Dataset of Land Reclamation in the Continental United States from the 10th to 18th Centuries

Zhao, C. S. ^{1, 2}  He, F. N. ^1*  Yang, F. ³  Wang, Y. F. ^{1, 2}

1. Key Laboratory of Land Surface Pattern and Simulation, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China;

2. University of Chinese Academy of Sciences, Beijing 100049, China;

3. Key Research Institute of Yellow River Civilization and Sustainable Development & Collaborative Innovation Center on Yellow River Civilization jointly built by Henan Province and Ministry of Education, Henan University, Kaifeng 475001, China

Abstract: Reconstruction of historical land-use and land-cover change (LUCC) at the regional level can provide not only the reliable basic data for studying climate and ecological effects, but also a reference for enriching and improving global datasets. The research area for this study is the modern continental United States. The number of Indians dominated by agriculture, per capita cropland area of Indians, number of non-Indians, and per capita cropland area of each colony were calculated based on historical documents and prior findings. The cropland amount for each region of the continental United States from the 10th to 18th centuries has been reconstructed. The land suitability for cultivation model and cropland allocation model were established. Finally, the geographical distribution pattern of cropland in the continental United States from the 10th to 18th centuries was reconstructed. The findings are as follows: (1) the amount of cropland in the continental United States shows a fluctuating growth trend, increasing from 1.71×10³ km² in 1000 to 4.74×10⁴ km² in 1780. It can be divided into three periods: slow growth (1000–1500), slow decrease (1500–1700), and rapid increase (1700–1780). (2) During the Indian period, the cropland was primarily distributed in the southwest and the area of the eastern region adjacent to Plain. The cropland was distributed in the eastern coastline area during the colonial period.

Keywords: Historical land use; land reclamation; continental United States; 10th to 18th centuries

DOI: https://doi.org/10.3974/geodp.2022.02.17

CSTR: https://cstr.escience.org.cn/CSTR:20146.14.2022.02.17

Dataset Availability Statement:

The dataset supporting this paper was published and is accessible through the Digital Journal of Global Change Data Repository at: https://doi.org/10.3974/geodb.2022.02.04.V1 or https://cstr.escience.org.cn/CSTR:20146.11.2022.02.04.V1.

1 Introduction

Human-caused land-use and land-cover change (LUCC) not only has a direct influence on surface landscape pattern, but it also has a substantial impact on climate and environmental change via bio-geophysical and biogeochemical processes^[1–4]. As a result, LUCC and its impact studies have been critical in global environmental change. As an important part of LUCC study, historical LUCC became one of the fundamental issues of global change research. Quantitative reconstruction of historical LUCC is not only useful to better understand the influence of human activities on the earth’s surface over time, but also provides fundamental data for climate modeling and other studies^[5,6].

The reconstruction of historical LUCC has progressed considerably since the 1990s. The HYDE, SAGE, PJ, and KK10 datasets should be used to obtain the representative research data. In detail, Netherlands Environmental Assessment Agency established the HYDE^[7] global cropland and pasture dataset, with the coverage period from 10000 B.C. to 2015. SAGE cropland dataset covered the period from 1700 to 2007, which was established by Center for Sustainability and the Global Environment^[8]. Pongratz^[9] reconstructed the PJ dataset based on HYDE and SAGE, with the reconstruction period from 800 to 1992. Kaplan^[10] reconstructed the KK10 land cover data from 8000 years ago to 1850 A.D. Although the global datasets covered a long period of time and a wide range, there are still a lot of uncertainties at the regional level^[11–16]. As a result, it is critical to reconstruct the historical land use datasets at the regional level.

The academic community is concerned about the historical LUCC process of the United States, because it is one of the world’s greatest countries with economies and agricultural powers. The temporal and spatial variations of the cropland of the United States have been studied. Waisanen et al.^[17] examined the change features of cropland and its driving mechanisms in the United States from 1850 to 1997 using the statistical data. Based on the statistics, potential vegetation spatial pattern, and soil data, Steyaert et al.^[18] reconstructed the gridding land cover products with a spatial resolution of 20 km in the eastern United States in 1850, 1920, and 1992. Rhemtulla et al.^[19] reconstructed the land use in Wisconsin between the mid-nineteenth and the early-twentieth century. Based on remote sensing and statistical data, Zumkehr et al.^[20] reconstructed the amount and spatial distribution of cropland in the United States from 1850 to 2000. Yu et al.^[21] reconstructed the spatial distribution of cropland in the United States from 1850 to 2016. The findings show that the period of the reconstruction of cropland in the United States mostly after the United States was founded, especially after the middle of the nineteenth century. However, Indians and European colonists had already farmed and lived in the continental United States for a long time before the United States was founded. Furthermore, Indian agriculture had developed throughout the last 1,000 years. However, there are few studies on the reconstruction of cropland in the United States from the 10th to 18th centuries.

