The Bad Earth. Environmental Degradation in Communist China.

Back to the Enviro-Nazi Tactics of the Idiot Left

This article prints out to about 33 pages in its' original size font. I've compressed it a bit but it remains a very long article. It is remarkable in that this subject virtually never is brought up in the mainstream press. The Enviro-Nazis would have you believe that everything is perfect in communist dictatorships and nothing could be further from the truth. Vast areas of ALL of the world's dictatorships are terrible environmental disasters and more occur all the time. These countries which generally are financially bankrupt anyway spend almost nothing to cleanup their disasters and make no efforts whatsoever to prevent extinction of most species except those which are "politically correct" to make the effort to save such as Communist China's Pandas. Read this article if you have some time. It is memorable.

SMIL, V.(1984)Londres: M.E.Sharpe.
(Cap. 2: "Land", pp. 9-77)

LAND

 If there's a mountain, we'll cover it with wheat.

If there's water to be found, we'll use it all to plant rice.

We exhaust our energies in such undertaking,

All in the hope of enjoying a little peace.

Lu Yu, "The Farmer's Lament" (in L. S. Robinson translation)

There was a time when the trees were luxuriant on the Ox Mountain. As it is on the outskirts of a great metropolis, the trees are constantly lopped by axes. Is it any wonder that they are no longer fine?

Meng Ke, Book VI (in D. C. Lau translation)

 

The millennia-old course of environmental degradation-most manifestly represented by large-scale land cover and land use changes caused by deforestation, erosion, desertification, and losses of cropland-not only was not reversed by the Communist regime but all of these destructive trends have actually intensified since 1949. In some instances this has been the result of deliberate, and irrational, policies promoted by the state, such as the massive conversion of forests, grasslands, and lakes into grainfields; in others it has come about as a combination of careless resource expolitation by the state and the desperate actions of poor peasants: much deforestation is in this category.

And although any ranking is not only difficult but perhaps even meaningless, deforestation would appear to be the country's most critical environmental problem. Chinese writings, in the not so distant past so complacently citing huge newly afforested areas, have since 1978 revealed a wealth of truly stunning information about the problem, and their portrayal of the current situation and likely prospects adds up to, it can be stated without exaggeration, a grave environmental crisis of global importance.

 

2.1 Deforestation

Since the mid-1970s a great deal of belated attention, both in scientific and popular writings, has been focused on the growing extent and rapidity of deforestation throughout most of the poor world (see, for example, Eckholm 1976 and Meyers 1980). Unlike the rich countries of Europe and North America where the forests areas have either stabilized or actually increased during the past three decades, most of the Asian, African, and Latin American countries have experienced dangerously advancing deforestation. The continuing disappearance of tropical moist forests, the greatest repository of living mass on this planet, has been especially troubling (Myers 1980).

While numerous nationwide, regional, and local estimates and, less frequently, actual statistics on the decline of forestlands have been available for dozens of poor countries on the three continents, and while Brazilian and Indian deforestation have become almost textbook examples of environmental deterioration, there were no comparable Chinese figures. Indeed, outside China there was a widespread belief that the country was an encouraging exception to the dismal rule (South Korea being the other success story). These impressions were strengthened by the incessant roll of afforestation reports, put out by the Chinese media in general and Xinhua (New China News Agency) in particular, and by the enthusiastic descriptions of advances in tree planting offered by many (not particularly observant and touchingly unsuspicious) Western visitors to China in the early 1970s (Westoby 1979 is a good example).

John K. Galbraith summed it up with a grossly erroneous generalization: "The hills of China, which I had always heard of as being bare, are no longer so" (Galbraith 1973). Unfortunately inaccurate as this statement is, the fact that the Chinese have been engaged in massive afforestation campaigns ever since 1949 seems equally undeniable. Therefore some discussion is called for to show that, contrary to the claims, the extent of productive natural forests has actually declined during the past three decades.

 

2. 1. 1 China's forest resources

To begin with, official Chinese figures, old and new, are contradictory and, as in the case with most other area and volume values in global forestry, not particularly reliable. During the 1950s the published estimates of the area covered by forests ranged between 5 to 10 percent of China's territory (Richardson 1966). In recent Chinese publications shares of forest cover in 1949 are given as low as 5 percent (BR, March 2, 1979, p. 4) and as high as 8.6 percent (for example Yue Ping 1980), a considerable discrepancy. The latest official figure, first published in 1979 (and repeated since), is 12.7 percent, or 122 million ha, and when it was first published it was compared with the 8 percent (about 77 million ha) share in 1949 (Anonymous 1979a).

