Page 43

3

Land Use Change in Space and Time

The regions of India, China, and the United States described in this volume are each unique, as all places are.1 But do they have common discernible patterns of land use change that may provide insights into the dynamics of how people and places interact? The Tri-Academy Project examined and compared the timelines, patterns, and associated attributes of land use change in six study regions—Kerala, India; Haryana, India; Jitai Basin, China; Pearl River Delta, China; South Florida, USA; and Chicago, USA. In doing so, project researchers achieved a closer look at the intertwined fates of the forestland, grassland, wetland, agricultural land, and built-up land in those regions. The transformation from less intensively modified forests, grassland, and wetland, to land uses with greater human modification, such as agricultural fields and urban settings, is the most usual trajectory but not the only one (Arizpe et al., 1994). A comparison of these and other transformations provides the foundation for the analysis of population/land use dynamics described in this chapter. The goal of this analysis, as with other land use change studies, is to contribute to the integration of the natural and social sciences, the linkage of science to policy, and the development of pathways to sustainable development. (Turner, 1991; Fresco et al., 1997).

1In the following quote substitute “population/land use studies” for “history”: “Cross-national comparative history can undermine two contrary but equally damaging presuppositions—the illusion of total regularity and that of absolute uniqueness. Cross-national history, by acquainting one with what goes on elsewhere, may inspire a critical awareness of what is taken for granted in one's own country, but it also promotes a recognition that similar functions may be performed by differing means” (Frederickson, 1998).



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 43
Page 43 3 Land Use Change in Space and Time The regions of India, China, and the United States described in this volume are each unique, as all places are. 1 But do they have common discernible patterns of land use change that may provide insights into the dynamics of how people and places interact? The Tri-Academy Project examined and compared the timelines, patterns, and associated attributes of land use change in six study regions—Kerala, India; Haryana, India; Jitai Basin, China; Pearl River Delta, China; South Florida, USA; and Chicago, USA. In doing so, project researchers achieved a closer look at the intertwined fates of the forestland, grassland, wetland, agricultural land, and built-up land in those regions. The transformation from less intensively modified forests, grassland, and wetland, to land uses with greater human modification, such as agricultural fields and urban settings, is the most usual trajectory but not the only one (Arizpe et al., 1994). A comparison of these and other transformations provides the foundation for the analysis of population/land use dynamics described in this chapter. The goal of this analysis, as with other land use change studies, is to contribute to the integration of the natural and social sciences, the linkage of science to policy, and the development of pathways to sustainable development. (Turner, 1991; Fresco et al., 1997). 1In the following quote substitute “population/land use studies” for “history”: “Cross-national comparative history can undermine two contrary but equally damaging presuppositions—the illusion of total regularity and that of absolute uniqueness. Cross-national history, by acquainting one with what goes on elsewhere, may inspire a critical awareness of what is taken for granted in one's own country, but it also promotes a recognition that similar functions may be performed by differing means” (Frederickson, 1998).

OCR for page 43
Page 44 Using the common variable data set developed by the Tri-Academy Project, researchers examined the fate of forests, grassland, and wetland in the six study regions and tracked the nature of changes in agricultural areas and cropping patterns and of changes in urban or built-up areas. Among other things, the data allowed them to test the common perceptions that less intensively managed lands such as forests are declining ubiquitously and that built-up areas are rapidly expanding everywhere. The study focused, in particular, on the interrelated effects of population and land use against the backdrop of changing government policy. Population pressures influence land use, and changes in land use may affect social relations. Some types of land use policies also may attract or discourage settlement. Where data were available, this study also explored the environmental and social impacts that may accompany land use changes. LAND USE CLASSIFICATION SCHEME Each country in the Tri-Academy Project has its own land use classification scheme for data collection and planning purposes. Land use has been measured in the study regions by means of a variety of methods, including surveys, statistical reporting, air photogrammetry, and satellite data analysis. Thus the first task was to rationalize the national schemes into a consistent classification that would allow comparison across regions (see Table 3-1). The forest category includes both unmanaged and planted forests; the grassland/wetland category includes rangeland and pastures; the agriculture category includes plantation, grain, and horticultural cropping systems; and the built-up category includes urban areas and land used for transportation and communications. Grassland, rangeland, and wetland were combined because they represent more extensively managed short vegetation types than the plantation, grain, and horticultural crops represented in the agriculture category. The garden land found in the Chinese classification system denotes perennial crops such as orchard crops and tea and is included in the agriculture category of the Tri-Academy classification. TABLE 3-1 Tri-Academy Project Land Use Classification Scheme Class Definition Forest Less intensively managed and planted forests Grassland/wetland Grassland, rangeland, and wetland Agriculture Plantation, grain, and horticultural cropping systems Barren/wasteland Unused Built-up Urban areas, land used for transportation and communications Water Lakes and rivers