As a result, the modern continental United States and the 10th to 18th centuries were selected as the study region and the time period for this study, respectively. By systematically combining the relevant previous research results, we extracted the information about population and land reclamation in the continental United States during the last millennium. Then, we reconstructed the amount of cropland and gridding spatial pattern of cropland in the United States from the 10th to 18th centuries. The reconstruction findings show the regulation of land reclamation change in the continental United States. The results not only provide accurate fundamental data for in-depth studies of historical LUCC’s climatic and ecological effects, but also help to analyze the regional cases for improving the global dataset.

2 Metadata of the Dataset

The metadata of the Dataset^[22] is shown in Table 1.

Table 1  Metadata summary of the Dataset of land reclamation of United States of America during 1000–1780

3 Data Development Method

3.1 Data Sources

There are two types of basic data reported in this article. One is the historical population data used to determine the amount of cropland. Another is the modern base data used for spatially gridded reconstruction of cropland.

3.1.1 Sources of Historical Population Data

(1) Indian period (1000–1620). The population data at 1000 A.D. were quoted from the Historical Population Atlas of the World^[24].

The population of years 1500–1620 was quoted from the document “North American Indian Population Size, 1500 A.D. to 1985”^[25]. In detail, records of Indian population after year 1500 were already available, which was mainly obtained from Mooney’s study^[26]. By filtering and correcting early explorers’ estimations of a mount of tribes, the study estimated the population of North America around year 1500. Then, using Mooney’s work as a foundation, Ubelaker^[25] reconstructed the Indian population of the continental United States from year 1500 to year 1970 by adding and correcting more tribes’ demographic data. Snow^[27] estimated the population of Mohawk using the Mohawk sites. Then, Snow evaluated the accuracy of the estimation of previous population in proportion to the amount of sites in North America. The findings show that most studies overestimated the amount of Indian population. Ubelaker’s estimations are more similar to the population estimates based on site locations. As a result, Ubelaker’s reconstruction was used as the population of Indians after year 1500 in this study.

(2) Colonial period (1620–1780 A.D.). During this time, Europeans began colonizing the American continent and bringing Africans as slaves. The population was divided into Indians and non-Indians. The Indian data sources are the same as the Indian period sources mentioned above. The population of non-Indians is quoted from the United States Department of Commerce, Bureau of National Intelligence “Historical Statistics of the United States: Colonial Period to 1970” Volume 2.

3.1.2 Modern Basic Data

(1) Modern cropland data. We used the land use data with a spatial resolution of 5 km for a total of eight periods in 1982, 1985, 1990, 1995, 2000, 2005, 2010, and 2015. They were obtained from the study by Gong et al.[1]^[28]. The dataset, which is constructed based on the most recent version of GLASS (Global Land Surface Satellite) CDRs (Climate Data Records) from 1982 to 2015, covered cropland, woodland, grassland, scrubland, tundra, moorland, snow, and ice.

(2) Topographic data. The data include the ground height and slope. The United States Geological Survey (USGS) produced and published the Digital Elevation Model (DEM) product (V4.1)[2] collected by the Shuttle Radar Topography Mission (SRTM). The slope data were derived from the DEM data.

(3) Climate data. Active cumulative temperature (1960–1990) data were directly obtained from the Global Agro-ecological Zones (GAZ) product, which was published by the Food and Agriculture Organization (FAO) of the United Nations[3].

(4) Soil data. Soil-related data including coarse-grained matter content, sandy matter content, chalky matter content, and clay matter content of the soil were derived from SoilGrids[4].

3.2 Data Processing and Estimation

3.2.1 Population Data Processing

The political districts have not been set in the continental United States during the Indian and Colonial periods. The continental United States was split into eight regions as described below, based on the regional division scheme used by earlier researchers in historical population-related studies (Figure 1). The eight regions are as follows: the Northwest Coastal region, California region, Plateaus region, Great Basin region, Southwest region, Great Plains region, Northeast region, and Southeast region^[29].