But the new total does not agree with another official figure, which states that between 1949 and 1979, 28 million ha were newly afforested (Beijing home broadcast, January 14, 1980, SWB 1067); the 1979 total then should be only 105 million ha or, starting with the 8.6 percent share, no more than 110.6 million ha-assuming, of course, that no deforestation took place (of course, starting with the 5 percent coverage figure for 1949 would bring the total to just 76 million ha). The discrepancies do not end here. The new figure of 12.7 percent is claimed to be the result of a "comprehensive check of forestry resources" made originally nationwide between 1974 and 1976 and updated continuously since 1978, and it refers to "fully stocked" productive forests with canopies covering at least 30 percent of the ground (Lin Zi 1980; Hsiung and Johnston 1981).

In 1963, however, officials of the Ministry of Forestry admitted to Richardson (1966) that "nearly 50 percent" of the then estimated 96 million ha was a secondary forest following partial or complete exploitation and hence of low productivity: how those nearly 50 million ha became "fully stocked" just a decade or so later is left unexplained! The latest afforestation claim itself is very dubious: previously the almost identical total of 27 million ha of afforested land was claimed just for the years 1949-1959 (CIA 1976). Moreover, Zhang Anghe and colleagues (1981) state that the wood reserves in the surviving afforested areas are less than 200 million m3, or no more than 2 percent of the national total-although these new plantings account for 23 percent of officially "fully stocked" forests!

Not surprisingly, then, the official nationwide total of forest reserves -9.5 billion m3-reconfirms that Chinese forests are not "fully stocked": it translates to about 78 ^m3/ha, while, for example, the Soviet average is 109, and the U.S. mean is 106 m3/ha, a 35-40 percent difference! A similar situation exists in terms of average annual increments and harvests: in the Northeast, China's most productive forest area, the growth amounts to only 1.5 m3 per ha, and timber output is just 0. 7 m3 per ha, compared with Scandinavian values of, respectively, 3 and 2 m3 per ha (Zhang Anghe and others 198 1). And the total annual increment of 220 million m3 is barely equal to the officially acknowledged cuttings; the real disappearance is, of course, considerably higher.

In a way, all of these discrepancies and uncertainties, though very large, are irrelevant. China may now have some kind of tree cover over nearly 13 percent of its territory, but a good part of it will be less dense (after all, canopy densities for large areas cannot be reliably determined without expensive aerial surveys) and much less productive than the "fully stocked" label implies. Subtracting the poorly stocked secondary growth (at least 46 million ha), the newly planted areas (28 million ha), and the climax forests destroyed during the past three decades (conservatively estimated at about 5 million ha; for details see the next section) from the official 1979 figure leaves about 43 million ha of good, productive natural stands, or about 35 percent of the claimed total. Significantly enough, at the 1979 National Symposium on Forestry Economics it was stated that "only one-third" of China's forests is suitable for commercial logging (Anonymous 1979a).

But even if all the claimed area were a fine productive growth, China's share of forested land, absolute and per capita (0. 12 ha), would be very low, placing the country, respectively, 120th and 121st among 160 nations. Among large populous nations, forests occupy about 18 percent of total land area in India, 50 percent in Indonesia, 30 percent in Nigeria, and 40 percent in Brazil; respective per capita figures for these nations are roughly 0.1, 1.0, 0.5, and 4.0 ha (FAO 1980). In terms of commercial wood resources, China is, with about 9 m3 per capita, 57th among the 75 nations for which this measure is available, and the country's annual wood production translates into a minuscule 0.05 m3 per capita. The United States, with a nearly identical land mass, has 32 percent under forests (1.5 ha per capita, or 12.5 times that in China), and the U.S. per capita use of wood products is more than 40 times the Chinese figure. Clearly, China's poor forest resources put the country at a disadvantage in both environmental and economic terms, and the difficulty is compounded by the extremely uneven distribution of forested land.

The main forested region in China, containing some 60 percent of all timber resources, is in the Northeast: the Da Hinggan and Xiao Hinggan mountains in Nei Monggol; the provinces of Liaoning, Jilin, and Heilongjiang; and the Changbai Shan near the Korean border in Jilin. Heilongjiang alone, with its 16.66 million ha of productive forests (38.5 percent of the province's land), has 13.7 percent of the national total, but it produces nearly half of China's commercial timber (NCNA, November 20, 1980, SWB 1121). These boreal (northern coniferous) forests closely resemble their Siberian, European, and North American counterparts, being composed mainly of slow-growing pines (Pinus), larches (Larix), spruces (Picea), and firs (Abies). These trees may attain grand sizes, making them ideally suitable for railway ties and large construction timber, but their felling age is 40-50 years, even 100 years.

The forests of western Sichuan (in the Daxue Shan and the Shaluli Shan) and western Yunnan (especially near the Burmese and Laotian borders) are China's second largest store of forest biomass. A rich variety of broad-leaved evergreens predominate in the natural stands, and their felling age is usually less than 30 years. China's third largest area of natural forest is in southeastern Xizang, centered in Bomi and Baxoi counties. In this region there are nearly 100 different tree species, including the tropical Dipterocarpaceae in the humid, low-lying Yarlung Zangbo valley.