OCR for page 43
Page 45 The intent of the Tri-Academy classification scheme was to seek commonalities across the case study regions in terms of functionality given the often sparse and disparate land use data available. More detailed classifications that involve, for example, the quality of forest regrowth in China, are beyond the scope of the project. Although the coarse aggregation of land use categories made formal comparisons difficult, it did allow identification of common and contrasting factors and themes among the regions. The total area and percentages of land in each category over time are shown in Table 3-2. The land use data shown are for geographical regions comprised of groups of districts or counties within provinces or states, except for Haryana, where state-level data are provided. The land use data provided for Kerala are aggregates of that available for the Alleppey, Kottayam, and Idukki districts. Over the course of the study, several issues related to data and land use classifications arose. For example, land used for transportation and communications in China is included in the urban category. In South Florida, the grassland land use category includes wetland as well as rangeland. Another issue was that official data sometimes do not reflect actual conditions. In Kerala, the official data indicated no change over certain decades in forestland, whereas researchers noted a substantial decline. In other sites, particularly in China, a reported increase in the urban population was not necessarily caused by the movement of people but by a redefinition and reclassification of towns and villages. Length of data records varied considerably, but all regions are represented for the period 1975–1995. Finally, the percentage of total area classified according to land use has changed over time in some of the regions, most noticeably in the Jitai Basin. The data in the tables and figures in this chapter are based on the common variable database developed by the study participants. The data may differ slightly from the more detailed information or land use classification schemes used in the case studies. The study regions vary considerably in land area and population ( Figure 3-1). 2 Areas range from 0.965 million hectares for Chicago to 4.421 million hectares for Haryana. In 1950, South Florida had the smallest population (0.76 million) and the state of Kerala had the largest (13.55 million). Forty years later, in 1990, the Jitai Basin had the smallest population (2.35 million), and Kerala was still the most populous of the six regions (29.10 million—see Chapter 2 for a more complete discussion of demographics of the study areas). Because the land areas and total populations of the study regions vary considerably, population density was used to compare land use 2For Kerala, population and area data are given at the state level; land use data are for three districts: Alappuzha (Alleppey), Kottayam, and Idukki.

