The DOI matches its Crossref record; 55 of the 55 references it cites were found in a registry or an index; none of the 29 sources its claims cite is checked for support yet.
Unstructured satellite survey detects up to 20% of archaeological sites in coastal valleys of southern Peru
Open accessAbstract
Satellite survey is widely used for archaeological site discovery, but the efficacy of the method has received little systematic investigation. In this analysis, twelve study participants of different experience levels performed an unstructured remote survey of 197 km 2 in the Sama and Moquegua valleys of south central Peru where previous pedestrian surveys recorded 546 archaeological sites. Results indicate an average site discovery rate of 9.3% (0–18%, 95% range). The most experienced participants detect up to 20% (17–22%) of known archaeological sites. These detection rates can be used to derive reliable site frequency estimates on the Andean coast, which can be used in planning and budgeting for field efforts and estimating demographic patterns at large spatial scales that are difficult to achieve through pedestrian survey. More generally, this analysis offers a method for deriving correction terms specific to other parts of the world. Additionally, the results can serve as a baseline for evaluating the effectiveness of emerging artificial intelligence routines for archaeological site detection.
Overview
AI snapshot
Written from the paper’s full text.
TL;DR
In a 197 km² unstructured satellite survey of two coastal valleys in southern Peru with 546 known archaeological sites, 12 participants found an average of 9.3% of sites, with the most experienced reaching up to 20%. Detection rates varied significantly with education, Andean regional experience, GIS experience, and survey time, while false-positive rates averaged 19%.
Key findings
- Average site discovery was 9.3% (1–18% modeled 95% range; abstract reports 0–18%), with the most experienced participants detecting up to 20% (17–22%) and the least trained only 2% (1–3%).
- Significant predictors of discovery (p < 0.1) were education, Andean archaeology experience, GIS experience, and time spent surveying; pedestrian survey experience was not significant. Andean training produced a 335% increase from a median 4% to 13%; mean discovery was 5% for undergraduates, 8% for graduate students, and 13% for PhDs; GIS experience increased discovery ~106% from 5% to 10%; each additional minute surveying increased likelihood of discovery by 1%.
- False-positive rates ranged from 4–41% among participants, with a modeled mean of 19% (2–52%). Andean training increased false positives ~60% (10% to 16%); graduate students found ~310% more false positives than undergraduates (9% to 30%), with no significant difference between undergraduates and PhDs; most experienced surveyors had mean false-positive rates of 18% (8–23%), least experienced 15% (5–27%).
- Site age did not significantly affect discovery (p = 0.31), but site type did: Omo and Tumilaca sites were discovered more frequently than Chiribaya, Chen Chen, Estuquiña, Huaracane, and Inka sites (p < 0.01). Applying the rates, 100 detected sites imply 555–10,000 sites if surveyor experience is unknown, 454–588 if highly trained, and 3425–7388 if minimally trained.
Methods
- Twelve participants with varying archaeological training (four undergraduates, four graduate students, four PhDs) performed an unstructured remote survey of 197 km² in the Sama and Moquegua valleys of south central Peru, using Bing Maps imagery in QGIS with no prescribed search method or time limit.
- Ground control came from three prior pedestrian surveys totaling 546 known sites: Moquegua (160 km², 431 sites), Sama (7 km², 47 sites), and Lower Ilo (30 km², 68 sites). A mark within 118 m of a known site was counted as found; otherwise it was a false positive.
- Analysis used beta distribution models fit by maximum likelihood with Kolmogorov-Smirnov goodness-of-fit tests, and generalized linear mixed models in R with binomial (discovery) and Poisson (false positives) families, specifying site and surveyor as random effects and education, Andean specialty, GIS experience, pedestrian survey experience, and time as fixed effects; additional models tested site age and site type.
Limitations
- Paper-stated: the small participant pool (12) limits accurate quantification of effect sizes; participants were all archaeologists or archaeology students, not a random population, so results may not generalize to general crowdsourcing; results apply specifically to southern coastal Peru and should be extended cautiously; Bing imagery date and resolution are unreported; the ground surveys do not represent the full archaeological record; site size, composition, and preservation were outside scope.
- Paper-stated: the false-positive definition assumes pedestrian surveys are more reliable than satellite surveys; inter-coder reliability is a known issue in manual remote survey, though the authors argue this study helps address it.
- Evident from the text: the unstructured design left search strategy and effort uncontrolled, with no time limit; education categories had only four participants each; one ground-control survey (Goldstein 2005) lacked reported survey methods, and the authors assume comparability.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
0 of 29 cited sources checked · 0 supported · 0 flagged
- Findingown data
Training in Andean archaeology produces a 335% increase in discovery rate from a median site discovery rate of 4% to 13%.
“The strongest effect is found in Andean experience with training in Andean archaeology producing a 335% increase in discovery rate from a median site discovery rate of 4% to 13%.”
Quote in the paperSubstantial claim - Findingown data
Time spent surveying significantly (p < 0.1) affects site detection rates.
“Our binomial GLMM results demonstrate that the following variables significantly (p < 0.1) affect site detection rates: educational experience, experience in Andean archaeology, experience working with GIS, and time spent surveying (Table 3).”
Quote in the paperNot a substantial claim - Findingown data
The most experienced participants detect up to 20% (17-22%) of known archaeological sites.
“The most experienced participants detect up to 20% (17–22%) of known archaeological sites.”
Quote in the paperSubstantial claim - Findingown data
Increased time surveying produced a significant effect on discovery rates, with each additional minute spent surveying leading to a 1% increase in the likelihood of site discovery.
“Increased time surveying produced a significant effect on discovery rates, with each additional minute spent surveying leading to a 1% increase in the likelihood of site discovery.”
Quote in the paperNot a substantial claim - Findingown data
Formal archaeological training significantly affects site detection.
“We find that formal archaeological training significantly affects site detection.”
Quote in the paperSubstantial claim
- Findingown data
Experience in Andean archaeology significantly (p < 0.1) affects site detection rates.
“Our binomial GLMM results demonstrate that the following variables significantly (p < 0.1) affect site detection rates: educational experience, experience in Andean archaeology, experience working with GIS, and time spent surveying (Table 3).”
Quote in the paperSubstantial claim - Findingown data
Experience working with GIS significantly (p < 0.1) affects site detection rates.
“Our binomial GLMM results demonstrate that the following variables significantly (p < 0.1) affect site detection rates: educational experience, experience in Andean archaeology, experience working with GIS, and time spent surveying (Table 3).”
Quote in the paperNot a substantial claim - Findingown data
Educational experience significantly (p < 0.1) affects site detection rates.
“Our binomial GLMM results demonstrate that the following variables significantly (p < 0.1) affect site detection rates: educational experience, experience in Andean archaeology, experience working with GIS, and time spent surveying (Table 3).”
Quote in the paperNot a substantial claim - Findingown data
Pedestrian survey experience is not observed to produce a significant effect on site detection rates.
“Pedestrian survey experience is not observed to produce a significant effect.”
Quote in the paperNot a substantial claim - Findingown data
The participants identified a total of 824 archaeological sites, including false positives.
“The participants identified a total of 824 archaeological sites, including false positives (S4 File).”
Quote in the paperSubstantial claim- The authors state this as an assumption: Prior pedestrian surveys are necessarily more reliable than satellite surveys for defining false positives.
- Findingown data
Across the twelve study participants, the average site discovery rate was 9.3% (95% range 1-18%).
“Results indicate an average site discovery rate of 9.3% (0–18%, 95% range).”
- Also stated in: § Site discovery rates · plain
“From the modeled Beta distribution, we derive a mean discovery rate of 9.3% (1–18%).”