All three of these heavily wooded regions, containing three-quarters of China's forest resources, are far removed from the densely populated farming and industrialized areas of the North, the Center, and the East, which have been virtually stripped of any important productive forests; North China and the intensively cultivated central plains, with some 320 million people (about a third of China's total), have only 13 million ha of forests (mostly in smaller disjointed patches), or a mere 0.04 ha and one m3 per capita (Lin Zi 1980). And, naturally, forests are also extremely scarce in China's and Northwest, which embraces 37 percent of the country's territory but contains only 3.2 percent of its forest lands.

As the Xizang timber is not yet commercially exploited and the Sichuan-Yunnan resources serve mostly just regional needs, it is necessary to continue large-scale transfers of logs and sawn wood from the Northeast, which constitutes, together with coal transport, the major burden on the overextended North-South railway links. To achieve a better distribution of forest resources and also to improve the country's environment, the Chinese have been engaged for decades in truly massive afforestation campaigns, and if all the staggering claims issued since the early 1950s were taken seriously, the country would not be in the unenviable position it finds itself in today.

There are enough Chinese admissions (and outside observations), however, to conclude that most of the claimed afforestation totals reflect abstract targets, passed downward through the bureaucratic maze, whose fulfillment is reported upward in due time by the lower-level officials to be collated into impressive totals regardless of how well, if at all, the work has been done. Xu Dixin writes in Hongqi about many places that "reported to higher authorities every year the area and the number of trees and, having done this, considered that their task was completed. Few people care about the management and survival rate of the trees" (Xu 1981).

Shoddy planting and the absence of follow-up have been the undoing of China's afforestation campaigns (Figure 2). Undeniably, large numbers of trees have been planted in China since 1949, but Wang Jingcai, an associate professor of forestry at Northwest Agricultural College, quotes a revealing folk saying to explain why some areas have been planting trees everywhere and still have no forests: "Trees everywhere in spring, just half left by summer; no care taken in the fall, all trees gone by winter" (Wang Jingcai 1978).

Survival rates of new plantings have been appallingly low, often much below 10 percent, for several reasons. Careless planting (bareroot saplings put into rocky or sandy soil), no or inadequate follow-up care (no weeding, watering, protective fencing), and lack of scientific approaches to afforestation (choice of species inappropriate for the location or for the soil) have been the key causes. The best available nationwide appraisal of this persistent and extensive waste of resources can be found in a paper by the Policy Research Office of the Ministry of Forestry (1981), which states that of all China's plantings since 1949 6 4no more than one-third managed to survive." And not a few surviving trees have been recklessly destroyed long before maturity in the search for scarce fuel (see section 6.1.3).

But even those critical observers who distrusted the staggering official afforestation claims of the past and who were aware that as in any other poor, populous country appreciable deforestation had been taking place in China during the past three decades were surprised by the magnitude and ubiquity of deforestation reports that have come out of the country since the late 1970s.

 

2.1.2 Deforestation

Recently published accounts of deforestation portray a destruction comparable with the disappearance rates in other of the worst affected forests of the poor world, such as the tropical biomes in Southeast Asia, Central America, and parts of Africa (Myers 1980). Before I summarize the most appalling examples from the Chinese sources since 1979, let me make a few comments about the general causes of the problem.

To begin with, the growing timber demands of a rapidly industrializing society put a damaging stress on China's best commercial forest resources. Official figures for the years 1950-57 show that 1.332 million ha of productive forests were clear-cut, but only 241,977 ha (or 18.2 percent) were regenerated (Ministry of Forestry 1958). Severe damage was done to al I kinds of forests throughout China owing to the large-scale iron-making campaign started in 1958 as a key ingredient of that Maoist economic delusion called "The Great Leap Forward." Much of the accessible timber around tens of thousands of villages and towns was cut to provide charcoal for the primitive "backyard" furnaces producing useless pig iron.

For several years after the inevitable collapse of the "Great Leap," pressure on the forests eased; high lumber production targets were scaled down as a part of general economic readjustment, more attention was given to reforestation, and the demand for charcoal for the remaining small-scale pig iron furnaces dropped sharply. But after only a brief respite came a decade of lunatic policies labeled so inappropriately the "Cultural Revolution." The general lawlessness of its first few years, spilling into localized civil war and prolonged absences of any authority protecting the common good, left a tragic legacy that, combined with repeated policy reversals (during and after the period), punishments for taking initiative, and widespread rural poverty, has been obvious grounds for the destruction of any natural resource for a shortterm gain.