OCR for page 43
Page 46 TABLE 3-2 Common Variable Database of Land Use Change, Six Study Regions, Various Years Study Region Year Geographic Area (million ha) Forest (%) Grassland/Wetland (%) Agricultural Land (%) Barren/Wasteland (%) Built-up Areas (%) Water Areas (%) Total Area Classified (%) Kerala, 1960 0.813 30.7 .7 62.0 3.7 2.9 .0 100.0 India 1970 0.813 31.2 .5 63.5 1.0 3.8 .0 100.0 1980 0.917 29.4 .3 61.1 2.2 4.0 3.1 100.0 1990 0.871 30.9 .2 60.1 1.9 3.6 3.2 100.0 Haryana, 1970 4.421 2.2 1.2 80.6 4.1 7.0 .0 95.1 India 1980 4.421 3.0 .7 81.5 1.7 8.3 .0 95.2 1990 4.421 3.8 .5 80.9 2.2 7.6 .0 95.0 1995 4.421 2.5 .5 81.5 2.1 9.1 .0 95.7 Jitai Basin, 1965 1.251 41.8 – 17.1 – – 4.0 62.9 China 1981 1.251 33.6 20.1 15.9 20.8 4.2 5.4 100.0 1986 1.251 44.7 15.3 15.8 13.5 5.3 5.3 100.0 1994 1.251 51.6 10.8 15.7 9.7 6.7 5.6 100.0 Pearl River Delta, 1973 1.722 29.6 12.3 38.0 5.0 2.4 12.6 99.9 China 1982 1.722 32.3 3.0 34.6 5.8 7.5 16.8 100.0 1995 1.722 37.0 .2 34.7 .7 15.7 12.8 101.1 South Florida, 1900 2.712 31.9 60.1 .0 .0 .0 2.9 95.0 USA 1953 2.712 22.9 59.7 6.9 .0 2.4 2.8 94.7 1973 2.712 20.7 42.5 20.5 .0 8.1 2.8 94.6 1988 2.712 15.6 39.3 21.4 .4 13.8 9.5 100.0 Chicago, 1900 .965 4.0 – 90.0 – 6.0 – 100.0 USA 1955 .965 4.0 – 84.0 – 12.0 – 100.0 1972 .965 6.3 .9 55.7 2.7 33.0 1.4 100.0 1992 .965 11.6 4.2 47.8 .3 33.5 2.6 100.0 NOTE: Geographic area and land use data for Kerala are for the three districts of Alappuzha (Alleppey), Kottayam, and Idukki.

OCR for page 43
Page 47 ~ enlarge ~ FIGURE 3-1 Population of the six study regions, 1950–1990. changes with population trajectories (see Table 3-3 for the population densities of the six study regions from 1950 to 1990). Over the period 1950–1990, one study region in each country had relatively low population density (Haryana, Jitai Basin, South Florida) and one had relatively high population density (Kerala, Pearl River Delta, and Chicago). South TABLE 3-3 Population Density, Six Study Regions, 1950–1990 (persons per hectare) Study Region 1950 1960 1970 1980 1990 Kerala, India 3.488 4.350 5.495 6.551 7.490 Haryana, India 1.282 1.717 2.346 3.042 3.723 Jitai Basin, China .807 1.015 1.303 1.655 1.879 Pearl River Delta, China 3.293 4.454 4.971 5.959 8.926 South Florida, USA .280 .597 .900 1.324 1.715 Chicago, USA 5.439 6.537 7.355 7.531 7.676

OCR for page 43
Page 48 Florida had the lowest population density over time. In 1950 Chicago had the highest population density, but it was surpassed by the Pearl River Delta in 1990. All regions show increases in population density over the period 1950–1990, but South Florida, Haryana, and the Pearl River Delta had the most dramatic changes over time. For example, in 1950 the Pearl River Delta ranked third in population density; in 1990 it ranked first. PATTERNS OF LAND USE CHANGE Similarities and differences in the patterns of land use are discernible when the land use data are plotted as timelines of change over the period 1900–1995 ( Figure 3-2). This section uses such timelines, data on popula- ~ enlarge ~ FIGURE 3-2 Land use in the six study regions, various time periods. NOTE: “Less intensively managed” category includes forests, grassland/wetland, barren/wasteland, and water.