Quote in the paperSubstantial claim- The authors state this as an assumption: The survey methods and results of Goldstein (2005) are comparable to the other pedestrian surveys used as ground control.
- Also stated in: § Site discovery rates · plain
- Findingown data
Site discovery rates ranged from 1-20% among participants.
“Site discovery rates ranged from 1–20% among participants (see Table 2).”
Quote in the paperSubstantial claim - Findingown data
Experience with GIS had a strong effect on site discovery rate, producing an increase of approximately 106% from a mean of 5% to 10% between inexperienced and experienced GIS users.
“Finally, experience with GIS was also found to have a strong effect on site discovery rate, producing an increase of approximately 106% from a mean of 5% to 10% between inexperienced and experienced GIS users (Fig 3).”
Quote in the paperNot a substantial claim - Findingown data
Graduate students found approximately 310% more false positives than undergraduate students, from a mean of 9% to 30%.
“Graduate students found approximately 310% more false positives than undergraduate students, from a mean of 9% to 30%.”
Quote in the paperNot a substantial claim - Findingown data
Omo and Tumilaca associated sites in the Middle Moquegua Valley were discovered at a higher frequency than the Chiribaya, Chen Chen, Estuquiña, Huaracane, and Inka sites (p < 0.01).
“Omo and Tumilaca associated sites in the Middle Moquegua Valley were discovered at a higher frequency than the Chiribaya, Chen Chen, Estuquiña, Huaracane, and Inka sites (p < 0.01).”
Quote in the paperNot a substantial claim - Findingown data
Poisson general linear model analysis indicates that educational experience, experience in Andean archaeology, and the amount of time spent surveying exert a significant (p<0.1) effect on false positive rates.
“Poisson general linear model analysis indicates that the following variables exert a significant (p<0.1) effect on false positive rates: educational experience, experience in Andean archaeology, and the amount of time spent surveying.”
Quote in the paperNot a substantial claim - Findingown data
Participants found an average of 9.3% (1-18%) of archaeological sites identified by ground-based pedestrian survey.
“Participants found an average of 9.3% (1–18%) of archaeological sites identified by ground based pedestrian survey.”
Quote in the paperSubstantial claim - Findingown data
From the modeled Beta distribution, the mean false-positive rate is 19% (2-52%).
“From the modeled Beta distribution, we derive a mean false-positive rate of 19% (2–52%).”
Quote in the paperNot a substantial claim - Findingown data
The site-age model did not identify a significant relationship between the latest occupation of an archaeological site and its probability of discovery (p = 0.31).
“The site-age model did not identify a significant relationship between the latest occupation of an archaeological site and its probability of discovery (p = 0.31).”
Quote in the paperNot a substantial claim - Findingown data
Site type is observed to affect discovery rates.
“In contrast, site type is observed to affect discovery rates.”
Quote in the paperNot a substantial claim - Findingown data
The best-fit beta model produces shape parameters of α = 1.4 and β = 5.8.
“The best-fit beta model produces shape parameters of α = 1.4 and β = 5.8 (Fig 3).”
Quote in the paperNot a substantial claim - Findingown data
A KS goodness-of-fit test shows the beta model to be a plausible model for the data (KS D = 0.16; p = 0.84).
“A KS goodness-of-fit test shows the beta model to be a plausible model for the data (KS D = 0.16; p = 0.84).”
Quote in the paperNot a substantial claim - Findingown data
There was no significant difference in the number of false positives between undergraduate students and Ph.Ds.
“However, there was no significant difference in the number of false positives between undergraduate students and Ph.Ds.”
Quote in the paperNot a substantial claim - Findingown data
Whether or not a researcher has spent time in the field itself participating in pedestrian survey was not found to have a significant effect on site detection in satellite imagery.
“Whether or not a researcher has spent time in the field itself participating in pedestrian survey was not found to have a significant effect on site detection in satellite imagery.”
Quote in the paperNot a substantial claim - Findingreasoning
The results show that archaeological experience exerts a significant effect on site discovery rates and accuracy.
“The results show that archaeological experience exerts a significant effect on site discovery rates and accuracy.”
Quote in the paperSubstantial claim - Findingown data
False positive rates ranged from 4-41% among participants.
“False positive rates ranged from 4–41% among participants.”
Quote in the paperNot a substantial claim - Findingown data
Andean training had an effect on the frequency of false positives, with an increase of approximately 60%, from a mean of 10% to 16%.
“Andean training had an effect on the frequency of false positives, with an increase of approximately 60%, from a mean of 10% to 16%.”
Quote in the paperNot a substantial claim - Premisecitation
'Brute force' and automated approaches are time and resource intensive, often beyond the scope of research budgets.
“While ‘brute force’ and automated approaches offer distinct advantages for discovering archaeological sites in high resolution satellite imagery, they are also time and resource intensive–often beyond the scope of research budgets [29, 33, 34].”
Quote in the paperSubstantial claim0 of 3 sources checked- cites [29] J. Casana, 2014, Regional-Scale Archaeological Remote Sensing in the Age of Big Data: Automated Site Discovery vs. Brute Force Methods, Advances in Archaeological Practice source
- cites [33] DS Davis, 2019, Object-based image analysis: a review of developments and future directions of automated feature detection in landscape archaeology, Archaeological Prospection source
- cites [34] P VanValkenburgh, 2020 source
- Premisecitation
The most effective machine learning programs are still outperformed by teams of experts manually combing through large volumes of high resolution satellite imagery.
“However, these types of automated satellite surveys have a number of limitations, and previous research demonstrates that the most effective machine learning programs are still outperformed by teams of experts manually combing through large volumes of high resolution satellite imagery [28, 29].”
Quote in the paperSubstantial claim0 of 2 sources checked- cites [28] HA Orengo, 2020, Automated detection of archaeological mounds using machine-learning classification of multisensor and multitemporal satellite data, Proceedings of the National Academy of Sciences source
- cites [29] J. Casana, 2014, Regional-Scale Archaeological Remote Sensing in the Age of Big Data: Automated Site Discovery vs. Brute Force Methods, Advances in Archaeological Practice source
- Premisenone given
The general audiences employed in crowdsourcing approaches lack the anthropological, archaeological, or regional training needed to search successfully for archaeological sites.
“Foremost among these is that the general audiences employed in crowdsourcing approaches lack the anthropological, archaeological, or regional training needed to search successfully for archaeological sites.”
Quote in the paperSubstantial claim - Study design and conductown data
The geographic error between georeferenced archaeological sites and their corresponding positions in satellite imagery is less than 118 m 95% of the time.
“Our error analysis suggests a typical error of less than 118 m. In other words, the analysis shows that the geographic error between the georeferenced archaeological sites and their corresponding positions in satellite imagery is less than 118 m 95% of the time.”
Quote in the paperSubstantial claim - Study design and conductown data
The relatively small participant pool raises concern about the ability to accurately characterize effect sizes.
“However, the relatively small participant pool raises concern about the ability to accurately characterize effect sizes.”
Quote in the paperSubstantial claim - Premisecitation
To both the lay person and automated computer algorithms, it can be difficult or impossible to distinguish between archaeological and modern features.
“To both the lay person and automated computer algorithms, it can be difficult or impossible to distinguish between archaeological and modern features [22, 28, 30].”
Quote in the paperSubstantial claim0 of 3 sources checked- cites [22] J Casana, 2017, Satellite imagery-based monitoring of archaeological site damage in the Syrian civil war, PLOS ONE source
- cites [28] HA Orengo, 2020, Automated detection of archaeological mounds using machine-learning classification of multisensor and multitemporal satellite data, Proceedings of the National Academy of Sciences source
- cites [30] AYM Lin, 2014, Crowdsourcing the Unknown: The Satellite Search for Genghis Khan, PLOS ONE source
- Premisecitation
GLMM is relatively robust in terms of detecting effects.