Illegal felling of forest trees, always a problem in wood-short China, became truly rampant, and the worsening rural energy supply (see section 6.1.3) intensified the damage. Trees were also cut, and often sold illegally, for otherwise unobtainable construction lumber or raw material for various manufacturers. Villages in the plains would make such deals with loggers in nearby forested mountain areas on a not inconsiderable scale. By far the most important reason for accelerated deforestation in the late 1960s and during most of the 1970s, however, was another disastrous Maoist policy, "taking grain as the key link." As understandable as the concern is in the world's most populous nation with very limited area of farmland and barely adequate nutrition, the expansion of cereal production could not be achieved in a more damaging, environmentally and economically, and less sustainable manner than by cutting down the forests and planting grain in their place. The Policy Research Office of the Ministry of Forestry (1981) states that the reclamation of forests for grainfields and fires have destroyed at least 6.7 million ha since 1949.

In most places the inevitable vicious circle set in soon after slopes were deforested to make way for grainfields: after a few years, as the accumulated organic matter was sharply reduced and the thin soil rapidly eroded, yields on the newly reclaimed land plummeted and more land was deforested just to maintain the harvests. The abandoned, barren land then succumbed to erosion, often with the irreparable result of all soil being removed to the bedrock. Conversion to grainfields, state-run commercial logging, and illegal private cutting, mainly for fuel, have thus been responsible for the disappearance of China's forests.

As noted earlier, most of China's productive boreal forests are in the northernmost province of Heilongjiang, which supplies nearly 50 percent of the country's timber. Between 1949 and 1978 wood removal in Heilongjiang's forests totaled over 1.1 billion m3, while the new growth amounted to only 600 million m3; in the Da Hinggan mountains, the heart of the province's logging area, nearly 200,000 ha were cut without any reforestation during the years 1964-1978 (Heilongjiang provincial broadcast, March 22, 1979, SWB 1028). In the Yichun area of the province the number of trees felled annually is 2.5 times that of the newly planted ones, and the forests have receded by 100,000 ha in the past two decades (NCNA, October 8, 1979, SWB 1059). A recent survey shows that the province's forests are declining by 1.7 percent annually (Heilongjiang provincial broadcast, November 27, 1980, SWB 1121), and a reforestation survey completed in late 1980 shows that since 1949 the province has cut 2.418 million ha of forest and afforested only 2.073 million ha, a difference of 345,000 ha (Xinhua, February 27, 1981, SWB 1133).

Elsewhere in the Northeast, "regeneration is failing behind lumbering" in Nei Monggol's eastern forest areas (Nei Monggol regional broadcast, March 25, 1981, SWB 1133), and in Jilin "forest resources have gradually diminished during the past few years" because the reforestation rate of cut areas has been a mere 12.7 percent, and also owing to the large number of fires that could not be fought because there is "always a lack of professional firemen and funds" (Xinhua, November 26, 1978, JPRS 72421; Figure 3).

The destruction of the few surviving forests continues in the and North where, simultaneously, an immense planting project (the "Green Great Wall," see section 2.1.4) is under way to reverse the massive erosion and desertification (see sections 2.2 and 2.3). Devastated forests in some provinces and autonomous regions in that area are twice the size of afforested land (Xinhua, June 10, 1979, JPRS 73796), and in some counties nine times; where millions of trees are to be planted before 1985 as a "strategic measure" to control erosion and desertification, animal or tractor-drawn carts can be seen on the roads, loaded with indiscriminately and illegally cut trunks, branches, and roots (Jiang 1979), and on the already heavily eroded Loess Plateau and in the Wei He (Wei River) valley unscrupulous lumbering has not only not ceased, but is actually increasing in some places.

The situation is hardly different in the densely settled farming provinces on the North China Plain were forests survive only on remote. mountain slopes. In Anhui, timber consumption exceeds the natural growth of 400,000 m3, forest resources in many localities have been totally exhausted, "while an insufficient number of trees has been planted in time to replace them" (Anhui provincial broadcast, March 12, 1981, SWB 1133). In Hebei, one half of the mountain slopes (about 33,000 ha) are now barren, and as recently as 1980 a provincial broadcast stated that "unscrupulous lumbering is rampant in quite a number of areas, and the destruction of forests to open up more land is still very serious" (Hebei provincial broadcast, March 6, 1980, SWB 1090).

Perhaps the most disquieting reports on deforestation come from the Chang Jiang basin, and notably from Sichuan (Figure 4), where the decline did not start until the late 1960s but has since accelerated rapidly. What must be termed planned destruction has been one of the leading causes of Sichuan deforestation: in 104 state forest areas of the province 760,000 m3 of timber could be cut annually in a sustainable manner-but every year the quotas fixed by the state departments have been reaching as much as 2 million m3, occasionally even higher (Xu Dixin 198 1).