OCR for page 43
Page 49 tion density, and supporting information from the case studies to address four questions about land use patterns in the study regions: 1. Are less intensively managed lands declining in the study areas? There is a widespread perception that as populations expand, they encroach progressively on less intensively managed lands (forests, grassland, and wetland), and convert them to agricultural and built-up areas. This perception is tested by examining the fates of these land use types in the six study regions. 2. What is the pattern of change in agricultural land? It is commonly believed that the agricultural area of a region first expands to provide food for local population growth and then contracts as urban areas spread. Is this trajectory evident in the study data? 3. What is the dynamic of change for subsistence and market crops? As regions develop, it is often assumed that subsistence crops give way to market crops. If so, how are such changes related to the trajectories of agricultural land change discussed above? What do the study data reveal about increases in crop productivity and their relationship to expansion or contraction of agricultural land area? 4. What has happened to built-up areas? Built-up areas are perceived to be “taking over” less intensively managed and agricultural lands. Is this true in the study regions? What is the rate of change in built-up areas in the six regions? 1. Are Less Intensively Managed Lands Declining? The answer to this question depends on a comparison of trends in forest areas, grassland, and wetland with trends in population density. These land use categories include managed land uses such as agroforestry and pastures, but they may be used to indicate trends in areas that have undergone less-intensive use by humans. Contrary to common perceptions, the recent official data indicated that forested areas in the six study regions have changed minimally or have increased in most areas, even while population density has increased in all areas ( Figure 3-3). More specifically, for the period 1970 to the present, the study's common data set shows that forestland has not declined dramatically and is even increasing in the two study regions of China. After a sharp decline from 1900 to 1950, South Florida shows only a small decline in forestland over the later period. In Kerala, the official published data show little change in forested areas, but other sources show a significant decline. In contrast to forests, grassland and wetland areas are declining in several of the study regions: Jitai Basin, Pearl River Delta, and South Florida ( Figure 3-4). In Kerala, the official data show that the forested area remained constant over the period 1970–1995, at about 30 percent, despite state-level

OCR for page 43
Page 50 ~ enlarge ~ FIGURE 3-3 Forestland and population density of the six study regions, various time periods. NOTE: For Kerala, population density is at the state level; land use change is for the Alappuzha (Alleppey), Kottayam, and Idukki Districts. population density increasing by more than a third. However, the study participants point out that the land use data may not reflect actual conditions (see Chapter 5). Topographical maps from the early part of the century, LANDSAT satellite images from 1973, and India remote sensing (IRS) images since 1981 indicate a substantial decline in forest vegetation cover over time. Chattopadhyay (1985) estimates that forest vegetation covered 44.4 percent of Kerala in 1905, 27.7 percent in 1965, 17.1 percent in 1973, and 14.7 percent in 1983. Field-level observations also indicate that

OCR for page 43
Page 51 ~ enlarge ~ FIGURE 3-4 Grassland/wetland in the six study regions, various time periods. NOTE: For Kerala, land use change is for the Alappuzha (Alleppey), Kottayam, and Idukki Districts. government incentives and the resulting migration into forested areas resulted in the conversion of forest areas into cropland. In the arid and semiarid state of Haryana, virtually all land was converted to agriculture before the 1970s (see Chapter 6). Forest, which covers only about 3 percent of the land area, actually increased slightly during the study period. Meanwhile, from 1950 to 1990 population density almost tripled. Although the small increase in area under forest cover in the 1980s is an indication of efforts to promote environmental conservation, Haryana's percentage of land under forest cover remains far below the Indian national average of 33 percent. The national Ninth Five-Year Plan (1997–2002) specifies that not less than 2 percent of the land area in each district should be classified as forest cover. In Haryana, major land use classifications were remarkably static over the period 1971–1990, implying that the most important land use changes were modifications rather than conversions.