“A recent simulation study shows that although GLMM is relatively robust terms of detecting effects, the effect of sample size on accurate quantification of effect size remains poorly understood [53].”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [53] SG Luke, 2017, Evaluating significance in linear mixed-effects models in R, Behav Res source
- Premisecitation
The effect of sample size on accurate quantification of effect size remains poorly understood.
“A recent simulation study shows that although GLMM is relatively robust terms of detecting effects, the effect of sample size on accurate quantification of effect size remains poorly understood [53].”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [53] SG Luke, 2017, Evaluating significance in linear mixed-effects models in R, Behav Res source
- Interpretationreasoning
Unstructured approaches may make satellite survey useful for answering substantive questions about past human demographics and land-use patterns.
“At most, such unstructured approaches may make satellite survey useful for answering substantive questions about past human demographics and land-use patterns.”
Quote in the paperNot a substantial claim - Interpretationown data
The detection rates can be used to derive reliable site frequency estimates on the Andean coast.
“These detection rates can be used to derive reliable site frequency estimates on the Andean coast, which can be used in planning and budgeting for field efforts and estimating demographic patterns at large spatial scales that are difficult to achieve through pedestrian survey.”
Quote in the paperNot a substantial claim - Study design and conductown data
The beta model offers a plausible model for the data (KS D = 0.15; p = 0.90).
“A KS goodness-of-fit test shows the beta model to offer a plausible model for the data (KS D = 0.15; p = 0.90).”
Quote in the paperNot a substantial claim - Interpretationreasoning
If the surveyor were highly trained, the model would estimate that 454-588 sites could be expected on the ground.
“For example, if the surveyor were highly trained, then the model would estimate that 454–588 sites could be expected on the ground.”
Quote in the paperNot a substantial claim - Study design and conductnone given
The participant population entirely consists of individuals pursuing degrees in anthropological archaeology or who have dedicated their careers to archaeological research.
“Our participant population entirely consists of individuals pursuing degrees in anthropological archaeology or have dedicated their careers to archaeological research.”
Quote in the paperSubstantial claim - Interpretationreasoning
Surveying a region through satellite imagery before entering the field, and then applying the correction-factors deduced from this analytical approach, may allow archaeologists to estimate the frequency of archaeological sites.
“Surveying a region through satellite imagery before entering the field, and then applying the correction-factors deduced from this analytical approach, may allow archaeologists to estimate the frequency of archaeological sites.”
Quote in the paperSubstantial claim - Interpretationreasoning
The wide range of possibilities in the uninformed model demonstrates that failure to account for experience strongly limits the predictive capacity of the model.
“The wide range of possibilities in our uninformed model demonstrates that failure to account for experience strongly limits the predictive capacity of our model.”
Quote in the paperSubstantial claim - Premisecitation
One of the primary issues with manual remote survey-regardless of the type of imagery used-is inter-coder reliability.
“One of the primary issues with manual remote survey—regardless of the type of imagery used—is inter-coder reliability [55, 59].”
Quote in the paperSubstantial claim0 of 2 sources checked- cites [55] S Quintus, 2017, The Efficacy and Analytical Importance of Manual Feature Extraction Using Lidar Datasets, Advances in Archaeological Practice source
- cites [59] K. Sadr, 2016, The Impact of Coder Reliability on Reconstructing Archaeological Settlement Patterns from Satellite Imagery: a Case Study from South Africa., Archaeological Prospection source
- Study design and conductown data
The use of the models as a quantitative guide for satellite survey is restricted by the challenges of the relatively small sample size.
“The use of our models as a quantitative guide for satellite survey is further restricted by the challenges of our relatively small sample size.”
Quote in the paperSubstantial claim - Premisecitation
Sadr et al.'s study was limited by their small sample size of only two individuals.
“However, this study was limited by their small sample size of only two individuals.”
Quote in the paperNot a substantial claim - Interpretationcitation
One reason that site type may affect detectability is that certain culturally specific architectural elements are more visible in satellite imagery.
“One reason that site type may affect detectability is simply that certain culturally specific architectural elements are more visible in satellite imagery.”
Quote in the paperNot a substantial claim - Interpretationown data
The degree to which different types and levels of training affect survey results was quantified, which theoretically allows analysts to calibrate satellite survey results based on the experience level of a surveyor.
“More interestingly, we were able to quantify the degree to which different types and levels of training affect survey results, which theoretically allows analysts to calibrate satellite survey results based on the experience level of a surveyor.”
Quote in the paperSubstantial claim - Interpretationreasoning
If a satellite survey of a previously unsurveyed region in coastal south-central Andes detected 100 sites, assuming a satellite surveyor of unknown archaeological experience, the basic model would predict that anywhere between 555 to 10,000 sites could be expected on the ground.
“For example, suppose a satellite survey of a previously unsurveyed region in coastal south-central Andes detected 100 sites. Assuming a satellite surveyor of unknown archaeological experience, our basic model would predict that anywhere between 555 to 10,000 sites (100/0.18 and 100/0.01, respectively) could be expected on the ground.”
Quote in the paperSubstantial claim - Interpretationreasoning
The results may provide a baseline for evaluating the efficacy of machine learning approaches to remote survey.
“Our results not only offer a tool for generating rough estimates archaeological site frequencies, but may also provide a baseline for evaluating the efficacy of machine learning approaches to remote survey.”
Quote in the paperSubstantial claim - Interpretationreasoning
The results offer a tool for generating rough estimates of archaeological site frequencies.
“Our results not only offer a tool for generating rough estimates archaeological site frequencies, but may also provide a baseline for evaluating the efficacy of machine learning approaches to remote survey.”
Quote in the paperSubstantial claim - Interpretationreasoning
If the surveyor were minimally trained, the model would predict 3425-7388 sites on the ground.
“Similarly, if the surveyor were minimally trained, the model would predict 3425–7388 sites on the ground.”
Quote in the paperNot a substantial claim - Premisecitation
Looters tend to target certain archaeological sites associated with cultures that produced polychrome ceramics, colorful textiles, precious metals, and other goods that can be sold for profit on black markets.
“Looters tend to target certain archaeological sites associated with cultures that produced polychrome ceramics, colorful textiles, precious metals, and other goods that can be sold for profit on black markets [56].”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [56] N Brodie, 2005, Looting and the World’s Archaeological Heritage: The Inadequate Response, Annual Review of Anthropology source
- Interpretationown data
The research project presented here helps to resolve the problem of inter-coder variability by demonstrating that coders with different educational experience levels can have broadly similar behavior when recording archaeological sites through remote survey.
“The research project presented here helps to resolve the problem of inter-coder variability by demonstrating that coders different educational experience levels can have broadly similar behavior when recording archaeological sites through remote survey.”
Quote in the paperSubstantial claim - Interpretationown data
The more experience a researcher has working in the region of interest and working with remote sensing technology and GIS, the more likely they are to find archaeological sites through remote survey.
“Not surprisingly, these data suggest that the more experience a researcher has working in the region of interest and working with remote sensing technology and GIS, the more likely they are to find archaeological sites through remote survey.”
Quote in the paperSubstantial claim - Interpretationcitation
By establishing the capabilities of human satellite surveyors using freely available imagery under ideal conditions, the paper provides an important point of comparison for machine learning results, especially those conducted in south central Peru and other desert regions.