Logging, clearing of forests for cultivation, expansion of pastures, and forest fires have so seriously upset the ecosystem balance in mountainous Aba, Garzê, and Liangshan prefectures that even the premier of the country has been speaking of the problem since 1980, and some environmentalists fear that the Chang Jiang, whose tributaries drain the prefectures, is in danger of becoming a second Huang He (see section 2.2.4). In total, Sichuan's forests have been reduced by about 30 percent since the 1950s (from 19 to 13.3 percent of the province's territory), but in Aba Prefecture the loss has been a staggering 68 percent; of the province's 139 counties only 12 now have forest covering more than 30 percent of the land, 22 have between 20 and 30 percent, but 91 have less than 10 percent, and 14 counties have below one percent (Wang and Chen 1981)-a very serious handicap for the country's most populous province with 100 million people!

Yunnan still ranks fourth in China with its total timber resources, but in relative terms the province's deforestation has been even more extensive than in Sichuan, and its loss of forest land appears to be by far the worst in China: about 55 percent of Yunnan's territory was covered by forests in the early 1950s, but by 1975 the share dropped to 30 percent; and the annual wood consumption rate of 26 million m3 is double the growth rate (Xinhua, August 20, 1980, JPRS 76376). Yunnan's southernmost region-Xishuangbanna on the Thai border, formerly the country's richest intact tropical moist forest-is being devastated: illegal logging, forest fires, and shifting cultivation have been destroying 3,300 ha a year. Although the area is a protected national reservation, more than 7,000 people have migrated in, establishing 40 villages and cutting the trees for buildings and firewood (Yang 1980). China's other large area of tropical moist forest, Hainan Island (Guangdong Province) in the South China Sea, has suffered even greater ravages. In 1949, 25.7 percent of the island's area (863,000 ha) was covered by tropical forests; today only 7.2 percent (242,000 ha) is left, with timber resources down from 64 million m3 in 1949 to 29 million m3 in 1980 (Zhang Tianxiong 1980; Figure 5). Elsewhere in Guangdong, Lin Xi (1979) reports on the destruction in the north of the province, where most of the standing trees are less than 10 years old, more is logged than planted, and stealing even from the national forests is flagrant. New plantings along the roads fare no better: a peasant complains in a letter to a Guangdong newspaper that one-third of his county's roadside trees were destroyed by breaking off branches, peeling off bark, and digging up roots (Guangdong provincial broadcast, January 28, 1978, JPRS 70685). The Policy Research Office of the Ministry of Forestry (I 98 1) estimates that in just 3 years between 1978 and 1980 nearly 470,000 ha of forest containing more than 9 million m3 of wood were destroyed in the province.

In Zhejiang, forest reserves declined rapidly from 29 million m3 in 1973 to 18 million m3 in 1977; at this rate all forests would be cut in about 7 years (Xu and Qi 1981). China's richest subtropical evergreen forest used to be in the moist maritime province of Fujian: in 1949 timber reserves totaled 178 million m3-now they stand at 89 million m3 (Anonymous 1980a). Forest fires are destroying millions of trees every year: more than 54 million in 1977, more than 30 million in 1979. The province's annual harvesting plans call for cutting 3 million m3, but actually removals are over 20 million m3, and in the second half of the 1970s Fujian was losing almost 270,000 ha of forest every year, a rate at which all mature and maturing forests would be destroyed in just 8 years (Policy Research Office of the Ministry of Forestry 198 1; Zhang, Xiao, and Xu 1980).

A flood of deforestation reports from virtually all other parts of the country has been featured in Chinese newspapers and journals. Even the minuscule remaining forests and groves of tree crops (tung oil, mulberries, bamboo, oranges) have not been spared; they were cut down, the land sown to grain crops, and the yield from these officially nonexistent fields added to the output from long-established fields to boost artificially their yield and overfulfill the planned targets. This socalled "helping field" phenomenon was still widespread in 1979 (Xing 1979), although a greater ecological lunacy is hard to imagine.

Thus China's forests-from the boreal stands of firs and pines in Heilongjiang to the mangroves of Hainan, from the mixed subtropical forests of Sichuan to the bamboo groves of Zhejiang-are rapidly receding. Adding up just the cited provincial deforestation totals -345,000 ha for Heilongjiang (1949-1980), 3.24 million ha for Sichuan (1949-1980), a staggering 10.83 million ha for Yunnan (1949-1980), 470,000 ha for Guangdong (1978-1980), and 1.3 million ha for Fujian (1976-1980)-produces a sum of over 16 million ha. Because this partial nationwide loss figure includes the values for the whole three decades in the three most forested provinces-Heilongjiang, Sichuan, and Yunnan-as well as most of Fujian's loss, the all-China aggregate for the years 1949-1980 would likely be no more than 20 percent higher, or some 20 million ha. Assuming that the 1949 total was 83 million ha (8.6 percent), the country's loss in just three decades amounts to no less than 24 percent, much of it involving the destruction of rich natural climax forests, a loss which cannot possibly be made up by extensive but inadequate and largely monocultural plantings.