Because the population data for 1000 A.D. was only the total data, it is necessary to divide them into regions. We divided the population data for 1000 A.D. by the proportion of the population in each region in 1500 A.D., because of the less change in the spatial distribution of the population between 1000 and 1500 A.D. Furthermore, from 1620 to 1780 A.D., the amount of colonies continuously expanded. However, it is difficult to obtain the population for each colony. The Bureau of Statistics reported the total popu­lation for the colonial district from 1620 to 1780 A.D., which is credible. As a result, the population of the colonial district was divided into each colony in 1620, 1700, and 1780 A.D., based on the proportion of each colony from the Historical Popula­tion Atlas.

3.2.2 Estimation of Cropland Area

(1) Indian period. It is important to estimate the amount of primarily agr­ic­ultural Indian population beca­use not all the Indians rely on agriculture for their livelihood.

Based on historical documents and previous findings, we first determ­ined the distribution range of the Indians, mostly agriculturalists. Then, we calculated the ratio of the area of agricultural distribution to the total area for each region. The amount of people in each region with agriculture as the main mode of production was calculated by multiplying the ratio and the population of each region. Finally, the cropland area and reclamation rate were calculated by multiplying the amount of per capita cropland with the population.

Because of the lack of historical documents, it is difficult to obtain per capita cropland throughout the Indian period directly from historical documents. It can be assumed that the degree of farming at the time of the colonists’ first arrival in North America was equivalent to that of the Indians, because the Indians originally taught agricultural skills to the colonists^[30]. “A settler could generally only farm 1 to 3 acres each year,” according to the General History of the United States^[30]. Indians had to farm more land to satisfy their own needs because of the poor farming techniques and food production during the primitive agricultural age. The land was sparsely inhabited and fertile, giving the Indians more per capita land to farm than the colonists. Therefore, four acres was used as the per capita cropland to calculate the cropland area and reclamation rate during the Indian period in this study.

(2) Colonial period. The reclamation rate was individually estimated for each region based on the Indian and colonial territories in this study. In detail, the reclamation in Indian region was calculated using the same method as it was for Indian times. Moreover, the cropland distribution was the same as it was in 1620 A.D. The colonial reclamation rate calculation is shown below.

               (1)

where Rcropz(I,t), perland(i,t), ρ(i,t), and Ф denote the reclamation rate, per capita cropland, population density, and proportion of reclamation (constant) for colony i in year t, respectively.

Per capita land area. Because the continental United States was mainly Indian throughout the early colonial period, colonial land development was restricted. Then, to obtain access to fertile land, the British administration began to evict and kill Indians. Moreover, the government implemented a program of “manumission,” in which huge swaths of land were granted to settlers, drawing a substantial number of non-Indians.

In Virginia, for example, a system of land grants was established in 1619, with each settler receiving 50 acres of land. In Maryland, the “manumission” was 100 acres per settler. In New Jersey, the “manumission” was 150 acres per settler. In Georgia, the “manumission” was 500 acres per settler^[31]. Therefore, the land area per capita for the period should be equal to the area set for the “manumission” of each region. The acreage per capita in that period should therefore be equal to the acreage set for the “right of man” in each region.

Reclamation proportion. The colonial authority decreed that three acres should be farmed for every 50 acres of land acquired by each immigrant. In actuality, although certain immigrants’ reclamation proportion may be somewhat greater or lower, it is difficult to obtain the necessary data. Therefore, we used 0.06 as the colonial reclamation proportion, which was calculated by dividing three acres by 50 acres.

3.2.3 Modern Base Data Resampling and Normalization

The geographic resolution was united by resampling each modern data to 10 km×10 km, because the gridded reconstruction of cropland has a resolution of 10 km×10 km. In addition, when the data were dimensionless, two normalization methods were used in this study to characterize the positive and negative connections between the impact factors and geographical distribution of cropland. The formulas are written below:

                    (2)

where M_norm(i,j) and N_norm(i,j) represent the normalized positive and negative correlation factors for grid i in region j, respectively, with the range from [0,1]. M_max(i,j) and N_max(i,j) are the maximum of impact factor values in region i at 10 km×10 km grid size. M(i,j) and N(i,j) are the original values of influencing factors in j grid of region i.