OCR for page 43
Page 52 In the Jitai Basin region of China, forests have been increasing since the 1980s. The original vegetation was subtropical mixed broadleaf and needleleaf forest, located on low to moderate slopes. Over the past 50 years, much of the original forest was destroyed by human activities and later replaced by secondary woods and shrubs (see Chapter 8). Between 1950 and 1978, a great deal of forest was cut down to serve as fuel for steel production, to clear fields for agriculture, and to provide firewood for domestic purposes. In Taihe County, the forest cover decreased from 48.7 percent in 1956 to 44.4 percent in 1961 to 31.3 percent in 1978, which caused soil erosion and loss of biodiversity. In the early 1980s the distribution of forestland to individual ownership added incentive for cutting. Since the 1980s the central government has regulated forest cutting and encouraged the return of marginal cultivated land to forestland. Thus forestland in the Jitai Basin increased from 0.42 million hectares in 1981, to 0.56 million hectares in 1986, to 0.64 million hectares in 1993, and it now covers more than 50 percent of the total area. The Pearl River Delta has long been exploited for agriculture and settlement because of its level terrain, high land quality, and convenience for transportation (see Chapter 9). Thus in this region deforestation of the original subtropical and tropical vegetation has been greater than in other parts of Guangdong Province. Forests covered only about 30 percent of the Pearl River Delta until very recently. Since 1950 population density has increased from 3.3 persons to 8.9 persons per hectare because of the massive and rapid economic development in the region, particularly since 1980. The forest area now appears to be increasing, however, because of replanting policies. The reported percentage of grassland/wetland area has declined from over 10 percent in the early 1970s to nearly zero in the 1990s, although this latter percentage appears to be extremely low. In South Florida, longer-term data reveal that forestland has been declining gradually since 1900 as population density in the region has grown from very low levels, 0.3 persons per hectare in 1950, to 1.7 persons per hectare in 1990. Total forest cover declined from 31.9 percent in 1900 to 15.6 percent in 1988. The pine forest that once covered the eastern coastal ridge has disappeared except for one patch preserved within the boundaries of Everglades National Park (see Chapter 10). The land in either grassland or wetlands also has declined—from 60.1 percent in 1900 to 39.2 percent in 1988. The extensive Everglades marsh, built through peat depositions over the past 5,000 years and once covering 12,000 square kilometers, has been reduced by 50 percent in this century to its present 6,000 square kilometers. Total land in the less intensively managed categories of forest and grassland/wetland has dropped dramatically in the region, from 92.1 percent to 54.9 percent since 1900. In the Chicago study region, most of the land to the west is well-watered grassland; the land to the east was timberland, but it was cleared

OCR for page 43
Page 53for farms by early in the nineteenth century (see Chapter 11). The land north of Chicago also was wooded, but by 1900 most of the usable timber had been cut and the lumber business in the region shrank. The cut timberland was converted primarily to farming and grazing. Like in other regions, however, recent data (1955–1992) reveal that forest and grassland/wetland areas in the Chicago region have increased, to over 10 percent and 4 percent, respectively. Restoration of wetland may account for some of latter increase. These increases in less intensively managed land have occurred as population density has leveled off to about 7.5 persons per hectare. In summary, a comparison across regions confirms that forestland had indeed declined up to 1970, but that, according to the official data, forested areas have been stabilizing or even increasing in many of the study regions since that time, even as population density has continued to rise. The current stability in forestland may stem in part from significant declines early in the century, when most expansion for agriculture took place. Some of the Tri-Academy study regions probably have experienced less change in forested areas because of a long history of dense population. A concern, however, is that official data may not reflect actual forested areas in some study regions. The Jitai Basin, where forests are increasing significantly as a result of active government policy intervention encouraging reforestation, illustrates the importance of government intervention in the maintenance and growth of less intensively managed areas. Similarly, in South Florida a dominant force in land use change has been the federal government's efforts to reclaim the wetland environment of the Kissimmee River. Yet in several of the study regions, including the Jitai Basin, Pearl River Delta, and South Florida, grassland and wetland are declining. Thus it appears that grassland and wetland may now be more at risk of conversion than forested areas. 2. What Is the Pattern of Change in Agricultural Land? During the period 1975–1995, when comparable data are available for all six regions, agricultural land in the study regions remained remarkably constant, despite increases in population density in all regions ( Figure 3-5). Even in Haryana, India, which is still in the process of adopting the set of agricultural management techniques known as the Green Revolution, agricultural land cover appears to have reached the limit of expansion and was even beginning to decline slightly in the most recent decade (1980–1990) included in the study (see Chapter 6). Since 1971 agricultural land in Haryana has remained basically static at about 80 percent of total area. As land cover categories, cultivable wasteland and fallow land, other than current fallow, have virtually disappeared in Haryana.