China's forestry, farming, and environmental scientists clearly realize the cripplin dangers inherent in the continuation of this destructive trend, and they have been able to impress their arguments on economists and politicians. The top central leadership, political and managerial, and the country's scientific elite are now fully aware of the scope of the past damage and of the urgent need to stop further extensive destruction. The premier of the country pleads for greater understanding and more attention to the problem, the official Communist Party daily, Renmin ribao, runs frequent first page appeals to save "the vital resources of the nation," and a new forestry law went into effect in 1979-but the situation continues to deteriorate.

The causes are various and intractable. The forestry law is a simply unenforceable legalistic document. Its provisions are strict: the illegal felling of trees in state forests is absolutely forbidden, and the guilty party not only has to pay high fines but also has to plant three new trees in place of the destroyed one and is held responsible for the saplings' survival. The regulations also forbid any forest clearing to reclaim farmland and hold the leading forestry officials responsible for losses caused by large forest fires.

The reality is quite different. The Policy Research Office of the Ministry of Forestry (1981) asserts that since the latter half of 1979 forest destruction on a national scale has actually spread, with "hordes of people" illegally cutting and buying wood. The already mentioned legacy of lawlessness, the absence of personal responsibility in a system of collective leadership, unrealistic economic goals, the desperation of poor peasants, and now the understandable wish to live better while the circumstances allow are the deeply intractable causes of deforestation-and of other environmental and societal mals of today's China. A recent description of the situation in Fujian, published in Renmin ribao, is a remarkable example worth recounting in detail (Anonymous 1980a):

All levels of the leadership, preoccupied in the past with implementing the policy of "taking grain as the key link" (i.e., cutting the forests to plant the cereals) now have as their only concern "seizing money before their eyes" (a not so surprising personal interpretation of the current "Four Modernizations" policy which is to make the country rich). As long as everybody's income rises during a cadre's tenure all is now well, and therefore some responsible persons "even lead the masses in large-scale cutting"-surely an ominous portent for the future!

The management of forests is chaotic, with overlapping, competing, uncoordinated bureaucracies trying blindly to fulfill assorted plans or to make quiet money: "One hoe making forests, but several axes cutting them down" goes a new Chinese saying. Past destructive practices continue: forest fires, often intentionally started (!), illegal purchases of lumber by mining enterprises and army units, widespread (of course, also mostly illegal) use of wood as industrial fuel. Two examples from Fujian are again most illustrative: exporting out of the province more than 650,000 m3 of timber outside the state plan in just one three-month period, and burning an equivalent of 230,000 m3 of lumber in only one county with 103 industrial plants and 55,000 inhabitants.

Especially popular throughout Fujian now is the conversion of large lumber pieces into small manufactured items that fetch higher prices (threeor four-fold the price of the original wood). Needless to say, the waste of wood in the process is phenomenal, but some counties have literally hundreds of small processing enterprises engaged in this practice. A decree to stop this waste was issued in June 1978, but it was still being completely ignored as of spring 1980. And the news of indiscriminate cutting of trees in Wuyishan, China's major natural natural preserve (where "no one obeys . . . the forestry ordinance" and the "illegally cut timber is stacked all over the place so that vehicles cannot get through, but no one does anything about it") rounds out a picture that tells more about real concerns and public order in China than all the silky travelogs brought back by the friendship delegations of the past decade.

The destruction of the most productive climax ecosystems on such a scale, moreover, in a nation whose phytomass resources had been in decline for centuries yet whose population has nearly doubled in just 25 years and whose aspirations to modernize are all but modest, has had predictable direct economic and environmental consequences, the tatter bringing, with a delay of just few years or a few decades, additional and stronger economic and social impacts.

2.1.3 Consequences

The direct economic effects of the destruction and misuse of China's forests are everywhere to be seen: pervasive shortages of roundwood, sawn wood, and paper. Coal remains China's principal source of primary energy, and further expansion of its mining is imperative for China's economic survival. Each 1,000 tons of coal extracted in underground mines (and, unlike the United States and the Soviet Union, China has almost no surface coal mining industry) requires 13-25 m3 of softwood pitprops. Each km of railway track requires 200-300 m3 of roundwood to manufacture the ties, which last at best 7 years in the North, much fewer in the South; and the needs for new track are enormous in a nation with mere 50,000 km of railways (the United States, with a virtually identical amount of territory had that much more than 100 years ago). And for each 1,000 m3 of new floor space, 50-70 m3 of timber is consumed (Yue Ping 1980).