“By establishing the capabilities of human satellite surveyors using freely available imagery under ideal conditions, we provide an important point of comparison for machine learning results–especially those conducted in south central Peru and other desert regions [58].”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [58] A Karamitrou, 2022, Towards the use of artificial intelligence deep learning networks for detection of archaeological sites, Surf Topogr: Metrol Prop source
- It may state two separate claims as one. 16
- Probably not a substantial claim. 29
It may state two separate claims as one.
Findingown dataGraduate students produced a discovery rate roughly 70% higher than undergraduate students, with mean discovery rates of 8% and 5%, respectively.
“With a mean discovery rate of 8%, graduate students produced a discovery rate roughly 70% higher than that of undergraduate students with a mean of 5%.”
Quote in the paperSubstantial claimIt may state two separate claims as one.
Findingown dataParticipants with a Ph.D. produced a discovery rate roughly 70% higher than graduate students, with mean discovery rates increasing from 8% to 13%.
“Participants with a Ph.D. furthermore produced a discovery rate roughly 70% higher than that of graduate students, with mean discovery rates increasing from 8% to 13%, respectively.”
Quote in the paperSubstantial claimIt may state two separate claims as one.
FindingreasoningSite type, but not age, significantly affects the probability of site detection.
“Furthermore, we observed that site type, but not age, significantly affects the probability of site detection.”
Quote in the paperSubstantial claimIt may state two separate claims as one.
Findingown dataThe GLMM shows that the most experienced surveyors (those with PhD degrees, regional experience, GIS experience, and pedestrian survey experience) produce mean false positive rates of 18% (8-23%), while the least experienced surveyors produce mean false positive rates of 15% (5-27%).
“In sum, the GLMM shows that the most experienced surveyors–those with PhD degrees, regional experience, GIS experience, and pedestrian survey experience–produce mean false positive rates of 18% (8–23%), while the least experienced surveyors produce mean false positive rates of 15% (5–27%).”
Quote in the paperNot a substantial claimIt may state two separate claims as one.
Findingown dataThe model shows that highly trained individuals find 20% (17-22%) of archaeological sites in a given region, with the least trained individuals finding only 2% (1-3%) of sites.
“Our model further shows that highly trained individuals find 20% (17–22%) of archaeological sites in a given region, with the least trained individuals finding only 2% (1–3%) of sites.”
Quote in the paperSubstantial claimIt may state two separate claims as one.
Findingown dataThe level of education, experience in Andean archaeology, experience working with GIS, and time spent surveying all impacted the number of verified archaeological sites found through satellite survey as anticipated.
“The level of education, experience in Andean archaeology, experience working with GIS, and time spent surveying all impacted the number of verified archaeological sites found through satellite survey as anticipated.”
Quote in the paperSubstantial claimProbably not a substantial claim.
Premisenone givenSatellite survey will fail to observe some proportion of archaeological sites, especially small artifact scatters.
“While satellite survey will certainly fail to observe some proportion of archaeological sites–especially small artifact scatters–it makes up for this shortcoming by allowing coverage of large areas at little to no cost, thus increasing detection rates for large sites.”
Quote in the paperSubstantial claimProbably not a substantial claim. The paper’s conclusions may not depend on it.
PremisecitationCasana (2014) demonstrates that the most effective form of satellite survey is the 'brute force' approach.
“Casana (2014) convincingly demonstrates that the most effective form of satellite survey is what they term the ‘brute force’ approach–a team of experts painstaking reviewing high volumes of satellite imagery, though their study did quantify detection rates [29].”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [29] J. Casana, 2014, Regional-Scale Archaeological Remote Sensing in the Age of Big Data: Automated Site Discovery vs. Brute Force Methods, Advances in Archaeological Practice source
Probably not a substantial claim. The paper’s conclusions may not depend on it.
PremisecitationMenze and Ur (2012) report a success rate of 70-90% for identifying known archaeological sites.
“For example, one automated satellite survey by Menze and Ur (2012) reports a success rate of 70–90% for identifying known archaeological sites, but also erroneously identifies a high volume of modern archaeological and geological features–with nearly 40% of all identified features being false positives [20].”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [20] BH Menze, 2012, Mapping patterns of long-term settlement in Northern Mesopotamia at a large scale, Proceedings of the National Academy of Sciences source
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
InterpretationreasoningSatellite survey makes up for this shortcoming by allowing coverage of large areas at little to no cost, thus increasing detection rates for large sites.
“While satellite survey will certainly fail to observe some proportion of archaeological sites–especially small artifact scatters–it makes up for this shortcoming by allowing coverage of large areas at little to no cost, thus increasing detection rates for large sites.”
Quote in the paperNot a substantial claimProbably not a substantial claim. The paper’s conclusions may not depend on it.
PremisecitationMenze and Ur (2012) erroneously identify a high volume of modern archaeological and geological features, with nearly 40% of all identified features being false positives.
“For example, one automated satellite survey by Menze and Ur (2012) reports a success rate of 70–90% for identifying known archaeological sites, but also erroneously identifies a high volume of modern archaeological and geological features–with nearly 40% of all identified features being false positives [20].”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [20] BH Menze, 2012, Mapping patterns of long-term settlement in Northern Mesopotamia at a large scale, Proceedings of the National Academy of Sciences source
Probably not a substantial claim. It may state two separate claims as one.
PremisecitationThe quantitative efficacy of unstructured satellite survey remains unknown, as does the extent to which experience affects survey efficacy.
“While previous studies have quantitatively demonstrated the utility of highly structured, systematic satellite surveys performed by teams of experts, the quantitative efficacy of unstructured satellite survey—the most pervasive form of satellite survey—remains unknown as does the extent to which experience affects survey efficacy [13].”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [13] J. Casana, 2020, Global-Scale Archaeological Prospection using CORONA Satellite Imagery: Automated, Crowd-Sourced, and Expert-led Approaches, Journal of Field Archaeology source
Probably not a substantial claim.
PremisecitationPrevious studies have quantitatively demonstrated the utility of highly structured, systematic satellite surveys performed by teams of experts.
“While previous studies have quantitatively demonstrated the utility of highly structured, systematic satellite surveys performed by teams of experts, the quantitative efficacy of unstructured satellite survey—the most pervasive form of satellite survey—remains unknown as does the extent to which experience affects survey efficacy [13].”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [13] J. Casana, 2020, Global-Scale Archaeological Prospection using CORONA Satellite Imagery: Automated, Crowd-Sourced, and Expert-led Approaches, Journal of Field Archaeology source
Probably not a substantial claim. The paper’s conclusions may not depend on it.
PremisecitationModern development can further hamper satellite survey, potentially obscuring archaeological sites and rendering them invisible to satellite photographs.
“Modern development can further hamper satellite survey, potentially obscuring archaeological sites and rendering them invisible to satellite photographs [32].”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [32] A Sarris, 2013, Integration of geophysical surveys, ground hyperspectral measurements, aerial and satellite imagery for archaeological prospection of prehistoric sites: the case study of Vészt\Ho-Mágor Tell, Hungary, Journal of Archaeological Science source
It may state two separate claims as one.
Study design and conductcitationThe ground surveys chosen for comparison do not represent the full archaeological record of the study areas, nor do they comprise the complete history of archaeological survey in each region.
“The ground surveys chosen for comparison with satellite survey in this study do not represent the full archaeological record of the study areas, nor do they comprise the complete history of archaeological survey in each region [47–49].”