OCR for page 43
Page 54 ~ enlarge ~ FIGURE 3-5 Agricultural land use and population density of the six study regions, various time periods. NOTE: For Kerala, population density is at the state level; land use change is for the Alappuzha (Alleppey), Kottayam, and Idukki Districts. In Kerala, agricultural land accounted for about 60 percent of total area over the period 1960–1990, as represented by the three districts aggregated in the Tri-Academy common variable data set. This steady percentage of agricultural land is similar to that throughout the state of Kerala over the same period, indicating that spatial expansion of agricultural area has probably reached its limit despite increases in population density. In the Pearl River Delta, farmland declined by over 10 percent in the decade between the early 1970s and the early 1980s. As the special economic zones were created, industrial and commercial development came

OCR for page 43
Page 55 to an area that had been one of the prime “rice bowls” of China. Vast amounts of highly productive cropland were lost as population density increased rapidly. Since the early 1980s, agricultural land overall has remained static because of the increasing cultivation, on the more marginal upland sites, of horticultural crops for consumption in the cities of Shenzhen and Guangzhou. Agricultural land use in the Jitai Basin has been steady at about 16 percent of total land area since 1965. Chicago and South Florida, where changes in agricultural land are the most striking, demonstrate two contrasting land use-population density trajectories. The Chicago study covers the longest period, beginning with the almost complete conversion to agriculture by the 1870s and ending with the virtual disappearance of farmland within the city's borders and in the surrounding suburbs by the present day. In this case, the most rapid declines in agricultural land, which occurred from the 1950s to the 1970s, were accompanied by the greatest increases in population density. Agriculture continues to play a role in the larger region, however, because of the highly fertile soils. In 1992, 48 percent of the total area was still in farms, producing corn for feedlots elsewhere in the Midwest. By contrast, in South Florida increases in agricultural land and population density have gone hand in hand. Agricultural land grew from nearly zero in 1900 to about 20 percent by the early 1970s as the Everglades wetlands were gradually drained for sugarcane and other crop production. Population density tripled from 0.3 to 0.9 persons per hectare from 1950 to 1970. Since the early 1970s, the percentage of agricultural land has remained stable, and population density has continued to grow, reaching 1.7 persons per hectare in 1990. 3. What Is the Dynamic of Change for Subsistence and Market Crops? In general, subsistence crop areas have decreased over time, and areas sown to market crops have increased in most of the study regions. Haryana is a major Green Revolution site that now provides rice and wheat for all of India. Total agricultural area has not changed greatly in Haryana, but the area sown to pulses (edible legumes), which are subsistence crops, decreased by 10 percent in the 1960s and by an additional 12 percent in the 1980s (see Table 6-9 in Chapter 6). The absolute area as well as the proportion of cropped area allocated to traditional crops such as jowar, bajra, maize, barley, and gram also declined. By contrast, the area allocated to Green Revolution crops—rice, wheat, and cotton—increased. Rice and wheat production climbed by 9 percent in the 1960s and by 2 percent in the 1980s. Oilseeds, another market crop in Haryana, declined slightly (by 1 percent) in the 1960s, but increased by 9 percent in the 1980s. Most of these changes have occurred as a result of crop switching as opposed to putting more land under cultivation.