Even with considerable recycling, each ton of typical paper products requires about 1.5 tons of new wood, while new paper consumes 3.55.5 m3 per ton. And because China's population will increase by a quarter billion during the next generation, schoolbook needs alone will be staggering; the spread of the massive amounts of technical information without which China's modernization is unthinkable cannot be accomplished without paper in a nation where most people do not have electricity. The timber requirements of the Chinese construction industry have been going up exponentially by 6 percent a year for the past decade, but housing conditions in cities remain unspeakably poor, with an average of 3.6 m2 of floor space per person (Xinhua, June 1, 1980, SWB 1087).

Pent-up demand for just replacing the rundown housing in the villages is even greater, creating enormous potential needs for wood. And as the unprecedented wave of marriages, the legacy of no population controls in the 1960s, rises in the coming years, the young couples will want a roomful of furniture, a sign of prosperity difficult to achieve for tens of millions of newlyweds. And the shortages of fuel wood are already cripplingly acute in a large part of China's countryside (see section 6.1.3).

The current situation is serious. Mine pitprop shortages are chronic, delaying coal extraction; railway ties have to be made of concrete, consuming scarce fossil fuels in the process; and the same is true for transmission poles needed in huge numbers to electrify the countryside; people line up for an overnight wait to buy a single piece of plywood; more than a million sewing machines stand unfinished in Shanghai waiting for little wooden boards; urban newlyweds have to wait another year to get a couple of pieces of simple furniture (examples are from Xinhua, October 10, 1979, SWB 1054). And today's shortages are bound to increase in the future, affecting nearly every aspect of the modernizing economy and the people's aspirations for a life a bit more affluent.

As serious and retarding as these wood shortages are, the problem hardly bears a comparison with the deep environmental effects of China's deforestation: indeed, few of the difficulties facing the nation are as grave in their impact on the society's well-being as the degradative processes set in motion by the destruction of forests. No other ecosystem offers as many irreplaceable services as natural climax forests. Before citing a variety of specific Chinese examples detailing the loss of these benefits, I will briefly review their general nature.

The influences and benefits of forests encompass a wide range of climatic, soil, water flow, and biotic phenomena, and these have been extensively studied in many locations around the world (Kittredge 1948; FAO 1962; Heinsdijk 1975). Climatic effects include moderation of temperature (lowering of the maxima and raising of the minima)-an influence similar to that of large bodies of water and most valuable for crop farming in nearby areas; dissipation of wind and reduction of wind speeds for a distance of more than 40 times the tree height and hence a high effectiveness in combating wind erosion; and high potential evapotranspiration (although rarely any meaningful increase in precipitation).

The most beneficial influence on soils is the outstanding protection trees offer against water erosion. Owing to the interception of rain in the canopies and to the presence of organic matter (litter) on the soil, differences of an order of magnitude between forested and barren surfaces are not uncommon. The rich organic layer on the forest floor also reduces the runoff through increased infiltration and retention.

These abilities are critical for removing water more rapidly from the surface between storms and thus decreasing both the frequency of flooding and its peak flows; in turn this, again, reduces soil erosion and the resulting stream and reservoir silting. And, obviously, the decomposing litter adds significantly to the soil's fertility. Ecosystemically, forests are the richest assemblies of plants and heterotrophs, the irreplaceable repositories of genetic diversity and variability.

The importance of all these always invaluable services is heightened in the world's most populous nation, whose fields are open to strong seasonal winds, whose fluctuating water supply is a key determinant of the harvest, and whose soil erosion problems have been traditionally severe. Accelerated erosion is, undoubtedly, the most widespread and most dangerous consequence of China's rapid deforestation, and as such it will be treated separately in some detail (see section 2.2), as will desertification (2.3). Below are just a few illustrations of some largescale and some local effects of deforestation.

That the aggregate water retention capacity of forests is very large can be easily illustrated by this simple calculation. Assuming that each ha of forest retains 300 m3 of water, the afforestation of all barren hills and mountains in the Chang Jiang basin would result in an effective storage of 20-30 billion m3 of water, twice the capacity of Danjiangkou, China's largest reservoir on the Han Shui (Han River) (Wang and Chen 1981). The removal of tree cover from extensive areas, such as the reductions on the order of 30 or 50 percent in some large southern provinces during just two or three decades, therefore leads inevitably to a markedly decreased water retention capacity and hence to the greater frequency and deeper impact of natural disasters. Chinese scientists believe that the link is unmistakable.

During the years 1950-58, 20 million ha of cropland was annually affected by natural disasters (mainly prolonged droughts or extensive flooding), while between 1972 and 1977 the annual average rose to more than 35 million ha, or a third of all farmland (Yi Zhi 1980). The same source states that in wet South China where two out of three days used to be rainy, now two out of three years are excessively dry. In Heilongjiang, formerly China's most forested province, overcutting and grassland destruction have brought previously unknown serious droughts and duststorms. On the opposite end of the country, in subtropical Yunnan, long-range climatic records show a severe drought once in every 9.6 years between 1470 and 1950-yet since the establishment of the People's Republic of China (PRC) in 1949 until 1978, extreme spring and spring-summer drought occurred every 3.2 years, and the frequency of floods also tripled (Wang and Zhou 1981).

