaeological evidence for the impacts of past hunter-gatherers, agriculturalists and urban settlements on global tropical forests. We compare the challenges faced, as well as the solutions adopted, by these groups with those confronting present-day societies, which also rely on tropical forests for a variety of ecosystem services. We emphasize archaeology's importance not only in promoting natural and cultural heritage in tropical forests, but also in taking an active role to inform modern conservation and policy-making.

The deep human prehistory of global tropical forests and its relevance for modern conservation.

Significant human impacts on tropical forests have been considered the preserve of recent societies, linked to large-scale deforestation, extensive and intensive agriculture, resource mining, livestock grazing and urban settlement. Cumulative archaeological evidence now demonstrates, however, that Homo sapiens has actively manipulated tropical forest ecologies for at least 45,000 years. It is clear that these millennia of impacts need to be taken into account when studying and conserving tropical forest ecosystems today. Nevertheless, archaeology has so far provided only limited practical insight into contemporary human-tropical forest interactions. Here, we review significant archaeological evidence for the impacts of past hunter-gatherers, agriculturalists and urban settlements on global tropical forests. We compare the challenges faced, as well as the solutions adopted, by these groups with those confronting present-day societies, which also rely on tropical forests for a variety of ecosystem services. We emphasize archaeology's importance not only in promoting natural and cultural heritage in tropical forests, but also in taking an active role to inform modern conservation and policy-making.

Correction: Mapping and Characterizing Selected Canopy Tree Species at the Angkor World Heritage Site in Cambodia Using Aerial Data.

[This corrects the article DOI: 10.1371/journal.pone.0121558.].

Incorporating Canopy Cover for Airborne-Derived Assessments of Forest Biomass in the Tropical Forests of Cambodia.

This research examines the role of canopy cover in influencing above ground biomass (AGB) dynamics of an open canopied forest and evaluates the efficacy of individual-based and plot-scale height metrics in predicting AGB variation in the tropical forests of Angkor Thom, Cambodia. The AGB was modeled by including canopy cover from aerial imagery alongside with the two different canopy vertical height metrics derived from LiDAR; the plot average of maximum tree height (Max_CH) of individual trees, and the top of the canopy height (TCH). Two different statistical approaches, log-log ordinary least squares (OLS) and support vector regression (SVR), were used to model AGB variation in the study area. Ten different AGB models were developed using different combinations of airborne predictor variables. It was discovered that the inclusion of canopy cover estimates considerably improved the performance of AGB models for our study area. The most robust model was log-log OLS model comprising of canopy cover only (r = 0.87; RMSE = 42.8 Mg/ha). Other models that approximated field AGB closely included both Max_CH and canopy cover (r = 0.86, RMSE = 44.2 Mg/ha for SVR; and, r = 0.84, RMSE = 47.7 Mg/ha for log-log OLS). Hence, canopy cover should be included when modeling the AGB of open-canopied tropical forests.

Mapping and characterizing selected canopy tree species at the Angkor World Heritage site in Cambodia using aerial data.

At present, there is very limited information on the ecology, distribution, and structure of Cambodia's tree species to warrant suitable conservation measures. The aim of this study was to assess various methods of analysis of aerial imagery for characterization of the forest mensuration variables (i.e., tree height and crown width) of selected tree species found in the forested region around the temples of Angkor Thom, Cambodia. Object-based image analysis (OBIA) was used (using multiresolution segmentation) to delineate individual tree crowns from very-high-resolution (VHR) aerial imagery and light detection and ranging (LiDAR) data. Crown width and tree height values that were extracted using multiresolution segmentation showed a high level of congruence with field-measured values of the trees (Spearman's rho 0.782 and 0.589, respectively). Individual tree crowns that were delineated from aerial imagery using multiresolution segmentation had a high level of segmentation accuracy (69.22%), whereas tree crowns delineated using watershed segmentation underestimated the field-measured tree crown widths. Both spectral angle mapper (SAM) and maximum likelihood (ML) classifications were applied to the aerial imagery for mapping of selected tree species. The latter was found to be more suitable for tree species classification. Individual tree species were identified with high accuracy. Inclusion of textural information further improved species identification, albeit marginally. Our findings suggest that VHR aerial imagery, in conjunction with OBIA-based segmentation methods (such as multiresolution segmentation) and supervised classification techniques are useful for tree species mapping and for studies of the forest mensuration variables.

A comprehensive archaeological map of the world's largest preindustrial settlement complex at Angkor, Cambodia.

The great medieval settlement of Angkor in Cambodia [9th-16th centuries Common Era (CE)] has for many years been understood as a "hydraulic city," an urban complex defined, sustained, and ultimately overwhelmed by a complex water management network. Since the 1980s that view has been disputed, but the debate has remained unresolved because of insufficient data on the landscape beyond the great temples: the broader context of the monumental remains was only partially understood and had not been adequately mapped. Since the 1990s, French, Australian, and Cambodian teams have sought to address this empirical deficit through archaeological mapping projects by using traditional methods such as ground survey in conjunction with advanced radar remote-sensing applications in partnership with the National Aeronautics and Space Administration (NASA)/Jet Propulsion Laboratory (JPL). Here we present a major outcome of that research: a comprehensive archaeological map of greater Angkor, covering nearly 3,000 km2, prepared by the Greater Angkor Project (GAP). The map reveals a vast, low-density settlement landscape integrated by an elaborate water management network covering>1,000 km2, the most extensive urban complex of the preindustrial world. It is now clear that anthropogenic changes to the landscape were both extensive and substantial enough to have created grave challenges to the long-term viability of the settlement.