OCR for page 43
Page 56 In Kerala, the land devoted to rice and tapioca (cassava), both subsistence crops, increased by 14 percent and 50 percent, respectively, over the 20-year period from 1956 to 1976, but declined by 50 percent and 57 percent over the most recent two decades (see Table 5-5 in Chapter 5). Market crops in Kerala, represented by coconut and rubber, increased in area—coconut by 50 percent and rubber by 107 percent—between 1957 and 1976 and by 45 percent and 117 percent between 1975 and 1997. Export crops such as pepper, ginger, coffee, cashew, and fruits also have increased in area. The year 1975 is considered to be a major turning point in the cropping pattern, because the area used for rice cultivation reached its peak, followed by a clear shift away from food crops. Sugarcane and tea areas also have declined. These declines have stemmed from the far-reaching land reform of recent decades that reduced the average size of landholdings. The smaller landholdings, combined with the higher skill levels and alternative opportunities for the recently educated population, made paddy cultivation both uneconomical and undesirable. In the Pearl River Delta, farmland area decreased and garden area increased in the 1980s, indicating a shift toward horticulture to supply the rapidly expanding urban areas of Shenzhen and Guangzhou; much less rice is now grown in the Delta. In the Jitai Basin, areas of tea and citrus, high-value market crops, increased dramatically between 1965 and 1987. By contrast, the U.S. study regions, Chicago and South Florida, have continuously and primarily produced market crops, but the kinds of crops grown have changed significantly. The Chicago study region experienced the wheat boom of 1920, a sharp decline in area sown to oats after 1950, and the general ascendancy of corn in recent decades. In South Florida, pineapples were grown in the early decades of the 1900s, but they have now virtually disappeared. Citrus and sugarcane have gained in prominence. 4. What Has Happened to Built-up Areas? India, China, and the United States have very different approaches to structuring development and even to defining built-up areas, making direct comparisons of trends difficult. Given that caveat, however, it appears that changes in built-up areas appear to track changes in population density more closely than do changes in other land use categories ( Figure 3-6). Over the last 30 years, Shenzhen has been transformed from a small fishing village in the Pearl River Delta into a major economic center. As a result, Shenzhen and other newly designated cities are part of a dramatic increase in the built-up area (now approaching 20 percent of total area) of the Pearl River Delta; population density has increased from just below five to almost nine persons per hectare. In the more rural Jitai Basin,

OCR for page 43
Page 57 ~ enlarge ~ FIGURE 3-6 Built-up areas and population density of the six study regions, various time periods. NOTE: For Kerala, population density is at the state level; land use change is for the Alappuzha (Alleppey), Kottayam, and Idukki Districts. urban land has expanded steadily in areas adjacent to existing urban sites, from very little built-up land in 1965 to 6.7 percent in 1994. By contrast, the built-up areas of Haryana and Kerala in India appear to be relatively stable, although there is some potential for encroachment by New Delhi into the Haryana region. Haryana is less urbanized than many developed states of India: only one-fourth of its population resides in urban areas. There have been small increases, however, in areas devoted to nonagricultural uses (housing, industrial estates, and infrastructure such as roads) as a result of increased urbanization, growth in manufacturing, and the government policy of creating development zones. Kerala, characterized by small, well-distributed urban centers rather than one large city, displays a unique rural–urban continuum. Tradition-

OCR for page 43
Page 58ally dispersed linear settlements developed along the ridges and upper slopes of the highland region, and the intervening valleys between the two ridges were used for seasonal agriculture and primary sector activities. Communication networks, such as roads and highways, originally developed along the ridges. As population density and home building grows throughout Kerala, rural–urban distinctions are blurring, even though the expansion of designated urban areas is mostly stagnant or slow. In South Florida, growth in built-up area tracks growth in population density. The amount of built-up land remained relatively low through the end of World War II, but then began to grow rapidly. This development first took place on the Atlantic coast along a 160-kilometer region running north–south, and somewhat later on the western coast along the Gulf of Mexico. Today, over 14 percent of the total area of South Florida is urban. This percentage is significant because approximately two-thirds of the study region is publicly owned land that cannot be developed. Unlike the customary sequence in which less intensively managed land use categories are converted first to agricultural land and then to urban uses, here suburban growth is transforming forestland, grassland, and wetland directly as population density continues to rise. In Chicago, the first major period of urban land use expansion occurred in the early 1900s. The rapid suburbanization that followed World War II dramatically increased the amount of built-up land; population density increased concurrently. Much of the conversion occurred through the loss of agricultural land in areas outside of Cook County. Since the mid-1970s, urban growth within the study area has been essentially stagnant: the proportion of built-up area has held steady at about 33 percent. Growth in population density also has slowed. Thus Chicago might be designated a “mature” urban area. PROCESSES OF CHANGE Findings from the case studies in response to the four questions posed reveal that forested areas generally seem to be steady or even increasing in most sites, while grassland and wetland areas are declining. Government policies play an important role in protecting and enhancing these less intensively managed land uses. Agricultural land areas in most of the study regions have not changed substantially in the last 30 years, although cropping patterns have altered dramatically, moving away from subsistence crops to more marketable crops. In the regions studied, the major transformations in the use of agricultural land occurred in the earlier periods. As subsistence farming is replaced by market-oriented production, food consumption patterns also tend to change. More and more varied food usually becomes available as transportation and infrastruc