3.3 Grid-based Reconstruction Methods

3.3.1 Determination of the Maximum Range of Available Allocation of Cropland

This study determined the maximum range of available allocation of cropland in the eastern and western regions. The range of maximum available allocation cropland should be included in the modern distribution of cropland in the east, where the population of colonists rapidly increased. We used the 1982, 1985, 1990, 1995, 2000, 2005, 2010, and 2015 eastern United States cropland distribution ranges to create a 10 km×10 km grid data as the maximum range of cropland distribution of the eastern United States through time. The Western Region is the rest of the United States except for the East. The maximum available allocation of cropland was considerably different from that in modern times in the West. Therefore, based on the distribution of agricultural Indians in the West, we determined the maximum range of available allocation of cropland for each time in the West.

3.3.2 Establishment of Land Suitability Model

People’s reclamation activities first occurred in areas that were suitable for crop growth. After being reclaimed, cropland would eventually be expanded from high-quality land to low-quality land as the population increased. Natural and human factors both influenced the distribution of cropland. It is difficult to identify the pattern of population distribution on a grid size in the United States during the 10th to 18th centuries. Moreover, the land regulations were not yet firmly implemented at that time. It was difficult to access the human element. Thus, the main factor influencing the distribution of cropland in this study was natural factors. Human land use activities will be concentrated at lower elevations, on gentle slopes, and in agriculturally appropriate climatic regions. Therefore, factors including height, slope, and climate were selected and standardized in this study. Based on the above calculations, the equivalent weights were utilized to establish a grid-scale land suitability model.

              (3)

where Land_suit(i,j) is the land suitability of grid j in region i. H_norm(i,j), S_norm(i,j), and C_norm(i,j) represent the normalized values of altitude, ground slope, and climate in grid i of region j, respectively.

3.3.3 Establishment of Gridding Allocation Methods of Cropland

Based on the land suitability values calculated using the above model, the cropland area in each region was allocated to the grid. The allocation model can be described as follows.

           (4)

where Cropland(j,t) is the cropland area of grid j in time t; Land_suit(i,j) is the land suitability value of grid i in region j; Area_crop(i,t) is the cropland area of region i in time t; α and β are the index of maximum range of available allocation cropland and soil factors, respectively. If the area of grid j exceeds the maximum range of available allocation cropland or unsuitable for farming, it was given a value of 0 for α and β. Otherwise, it was given a value of 1. The soil factor was determined using a range of parameters for arable soils of Zhang’s work^[32].

3.4 Technical Route

First, historical demographic data and Indian distribution patterns were collected and compiled for eight regions, which were divided from those of continental United States based on the Indian production methods. We determined the amount of Indian agricultural population for each region during the Indian period. Then, we used the population data above and per capita cropland based on relevant data such as settlement history to calculate the amount of cropland during the Indian period. The amount of colonies was estimated by combining the amount of non-Indian people with the history of non-Indian settlement and policies. In the Indian region, the amount of cropland was calculated in the same manner as in the Indian period. A series of cropland amount in each region of the continental United States from the 10th to18th centuries was reconstructed by calculating and integrating the aforementioned data.

Second, the maximum range of cropland distribution was determined based on the remote sensing data and historical documents. A land suitability model and gridding allocation methods for cropland were established.

Finally, we reconstructed gridded cropland data with a 10 km resolution during the 10th to 18th centuries, based on the above methodology (Figure 2).

Figure 2  The technical route of this study

4 Data Results

4.1 Data Composition

The dataset of land reclamation in the continental United States from the 10th to 18th centuries consists of a table file, five .shp files, and five raster data with a resolution of 10 km (Figure 3). Five shp data are shown for the distribution of reclamation rates by district in the United States in the years 1000, 1500, 1620, 1700, and 1780 (Figure 4). The five raster data are shown for the spatial distribution of 10-km-resol­ution cropland in the continental United States for the years 1000, 1500, 1620, 1700, and 1780 (Figure 5).

4.2 Data Results

There is a general increasing trend in the amount of cropland in the continental United States during the 10th to 18th centuries. In detail, the amount of cropland first increased and then decreased during the Indian period. Between the years 1000 and 1500, the continental United States was ethnically Indian-only. Partial Indians relied on farming as the major mode of livelihood. The amount of cropland increased with the increase in agricultural population, from 1.71×10³ km² in year 1000 to 8.53×10³ km² in year 1500. The Spanish did not colonize the continental United States when Columbus found the continent in year 1492, but this resulted in the invasion of European diseases and quick reduction in the Indian population, leading to a decrease of cropland to 6.94×10³ km² in year 1620.