Quote in the paperNot a substantial claim0 of 3 sources checked- cites [47] C. Stanish, 1991, A Late Pre-Hispanic Ceramic Chronology for the Upper Moquegua Valley, Peru, Fieldiana Anthropology source
- cites [48] BD Owen, 2005, Distant Colonies and Explosive Collapse: The Two Stages of the Tiwanaku Diaspora in the Osmore Drainage, Latin American Antiquity source
- cites [49] G. Zaro, 2007, Diversity Specialists: Coastal Resource Management and Historical Contingency in the Osmore Desert of Southern Peru, Latin American Antiquity source
Probably not a substantial claim. It may be too vague to check against evidence. It may state two separate claims as one.
Study design and conductreasoningWe proceed with confidence in interpreting the model's ability to detect relationships but with caution in terms of the model's ability to quantify effect sizes.
“We therefore proceed with confidence in interpreting the model’s ability to detect relationships but with caution in terms of the model’s ability to quantify effect sizes.”
Quote in the paperNot a substantial claimProbably not a substantial claim. It may be too vague to check against evidence. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.
InterpretationreasoningUnstructured surveys may be especially useful for junior scholars without the time or finances to dedicate to large-scale systematic surveys or the connections required to organize teams of experts.
“Unstructured surveys may be especially useful for junior scholars without the time or finances to dedicate to large-scale systematic surveys or the connections required to organize teams of experts.”
Quote in the paperNot a substantial claimProbably not a substantial claim. It may be too vague to check against evidence. The paper’s conclusions may not depend on it.
Interpretationnone givenThe results can serve as a baseline for evaluating the effectiveness of emerging artificial intelligence routines for archaeological site detection.
“Additionally, the results can serve as a baseline for evaluating the effectiveness of emerging artificial intelligence routines for archaeological site detection.”
Quote in the paperNot a substantial claimProbably not a substantial claim. It may be too vague to check against evidence.
InterpretationreasoningQuantitative specification of unstructured satellite survey efficacy offers a formal approach to estimating actual site frequencies without the time and effort needed to organize pedestrian or 'brute force' approaches.
“Quantitative specification of unstructured satellite survey efficacy offers a formal approach to estimating actual site frequencies without the time and effort needed to organize pedestrian or ‘brute force’ approaches.”
Quote in the paperNot a substantial claimProbably not a substantial claim.
InterpretationreasoningUnstructured approaches provide site frequency estimates in advance of pedestrian survey or structured satellite survey, which may be useful for time and financial budgeting.
“At least, they provide site frequency estimates in advance of pedestrian survey or structured satellite survey, which may be useful for time and financial budgeting.”
Quote in the paperSubstantial claimProbably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.
Study design and conductnone givenSites within the study region vary greatly by size, type, composition, and preservation.
“Additionally, sites within the study region vary greatly by size, type, composition, and preservation.”
Quote in the paperNot a substantial claimIt may state two separate claims as one.
InterpretationreasoningThe detection rates can be used in planning and budgeting for field efforts and estimating demographic patterns at large spatial scales.
“These detection rates can be used to derive reliable site frequency estimates on the Andean coast, which can be used in planning and budgeting for field efforts and estimating demographic patterns at large spatial scales that are difficult to achieve through pedestrian survey.”
Quote in the paperSubstantial claimProbably not a substantial claim.
InterpretationreasoningWithin the study region, it is unlikely that satellite survey can be more effective than pedestrian survey at site discovery given the intensity of archaeological survey in a region with high surface visibility.
“Although there are situations where satellite survey can be more effective than pedestrian survey at site discovery, within our study region this is unlikely to be the case given the intensity of archaeological survey in a region with high surface visibility.”
Quote in the paperSubstantial claim- The authors state this as an assumption: Prior pedestrian surveys are necessarily more reliable than satellite surveys for defining false positives.
Probably not a substantial claim.
PremisecitationFree and publicly available software such as Google Earth and Bing Maps provide satellite images with resolutions down to 61 centimeters.
“Free and publicly available software such as Google Earth and Bing Maps provide satellite images with resolutions down to 61 centimeters [11, 12].”
Quote in the paperSubstantial claim0 of 2 sources checked- cites [11] DA Contreras, 2010, Landscape and Environment: Insights from the Prehispanic Central Andes, J Archaeol Res source
- cites [12] DA Contreras, 2010, The utility of publicly-available satellite imagery for investigating looting of archaeological sites in Jordan, Journal of Field Archaeology source
Probably not a substantial claim. The paper’s conclusions may not depend on it.
Interpretationnone givenThis analysis offers a method for deriving correction terms specific to other parts of the world.
“More generally, this analysis offers a method for deriving correction terms specific to other parts of the world.”
Quote in the paperNot a substantial claimProbably not a substantial claim. The paper’s conclusions may not depend on it.
InterpretationreasoningA subtler reason site type may affect detectability may be related to the looting of archaeological sites.
“A subtler reason may be related to the looting of archaeological sites.”
Quote in the paperNot a substantial claimProbably not a substantial claim. The paper’s conclusions may not depend on it.
PremisecitationPrevious studies of remote survey note that archaeological sites located at high elevations and consisting of large geometric shapes-such as architectural structures-are much more likely to be detected than sites which lack these characteristics.
“Previous studies of remote survey note that archaeological sites located at high elevations and consisting of large geometric shapes—such as architectural structures—are much more likely to be detected than sites which lack these characteristics [54, 55].”
Quote in the paperNot a substantial claim0 of 2 sources checked- cites [54] O Risbøl, 2013, Interpreting cultural remains in airborne laser scanning generated digital terrain models: effects of size and shape on detection success rates, Journal of Archaeological Science source
- cites [55] S Quintus, 2017, The Efficacy and Analytical Importance of Manual Feature Extraction Using Lidar Datasets, Advances in Archaeological Practice source
Probably not a substantial claim. It may be too vague to check against evidence.
InterpretationreasoningSuch comparisons should permit assessment of the specific contexts in which machine learning succeeds or fails relative to small groups of experts.
“Such comparisons should furthermore permit assessment of the specific contexts in which machine learning succeeds or fails relative to small groups of experts.”
Quote in the paperNot a substantial claimProbably not a substantial claim.
PremisecitationSadr et al. (2016) find that different coders record archaeological features differently, regardless of training, boots-on-the-ground survey experience, or experience.
“Sadr et al. (2016) find that different coders record archaeological features differently, regardless of training, boots-on-the-ground survey experience, or experience.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites Sadr et al. (2016) source
Probably not a substantial claim. The paper’s conclusions may not depend on it.
Interpretationnone givenThe lack of structure minimizes time spent strategizing the search plan and maintaining search structure.
“Second, the lack of structure minimizes time spent strategizing the search plan and maintaining search structure.”
Quote in the paperNot a substantial claimIt may state two separate claims as one.
Interpretationown dataThe analysis confirms some, but not all, predictions regarding the effects of experience on site detection.
“Our analysis confirms some, but not all of our predictions regarding the effects of experience on site detection.”
Quote in the paperSubstantial claimProbably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds. It may state two separate claims as one.
Interpretationnone givenUnstructured searches require minimal training, allowing researchers to take advantage of crowd sourcing approaches.
“First, unstructured searches require minimal training allowing researchers to take advantage of crowd sourcing approaches.”
Quote in the paperNot a substantial claimProbably not a substantial claim. The paper’s conclusions may not depend on it.
PremisecitationStructured surveys yield more easily interpreted results.
“Structured surveys thus yield more easily interpreted results [29].”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [29] J. Casana, 2014, Regional-Scale Archaeological Remote Sensing in the Age of Big Data: Automated Site Discovery vs. Brute Force Methods, Advances in Archaeological Practice source
Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.
InterpretationreasoningThe presence of 'looter pits' may actually change the visibility of archaeological sites in satellite imagery, making sites that would otherwise be invisible to satellite survey detectable.