OCR for page 43
Page 59ture systems improve with development and as rising incomes permit more variety in diets. Built-up areas, though differently defined, are increasing in almost all the case study regions, but at the expense of grassland and unused land rather than agricultural areas. As these land use changes have been under way, all the study regions have undergone increases in population density. Increases in this single demographic measure do not appear to be strongly linked with change in forest or agricultural land, but the growth in population density does appear to be closely linked to changes in built-up areas as defined in the study regions. At a more subtle yet still significant level, these land use trends also illustrate important shifts in the process of land utilization within the study regions. All sites appear to be at or approaching a steady state in their major land uses. The shifts now occurring appear to be driven in part by market forces and public policies and reflect the processes of land use intensification and changes in cultural and ecological values rather than significant changes in major land use categories. For example, while the amount of agricultural land in the study regions remained relatively stable in area, the land in all regions was used more intensively or for higher-value activities. In the Jitai Basin, land was increasingly utilized for higher-value export crops such as citrus and other fruits. At other sites, such as in the United States, intensification has resulted in significant subregional shifts in land use. Agricultural lands were converted to urban land uses, and more distant lands with native cover were developed agriculturally. All study regions witnessed a decline in land with low-intensity uses, particularly grassland and wetland. Although not a primary focus of this study, this shift has probably led to a drop in wildlife habitat, biodiversity, and natural flood control throughout the six study regions. By contrast, most regions also saw an increase in the amount of forestland. While initially this increase might seem to conflict with the trend of loss in some less intensively managed land uses, it is evident that the forest utility in all cases was explicitly defined with respect to human uses and mandated by policy instruments. Forests are used to control soil erosion, provide fuelwood, and, in limited examples, for recreation. Satellite measurement techniques are being used in several of the study regions for monitoring and analyzing changes in forest areas and other land cover types (Liverman et al., 1998). The shifts in land use processes illustrate the construction of increasingly human-dominated landscapes at each of the study sites, where land value and use are defined largely by human societal needs. This finding is congruent with the trajectory of increasing population density at all sites. The shifts and constructions in land use and the concurrent demographic changes are explored in more detail in the chapters on the individual study regions.

OCR for page 43
Page 60 REFERENCES Arizpe, L., M. P. Stone, and D. C. Majors, eds. 1994 . Population and Environment: Rethinking the Debate. Boulder : Westview Press . Chattopadhyay, S. 1985 . Deforestation in parts of Western Ghats region (Kerala). International Journal of Environmental Management 20: 219–230 . Fredrickson, G. M. 1998 . The Comparative Imagination: On the History of Racism, Nationalism, and Social Movements. Berkeley : University of California Press . Fresco, L., R. Leemans, B. L.Turner II, D. Skole, A. G. vanZeijl-Rozema, and V. Haarmann, eds. 1997 . Land Use and Cover Change (LUCC) Open Science Meeting Proceedings. LUCC Report Series No. 1. International Geosphere-Biosphere Programme and the International Human Dimensions Programme. Institut Cartografic de Catalunya , Barcelona. Liverman, D., E. F. Moran, R. R. Rindfuss, and P. C. Stern, eds. 1998 . People and Pixtels: Linking Remote Sensing and Social Science. Washington, D.C. : National Academy Press . Turner, B. L., II. 1991 . Thoughts on linking the physical and human sciences in the study of global environmental change. Research and Exploration 7(2): 133–135 .