Figure 4  Reclamation rates by region in the continental United States from 1000 to 1780 A.D.

Figure 5  Spatial patterns of cropland in the continental United States from 1000 to 1780 A.D. (10 km×10 km)

The amount of cropland increased during the colonial period, particularly after year 1700. The British began to settle in the continental United States after year 1620. Then, policies related to land settlement were enacted, attracting large-scale immigration activities and expanding the colonies. There were 13 colonies in 1780, including Virginia and Georgia. The amount of cropland rapidly increased as a result, to 4.74×10⁴ km² in 1780, with the increase of 4.83 times between 1700 and 1780.

The northeast, southeast, southwest, and middle Plain regions of the United States were the primary locations of American Indian colonization in 1000 A.D., with the greatest rate of reclamation in the southeast at only 0.07%. The reclamation rate in the continental United States had steady growth from 1000 to 1500 A.D. The reclamation rates increased by 0.45%, 0.24%, and 0.14% in the southwest, southeast, and northeast regions, respectively. The reduction of Indians population resulted in the decrease of the reclamation rate from 1500 to 1620 A.D. The southwest and southeast regions had the largest decrease with 0.07%.

The amount of colonists gradually increased from 1620. Unlike the Indians, who farmed for their own survival, the colonists colonized the country to profit from the sale of crops and tobacco. As a result of the “right of man,” the settler population rapidly increased, as did the reclamation rate in the colonies. In 1780, the reclamation rate increased in the eastern colonies. The reclamation rates of New Jersey, Rhode Island, Maryland, Connecticut, and Massachusetts all exceeded by 20%.

The spatial grid of cropland during the 10th to 18th centuries reveals that cropland was primarily located in the south-west and south-east parts with a small distribution and reclamation rate. At year 1500, the cropland had expanded. The rate of settlement in the south-eastern and southern parts of the plain decreased in 1620. In 1700, the amount of cropland in the southwest decreased. However, in the northeast part, the amount of cropland near the shore expanded. In 1780, the cropland area along the eastern coast quickly expanded.

5 Discussion and Conclusion

5.1 Discussion

Because the PJ and HYDE datasets are the worldwide datasets that provide cropland data in the continental United States from the 10th to 18th centuries, the total cropland area in the continental United States of this study was compared to that of the global dataset PJ and HYDE3.2 (Figure 6). The trend of cropland area change during the 10th to 18th centuries in this study is similar to that of the HYDE3.2 and PJ datasets, with an overall increasing tendency. The PJ dataset, in detail, can be separated into two periods during the 10th to 18th centuries, a period of relative stability from 1000 to 1600, and a period of linear development from 1600 to 1780. In contrast, the cropland data of both this study and HYDE 3.2 can be divided into three periods, which are more consistent with the historical facts: a period of slow growth from 1000 to 1500, a period of decrease from year 1500 to 1700, and a period of rapid growth from 1700 to 1780.

Quantitatively, compared with HYDE3.2, the cropland area of this paper is smaller before 1700 and higher after 1700. The reasons for these distinctions are as follows:

(1) There is a slight difference in the amount of per capita cropland between the two datasets. Because historical documents were used to estimate the per capita cropland in this study, it has not been used in HYDE3.2. The per capita cropland of HYDE3.2 was projected based on modern per capita cropland. For example, 0.031 km²/person and 0.025 km²/person were used as the per capita cropland area in the years 1000 and 1600, respectively. In contrast, the per capita cropland area was estimated in the Indian and non-Indian periods in this study, respectively. Furthermore, the data for per capita cropland area were based on historical documents.

(2) The method for measuring cropland was different between the two datasets. In HYDE 3.2, the total cropland area was calculated by directly multiplying the total population by the per capita cropland area. Different methods for measuring the cropland amount for the Indian and colonial periods were used in this study.

(3) To estimate the amount and distribution of cropland during the Indian period, we considered that agriculture, fishing, gathering, and hunting were all forms of livelihood for the Indians. Thus, we used historical documents to determine the distribution area of the agricultural population. Then, we calculated the amount of agricultural population based on the population density of the area. Finally, the cropland area and reclamation rate of each district were calculated.