“The presence of ‘looter pits’ may actually change the visibility of archaeological sites in satellite imagery, making sites that would otherwise be invisible to satellite survey detectable.”
Quote in the paperNot a substantial claimIt may state two separate claims as one.
InterpretationreasoningThe efficacy of unstructured satellite survey may vary by experience and expertise, but even under optimal conditions (advanced education, experience in the region and with both GIS and pedestrian survey) satellite survey using freely available imagery only identifies ~20% of sites.
“In sum, the efficacy of unstructured satellite survey may vary by experience and expertise, but even under optimal conditions (advanced education, experience in the region and with both GIS and pedestrian survey) satellite survey using freely available imagery only identifies ~20% of sites.”
Quote in the paperSubstantial claimProbably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.
Interpretationnone givenA second disadvantage is that unstructured surveys can result in search redundancy and thus inefficiencies in the use of time.
“A second disadvantage is that unstructured surveys can result in search redundancy and thus inefficiencies in the use of time.”
Quote in the paperNot a substantial claimProbably not a substantial claim. It may be too vague to check against evidence. The paper’s conclusions may not depend on it.
InterpretationreasoningIn determining rates of error in satellite survey, archaeologists can intelligently generate rough estimates of site densities for planning field efforts.
“In determining rates of error in satellite survey, archaeologists can intelligently generate rough estimates of site densities for planning field efforts, both of which are central tasks of archaeological research.”
Quote in the paperNot a substantial claimIt may state two separate claims as one.
Interpretationown dataThe data are clear in showing that site-detection rates approaching 20% are readily attainable and that individuals with more training are more likely to approach such levels of success.
“Nonetheless, the data are clear in showing that site-detection rates approaching 20% are readily attainable and that, unsurprisingly, the individuals with more training are more likely to approach such levels of success.”
Quote in the paperSubstantial claimProbably not a substantial claim. It may be too vague to check against evidence. It may state two separate claims as one.
Study design and conductreasoningThe reported extent of the effects should be interpreted with caution, as should any attempt to use these models to predict site densities given surveyor experience.
“However, the reported extent of the effects should be interpreted with caution as should any attempt to use these models to predict site densities given surveyor experience.”
Quote in the paperNot a substantial claimProbably not a substantial claim. It may be too vague to check against evidence. The paper’s conclusions may not depend on it.
InterpretationreasoningUnstructured surveys are likely to be more efficient, as surveyors gain information about hot spots as they go and can prioritize their approaches towards high density regions.
“Third, unstructured surveys are likely to be more efficient, as surveyors gain information about hot spots as they go and can prioritize their approaches towards high density regions.”
Quote in the paperNot a substantial claim
The authors state this as an assumption.
The survey methods and results of Goldstein (2005) are comparable to the other pedestrian surveys used as ground control.
“Unfortunately, survey methods are not reported for the Goldstein (2005) survey, but we assume here that the survey methods and results are comparable.”
The authors state this as an assumption.
Prior pedestrian surveys are necessarily more reliable than satellite surveys for defining false positives.
“Our definition of false positive assumes that prior pedestrian surveys will necessarily be more reliable than satellite surveys.”
The paper relies on this without support.
The arid Atacama Desert environment offers exceptional preservation with sites occurring on the surface and known to be detectable in satellite images.
“First, the arid Atacama Desert environment offers exceptional preservation with sites occurring on the surface and known to be detectable in satellite images.”
The paper relies on this without support.
The history of reported archaeological investigations in these regions provides the ground control needed to validate satellite survey results.
“Second, the history of reported archaeological investigations in these regions provides the ground control needed to validate satellite survey results.”
The paper relies on this without support.
The lack of dense surface vegetation provides ideal conditions for remote detection.
“Finally, the lack of dense surface vegetation provides ideal conditions for remote detection.”
The paper relies on this without support.
All of these more recent archaeological cultures produced ceremonial, domestic, and funerary architecture visible on the surface and in freely available satellite imagery.
“All of these more recent archaeological cultures produced ceremonial, domestic, and funerary architecture visible on the surface and in freely available satellite imagery.”
The paper relies on this without support.
Agricultural features are frequently re-used over time and by contemporary agro-pastoral populations, making the distinction between modern and archaeological features difficult.
“This is likely because dating such features in the Andes can be problematic, as agricultural features are frequently re-used over time and by contemporary agro-pastoral populations, thus making the distinction between modern and archaeological features difficult.”
The paper relies on this without support.
An undergraduate student picked at random from an entire university population would not perform as well as the undergraduate students who elected to participate in this study.
“We believe it is unlikely, for example, that an undergraduate student picked at random from an entire university population would perform as well as the undergraduate students who elected to participate in this study.”
The paper relies on this without support.
Unstructured searches are more engaging for participants who enjoy the organic discovery and exploration process.
“Last, and potentially most importantly, unstructured searches are more engaging for participants who enjoy the organic discovery and exploration process.”
The paper relies on this without support.
Structured surveys can be tedious, causing surveyors to lose focus and fail to identify sites in their field of view.
“Structured surveys, in contrast, can be tedious causing surveyors to lose focus and fail to identify sites in their field of view.”
The paper relies on this without support.
Unstructured surveys are a common exercise among practicing archaeologists.
“Nonetheless, we acknowledge some of the advantages of unstructured survey along with the reality that it is a common exercise among practicing archaeologists.”
The paper relies on this without support.
The scale and complexity of Andean architecture renders archaeological sites exceptionally visible.
“Moreover, the scale and complexity of Andean architecture renders these sites exceptionally visible.”
The paper relies on this without support.
The risk of bias in unstructured survey in areas preconceived to be productive while limiting search in other areas is a major disadvantage.
“One major disadvantage of unstructured survey is the well-known risk of bias in areas that are preconceived to be productive while limiting search in other areas that may actually be archaeologically interesting.”
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Stated false-positive range excludes table values
MinorTable 2 includes a false-positive proportion of 0.01 (1%) for participant 7 and 0.42 (42%) for participants 5 and 6, outside the stated 4-41% range.
“False positive rates ranged from 4–41% among participants.”
“| 5 | grad | no | no | yes | 19 (0.03) | 14 (0.42) | | 6 | grad | yes | no | no | 92 (0.16) | 66 (0.42) | | 7 | grad | yes | yes | no | 72 (0.13) | 1 (0.01) |”
Abstract and results give different discovery-rate ranges
MinorThe abstract gives a 0-18% range for the 9.3% discovery rate, while the results give 1-18% for that same modeled rate.
“Results indicate an average site discovery rate of 9.3% (0–18%, 95% range).”
“From the modeled Beta distribution, we derive a mean discovery rate of 9.3% (1–18%).”
False-positive significance summary omits two significant variables
MinorThe prose lists three variables with significant effects at p<0.1, but Table 3 also gives GIS experience p=0.07 and pedestrian survey experience p<0.01 for the false-positive model.
“the following variables exert a significant (p<0.1) effect on false positive rates: educational experience, experience in Andean archaeology, and the amount of time spent surveying.”
“| False positive | Education | -1.04 | -0.24 | p < 0.01 | | | Regional Specialty | 1.36 | 0.20 | p < 0.01 | | | GIS experience | 0.61 | 0.34 | 0.07 | | | Pedestrian survey experience | -0.61 | 0.20 | p < 0.01 |”
Omo significance claim conflicts with Table 3
MinorThe prose assigns p<0.01 to the higher discovery frequency of Omo and Tumilaca sites, but Table 3 reports p=0.03 for Omo.