(4) We accounted for the difference in per capita cropland area between Indians and non- Indians when calculating the cropland area during the colonial period. Therefore, we calculated the cropland area in the Indian and non-Indian regions. The cropland area in Indian regions were determined in the same manner as in Indian period. The per capita cropland area in each colony was multiplied by the amount of population in that region to obtain the amount of cropland in each colony.

5.2 Conclusion

The amount of cropland in the continental United States during the 10th to 18th centuries was reconstructed in this study based on the statistical data, historical documents, and archaeological data. On this basis, the land reclamation suitability model and cropland allocation model were established. The gridded cropland data in the continental United States during the 10th to 18th centuries was reconstructed, with a spatial resolution of 10 km×10 km. This data showed the characteristics of land reclamation in the continental United States during the 10th to 18th centuries.

(1) The amount of cropland in the continental United States increased steadily, rising from 1.71×10³ km² in year 1000 to 4.74×10⁴ km² in year 1780. It can be divided into three phases: a slowly increasing period (years 1000–1500), a slowly decreasing period (years 1500–1700), and a quickly increasing period (years 1700–1780).

(2) The cropland was primarily located in the south-west and east parts, next to the plains, during Indian periods. The cropland in Indian region was located in the southeast and southwest parts during the colonial period with its area rapidly narrowing. In contrast, the cropland was expanded from the initial north-eastern coastal area to the eastern coastal area in the colonial region during the colonial period.

Author Contributions

Zhao, C. S. and He, F. N. made the overall design for the development of the dataset. Zhao, C. S., Yang, F., and Wang, Y. F. collected and processed the data. Zhao, C. S. and Yang, F. designed the model and algorithms. All the authors jointly wrote and revised the paper.

Conflicts of Interest

The authors declare no conflicts of interest.

References

[1]    Turner, B. L., Meyer, W. B., Skole, D. L. Global land-use/land-cover change towards an integrated study [J]. Ambio, 1994, 23(1): 91–95.

[2]    Pielke, R. A., Marland, G., Betts, R. A., et al. The influence of land-use change and landscape dynamics on the climate system relevance to climate change policy beyond the radiative effect of greenhouse gases [J]. Philosophical Transactions of the Royal Society A Mathematical Physical and Engineering Sciences, 2002, 360(1797): 1705–1719.

[3]    Foley, J. A., Defries, R., Asner, G. P., et al. Global consequences of land use [J]. Science, 2005, 309(5734): 570–574. DOI: 10.1126/science. 1111772

[4]    Arneth, A., Sitch, S., Pongratz, J. Historical carbon dioxide emissions caused by land-use changes are possibly larger than assumed [J]. Nature Geoscience, 2017, 10 (2): 79–84.

[5]    Gaillard, M. J., Morrison, K. D., Madella, M., et al. Past land-use and land-cover change: the challenge of quanitification at the subcontinental to global scales [J]. Past Land Use and Land Cover, 2018, 26(1): 1–44.

[6]    Gaillard, M. J. LandCover6K: global anthropogenic land-cover change and its role in past climate [J]. Past Global Changes Magazine, 2015, 23(1): 38–39.

[7]    Klein Goldewijk, K., Beusen, A., Doelman, J., et al. Anthropogenic land use estimates for the holocene: HYDE3.2 [J]. Earth System Science Data, 2017, 9(2): 927–953.

[8]     Ramankutty, N. Global Cropland and Pasture Data from 1700–2007 [D]. Montreal: McGill University Press, 2012.

[9]    Pongratz, J., Reick, C., Raddatz, T., et al. A reconstruction of global agricultural areas and land cover for the last millennium [J]. Global Biogeochemical Cycles, 2008, 22(3): 1–16.

[10]  Kaplan, J. O., Krumhardt, K. M., Ellis, E. C., et al. Holocene carbon emissions as a result of anthropogenic land cover change [J]. The Holocene, 2011, 21(5): 775–791.

[11]  Fang, X. Q., Zhao, W. Y., Zhang, C. P., et al. Methodology for credibility assessment of historical global LUCC datasets [J]. Science China Earth Sciences, 2020, 50(7): 149–160.

[12]  He, F. N., Li, S. C., Zhang, X. Z., et al. Comparisons of cropland area from multiple datasets over the past 300 years in the traditional cultivated region of China [J]. Journal of Geographical Sciences, 2013, 23(6): 978–990.