“Omo and Tumilaca associated sites in the Middle Moquegua Valley were discovered at a higher frequency than the Chiribaya, Chen Chen, Estuquiña, Huaracane, and Inka sites (p < 0.01).”
“| | Omo | 1.93 | 0.90 | 0.03 |”
Figure captions give incorrect experience-group sizes
MinorThe captions give pedestrian-survey group sizes of 4 and 8 and GIS group sizes of 7 and 5, whereas the results state that seven participants had pedestrian-survey experience and four had GIS experience. Both Fig 2 and Fig 3 repeat the caption counts.
“C: Experience in pedestrian survey (n = 4,8), D: Experience with GIS (n = 7,5)”
“seven had experience with pedestrian survey, and four had experience working with GIS in a research capacity.”
Citations
Cites
R: A Language and Environment for Statistical Computing
R Core Team (2000) · doi:10.32614/r.manuals · Cited by 355,817
Fitting Linear Mixed-Effects Models Using lme4
Bates, D., Mächler, M., Bolker, B., & Walker, S. (2015) · Journal of Statistical Software · doi:10.18637/jss.v067.i01 · Cited by 86,835
Continuous Univariate Distributions.
Omori, Y., Johnson, N. L., Kotz, S., & Balakrishnan, N. (1995) · Journal of the American Statistical Association · doi:10.2307/2291547 · Cited by 9,085
Evaluating significance in linear mixed-effects models in R
Luke, S. G. (2017) · Behavior Research Methods · doi:10.3758/s13428-016-0809-y · Cited by 1,704
Method and Theory in American, Archaeology
Campbell, T. N. (1959) · Hispanic American Historical Review · doi:10.1215/00182168-39.2.309 · Cited by 551
Correcting temporal frequency distributions for taphonomic bias
Surovell, T. A., Finley, J. B., Smith, G. M., Brantingham, P. J., & Kelly, R. L. (2009) · Journal of Archaeological Science · doi:10.1016/j.jas.2009.03.029 · Cited by 377
Quebrada Jaguay: Early South American Maritime Adaptations
Sandweiss, D. H., McInnis, H. E., Burger, R. L., Cano, A., Ojeda, B., Paredes, R., Sandweiss, M. D. C., & Glascock, M. D. (1998) · Science · doi:10.1126/science.281.5384.1830 · Cited by 338
Satellite Remote Sensing for Archaeology
Parcak, S. H. (2009) · doi:10.4324/9780203881460 · Cited by 255
Detection of archaeological crop marks by using satellite QuickBird multispectral imagery
Lasaponara, R., & Masini, N. (2006) · Journal of Archaeological Science · doi:10.1016/j.jas.2006.04.014 · Cited by 211
Monumental architecture at Aguada Fénix and the rise of Maya civilization
Inomata, T., Triadan, D., López, V. A. V., Fernández-Diaz, J. C., Omori, T., Bauer, M. B. M., Hernández, M. G., Beach, T., Cagnato, C., Aoyama, K., et al. (2020) · Nature · doi:10.1038/s41586-020-2343-4 · Cited by 207
Knudson, K. J., & Price, T. D. (2006) · American Journal of Physical Anthropology · doi:10.1002/ajpa.20480 · Cited by 173
Orengo, H. A., Conesa, F. C., Garcia-Molsosa, A., Lobo, A., Green, A. S., Madella, M., & Petrie, C. A. (2020) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2005583117 · Cited by 165
Mapping patterns of long-term settlement in Northern Mesopotamia at a large scale
Menze, B. H., & Ur, J. A. (2012) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1115472109 · Cited by 139
Lidar reveals pre-Hispanic low-density urbanism in the Bolivian Amazon
Prümers, H., Betancourt, C. J., Iriarte, J., Robinson, M., & Schaich, M. (2022) · Nature · doi:10.1038/s41586-022-04780-4 · Cited by 124
Andean Diaspora: The Tiwanaku Colonies and the Origins of South American Empire
Goldstein, P. S. (2006) · The Journal of Latin American and Caribbean Anthropology · doi:10.1525/jlat.2006.11.2.507 · Cited by 116
Patterns of looting in southern Iraq
Stone, E. (2008) · Antiquity · doi:10.1017/s0003598x00096496 · Cited by 112
Satellite imagery-based monitoring of archaeological site damage in the Syrian civil war
Casana, J., & Laugier, E. J. (2017) · PLOS ONE · doi:10.1371/journal.pone.0188589 · Cited by 110
Satellite Imagery-Based Analysis of Archaeological Looting in Syria
Casana, J. (2015) · Near Eastern Archaeology · doi:10.5615/neareastarch.78.3.0142 · Cited by 107
Rayne, L., Bradbury, J., Mattingly, D. J., Philip, G., Bewley, R., & Wilson, A. I. (2017) · Geosciences · doi:10.3390/geosciences7040100 · Cited by 105
Satellite evidence of archaeological site looting in Egypt: 2002–2013
Parcak, S. H., Gathings, D., Childs, C., Mumford, G. D., & Cline, E. H. (2016) · Antiquity · doi:10.15184/aqy.2016.1 · Cited by 103
LOOTING AND THE WORLD'S ARCHAEOLOGICAL HERITAGE: The Inadequate Response
Brodie, N., & Renfrew, C. (2005) · Annual Review of Anthropology · doi:10.1146/annurev.anthro.34.081804.120551 · Cited by 98
Davis, D. S. (2019) · Archaeological Prospection · doi:10.1002/arp.1730 · Cited by 92
Casana, J. (2020) · Journal of Field Archaeology · doi:10.1080/00934690.2020.1713285 · Cited by 90
Sarris, A., Papadopoulos, N., Agapiou, A., Salvi, M. C., Hadjimitsis, D. G., Parkinson, W. A., Yerkes, R. W., Gyucha, A., & Duffy, P. R. (2013) · Journal of Archaeological Science · doi:10.1016/j.jas.2012.11.001 · Cited by 88
Detection of Ancient Settlement Mounds
Menze, B. H., Ur, J. A., & Sherratt, A. (2006) · Photogrammetric Engineering & Remote Sensing · doi:10.14358/pers.72.3.321 · Cited by 85
Contreras, D. A., & Brodie, N. (2010) · Journal of Field Archaeology · doi:10.1179/009346910x12707320296838 · Cited by 84
Owen, B. D. (2005) · Latin American Antiquity · doi:10.2307/30042486 · Cited by 84
Lasaponara, R., Leucci, G., Masini, N., & Persico, R. (2013) · Journal of Archaeological Science · doi:10.1016/j.jas.2013.10.032 · Cited by 78
Casana, J., & Cothren, J. D. (2012) · SpringerBriefs in archaeology. Archaeological heritage management/SpringerBriefs in archaeology/Springer briefs in archaeological heritage management · doi:10.1007/978-1-4614-6074-9_4 · Cited by 63
Huaquerosand remote sensing imagery: assessing looting damage in the Virú Valley, Peru
Contreras, D. A. (2010) · Antiquity · doi:10.1017/s0003598x0006676x · Cited by 60
Casana, J. (2014) · Advances in Archaeological Practice · doi:10.7183/2326-3768.2.3.222 · Cited by 60
Risbøl, O., Bollandsås, O. M., Nesbakken, A., Ørka, H. O., Næsset, E., & Gobakken, T. (2013) · Journal of Archaeological Science · doi:10.1016/j.jas.2013.07.002 · Cited by 58
Landscape and Environment: Insights from the Prehispanic Central Andes
Contreras, D. A. (2010) · Journal of Archaeological Research · doi:10.1007/s10814-010-9038-6 · Cited by 50
Architecture and Power on the Wari-Tiwanaku Frontier
Nash, D. J., & Williams, P. R. (2004) · Archeological Papers of the American Anthropological Association · doi:10.1525/ap3a.2004.14.151 · Cited by 50
Rethinking the Landscape: Emerging Approaches to Archaeological Remote Sensing
Casana, J. (2021) · Annual Review of Anthropology · doi:10.1146/annurev-anthro-101819-110344 · Cited by 45
The Efficacy and Analytical Importance of Manual Feature Extraction Using Lidar Datasets
Quintus, S., Day, S. S., & Smith, N. J. (2017) · Advances in Archaeological Practice · doi:10.1017/aap.2017.13 · Cited by 43
Big Archaeology: Horizons and Blindspots
VanValkenburgh, P., & Dufton, J. A. (2020) · Journal of Field Archaeology · doi:10.1080/00934690.2020.1714307 · Cited by 39
Lasaponara, R., & Masini, N. (2013) · International Journal of Applied Earth Observation and Geoinformation · doi:10.1016/j.jag.2013.09.006 · Cited by 37
Crowdsourcing the Unknown: The Satellite Search for Genghis Khan
Lin, A. Y.-M., Huynh, A., Lanckriet, G., & Barrington, L. (2014) · PLoS ONE · doi:10.1371/journal.pone.0114046 · Cited by 35
Goldstein, P. S., & Magilligan, F. J. (2010) · CATENA · doi:10.1016/j.catena.2010.11.001 · Cited by 32
Karamitrou, A., Sturt, F., Bogiatzis, P., & Beresford-Jones, D. (2022) · Surface Topography: Metrology and Properties · doi:10.1088/2051-672x/ac9492 · Cited by 29
Detection of Maya Ruins by LiDAR: Applications, Case Study, and Issues
Chase, A. F., & Chase, D. Z. (2017) · Geotechnologies and the environment · doi:10.1007/978-3-319-50518-3_22 · Cited by 22
GIS, Remote Sensing, and Landscape Archaeology
Parcak, S. H. (2017) · Oxford University Press eBooks · doi:10.1093/oxfordhb/9780199935413.013.11 · Cited by 22
A Pilot Study on Remote Sensing and Citizen Science for Archaeological Prospection
Stewart, C., Labrèche, G., & González, D. L. (2020) · Remote Sensing · doi:10.3390/rs12172795 · Cited by 22
Wernke, S. A., VanValkenburgh, P., & Saito, A. (2020) · Journal of Field Archaeology · doi:10.1080/00934690.2020.1713286 · Cited by 21
Wernke, S. A., Adams, J. A., & Hooten, E. R. (2014) · Advances in Archaeological Practice · doi:10.7183/2326-3768.2.3.147 · Cited by 19
Sadr, K. (2016) · Archaeological Prospection · doi:10.1002/arp.1515 · Cited by 16
Zaro, G. (2007) · Latin American Antiquity · doi:10.2307/25063102 · Cited by 15
Aldenderfer, M. S. (2003) · American Scientist · doi:10.1511/2003.38.542 · Cited by 15
Masini, N., & Lasaponara, R. (2021) · Journal of Computer Applications in Archaeology · doi:10.5334/jcaa.73 · Cited by 14
Oltean, I. A., & Abell, L. L. (2011) · Remote sensing and digital image processing · doi:10.1007/978-90-481-8801-7_13 · Cited by 10
Virtual survey on North Mesopotamian tell sites by means of satellite remote sensing
Menze, B. H., Mühl, S., & Sherratt, A. (2007) · Cited by 9
A late pre-Hispanic ceramic chronology for the upper Moquegua Valley, Peru / Charles Stanish.
Stanish, C. (1991) · doi:10.5962/bhl.title.5180 · Cited by 4
DOCUMENTATION OF HERITAGE STRUCTURES THROUGH GEO-CROWDSOURCING AND WEB-MAPPING
Dhonju, H. K., Xiao, W., Shakya, B., Mills, J. P., & Sarhosis, V. (2017) · The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences · doi:10.5194/isprs-archives-xlii-2-w7-17-2017
Andean Diaspora: The Tiwanaku Colonies and the Origins of South American Empire
Goldstein, P. S. (2006) · Journal of Latin American Anthropology · doi:10.1525/jlca.2006.11.2.507
Flagged first, then by citations
Research works printed but not linked: 4
- S. Parcak, 2019, Archaeology from space: How the future shapes our past
- QGIS Development Team, 2021, QGIS Geographic Information System
- SI Baitzel, 2019, Presencia humana, patrones de asentamientos prehispánicos y complementariedad ecológica en las lomas del Valle de Sama, Tacna, Perú, Chungará (Arica), 51, 381
- A Kunzetsova, 2017, lmerTest package: Tests in linear mixed effect models, J Stat Softw, 82, 1
Cited by
Sorted by citations
Landauer, J., Klassen, S., Wijker, A. P., van der Kroon, J., Jaszkowski, A., & Verschoof, W. B. (2025) · PLoS ONE · doi:10.1371/journal.pone.0320452 · Cited by 9
Eidshaug, J. S. P., Bjerck, H. B., Breivik, H. M., Risbøl, O., Tívoli, A. M., & Zangrando, A. F. (2024) · The Journal of Island and Coastal Archaeology · doi:10.1080/15564894.2024.2426828 · Cited by 7
Bourgeois, Q. P. J., Kaptijn, E., Verschoof, W. B., & Lambers, K. (2024) · Antiquity · doi:10.15184/aqy.2024.127 · Cited by 5
Oyaneder, A. (2025) · Antiquity · doi:10.15184/aqy.2025.10213 · Cited by 3
Ure, J. (2026) · Springer eBooks · doi:10.1007/978-3-032-07619-9_6 · Cited by 1
Archaeological sites in conflict zones: an in-depth case study of Tell Dothan, Palestine
Al-Houdalieh, S. H. (2026) · Contemporary Levant · doi:10.1080/20581831.2026.2628423 · Cited by 1
Snyder, T. J., & Haas, R. (2025) · PLoS ONE · doi:10.1371/journal.pone.0328409
Self-citationMesía-Montenegro, C., & Sánchez-Borjas, A. (2026) · Journal of Cultural Heritage · doi:10.1016/j.culher.2026.02.004
Metadata
- Authors
- Snyder, T. J., & Haas, R.
- Affiliations
- University of California, Davis (US)
- University of Wyoming (US)
- Corresponding author
- Snyder, Thomas J.
- Venue
- PLOS One
- Publisher
- Public Library of Science (PLoS)
- Published
- 2024-02-06
- Volume
- 19
- Issue
- 2
- Pages
- e0292272
- ISSN
- 1932-6203 (ISSN-L)
- PMC
- PMC10846729
- OpenAlex
- W4391578295
- Cited by
- 8
- References
- 52
- Citations by year
- 2026: 3 · 2025: 3 · 2024: 2
- FWCI
- 7.79
- Type
- journal article
- Language
- en
- Update notices
- correction (2025-07-15)
- Open access
- gold
- doi.org
- Licence
- Locations
- PLoS ONE (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Topics
- Social Sciences › Archeology › Archaeological Research and Protection
- Social Sciences › Arts and Humanities › Archeology (arts and humanities) › Archaeology and Cultural Heritage
- Physical Sciences › Earth and Planetary Sciences › Geophysics › Geophysical Methods and Applications
- Keywords
- satellite remote sensing, archaeological site detection, Moquegua Valley, coastal valleys, Andean coast
- MeSH terms
- Archaeology, Artificial Intelligence, Humans, Peru
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
A language model drew the claims from the paper on 2026-11-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-11-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