[13]  He, F. N., Li, S. C., Yang, F., et al. Evaluating the accuracy of Chinese pasture data in global historical land use datasets [J]. Science China Earth Sciences, 2018, 61(11): 1685–1696.

[14]  Yang, F., He, F. N., Li, M. J., et al. Evaluating the reliability of global historical land use scenarios for forest data in China [J]. Journal of Geographical Sciences, 2020, 30(17): 1083–1094.

[15]   Kaplan, J. O., Krumhardt, K. M., Gaillard, M. J., et al. Constraining the deforestation history of Europe: evaluation of historical land use scenarios with pollen-based land cover reconstructions [J]. Land, 2017, 6(4): 1–20.

[16]  Leite, C. C. , Costa, M. H., Soares-Filho, B. S., et al. Historical land use change and associated carbon emissions in Brazil from 1940 to 1995 [J]. Global Biogeochemical Cycles, 2012, 26(2): 1–13.

[17]  Waisanen, P. J., Bliss, N. B. Changes in population and agricultural land in conterminous United States counties, 1790 to 1997 [J]. Global Biogeochemical Cycles, 2002, 16(4): 84-1–84-19. DOI: 10.1029/2001GB001843.

[18]  Steyaert, L. T., Knox, R. G. Reconstructed historical land cover and biophysical parameters for studies of land -atmosphere interactions within the eastern United States [J]. Journal of Geophysical Research Atmospheres, 2008, 113(D2): 194–204.

[19]   Rhemtulla, J. M., Mladenoff, D. J., Clayton, M. K. Legacies of historical land use on regional forest composition and structure in Wisconsin, USA (mid-1800s-1930s-2000s) [J]. Ecological Applications, 2009, 19(4): 1061–1078.

[20]  Zumkehr, A., Campbell, J. E., et al. Historical U.S. Cropland areas and the potential for bioenergy production on abandoned croplands [J]. Environmental Science and Technology, 2013, 47(8): 3840–3847.

[21]  Yu, Z., Lu, C. Historical cropland expansion and abandonment in the continental U.S. during 1850 to 2016 [J]. Global Ecol Biogeogr, 2018, 27(3): 322–333.

[22]  Zhao, C. S., He, F. N., Yang, F., et al. Dataset of land reclamation of United States of American during 1000–1780 [J/DB/OL]. Digital Journal of Global Change Data Repository, 2022. https://doi.org/10.3974/ geodb.2022.02.04.V1. https://cstr.escience.org.cn/CSTR:20146.11.2022.02.04.V1.

[23]  GCdataPR Editorial Office. GCdataPR data sharing policy [OL]. https://doi.org/10.3974/dp.policy.2014.05 (Updated 2017).

[24]  McEvedy, C., Jones, R. Atlas of World Population History [M]. Penguin Books Ltd., London. 1978.

[25]  Ubelaker, D. H. North American Indian Population Size, A.D. 1500 to 1985 [J]. American Journal of Physical Anthropology, 1988, 77: 289–294.

[26]  Mooney, J. Population [M]. In FW Hodge (ed.): Handbook of American Indians North of Mexico. Bureau of American Ethnology, Bulletin 30, Part 2. Washington DC, 1910.

[27]  Snow, D. R. Microchronology and demographic evidence relating to the size of pre-columbian north american Indian populations [J]. Science, 1995, 268(5217): 1601.

[28]  Liu, H., Gong, P., Wang, J., et al. Annual Dynamics of Global Land Cover and its Long-term Changes from 1982 to 2015 [J]. Earth System Science Data, 2020, 12(2): 1217–1243.

[29]  Alvin, M., Josephy, Jr. 1968. The Indian Heritage of America [M]. Jia, S. H. Trans. Taibei, China: The Commercial Press, Ltd.

[30]  Li, J. M. A General History of the United States, VOL.1, Laying the Foundation of the United States 1585–1775 [M]. Beijing, People’s Publishing House of China, 2002.

[31]  John, T. S. A History of American Farming, 1607–1972 [M]. Gao T, Song P, Zhu R, Trans. Beijing: Agricultural Press of China, 1975.

[32] Zhang, C. P. Development of a gridded allocation algorithm of historical cropland derived from the physiogeographic factors—A case study of China (in Chinese) [D]. Beijing: Beijing Normal University, 2020.