One of China's worst floods in decades was certainly aggravated by widespread deforestation: the Chang Jiang floods in the summer of 1981-an extraordinarily large one in mid-July, the worst inundation since 1877, a smaller one in mid-August-directly affected 135 countries and cities with 11.8 million people, inundating 1.6 million rooms, 2,600 factories, and 830,000 ha of mostly level, fertile farmland (destroying or badly damaging potential harvests of some 1.5 million tons of grain) and damaging 38,000 water conservancy works; direct monetary losses were put at 412.5 billion (Du 1981). This crippling experience led both the central and provincial leadership to search for contributing causes, and a consensus on "painful lessons" quickly emerged: besides uncontrollable natural factors, the loss of 30 percent of the province's forested area, blind conversion of slopeland to grainfields, and improper construction of water conservancy projects were clearly to blame (Anonymous 1981a).

On Hainan Island, the massive destruction of forests has been accompanied by a greater frequency of floods, droughts, and blowing sand and the near extinction of several wildlife species (see section 5. 1). When the villages in the hilly and mountainous regions inunediately west of Beijing were required to become self-sufficient in grain they predictably cleared many slopes for grainfields, causing not only serious water losses and soil erosion but also becoming a source of repeated duststorms blanketing the capital (see section 2.3).

On a smaller scale, the experience of the Dayao Shan forest area (Liuzhou Prefecture, Guangxi) is notable. Two decades of rapid deforestation destroyed 21,300 ha of trees, and an additional 13,300 ha succumbed to uncontrolled fires. Because this forest region is the source of 25 rivers whose waters irrigate nearly 60,000 ha in eight neighboring counties, water shortages and droughts induced by this destruction have been crippling. A 1977 survey found the regular flow of 22 out of the 25 rivers reduced by one-third compared with the years before 1958. And the destruction continues even after the focusing of widespread attention on the problem: peasants in mountain villages, who are urged to protect the forests, have a low grain ration Oust 210 kg compared with over 300 kg in the surrounding farming regions), and because they "are not given enough food to eat, what can they do if they do not choose to reclaim some forest land to grow grain crops?" (Xinhua, October 22, 1980, SWB 1110).

The experience of Qingzhen County on the Guizhou plateau is similar (Wang Ganinei 1981). In 1949 the county had a 30 percent forest cover; by 1975 this was reduced to a mere 4.7 percent. The impacts: 54 percent of the nonirrigated farmland now suffers serious erosion; the average annual rainfall has dropped by 120-155 mm, and prolonged droughts (up to 74 rainless days) have become more common; despite large increases of cultivated areas there has been hardly any growth in grain output, and the degeneration of grasslands reduced sheep and goat counts by nearly 90 percent between 1957 and 1978. Now the reversion to forests has been begun, and by the year 2000 the county should again have a 30 percent tree cover. What an unnecessarily destructive cycle-and over the past three decades how many localities have developed along the same sad path!

2.1.4 Solutions

The conclusion is unequivocal: if the Chinese are not able to reverse the current deforestation trend very soon, they will face within a generation major environmental disasters deeply affecting the very capacity for decent human existence in the worst-off regions and undermining the modernization of the whole nation. What are the measures that must be taken to lead toward recovery? There are many, some traditional, some innovative-all promoted relentlessly and with the utmost commitment. No frenzied mass campaigns similar to the delusive outbursts of the past will do, but participation in the necessarily long-range efforts will have to be massive.

Strange as it might seem in a country with such a relatively minuscule per capita consumption of wood (and despite the chronic shortages!), there is a considerable potential to conserve wood in many ways and hence to relieve some of the overcutting. The Ministry of Forestry estimates that throughout China over 12 million m3 of timber and processed timber remnants are either rotting in the forests or are eventually collected for firewood instead of being turned into useful products (Xinhua, October 10, 1979, SWB 1054). In 1978 a mere 9 percent of timber remnants was used by the artificial board industry or in papermaking, a sharp contrast with the situation in the relatively so much richer in wood United States, where the current rate of wood residue utilization is around 75 percent and rising.

Nationwide sawmill refuse in China is 37 percent of the total timber output (i.e., nearly 20 million m3 a year), but the particle board industry, the best user for this prodigious waste, is so mismanaged (up to 10 different organizations and government departments run a single factory!), outdated, and lacking in investment (less than 8 percent of all investment in forestry has been earmarked in recent years for the industry) that it cannot absorb this precious raw material.

The papermaking industry is no less wasteful. In Fujian all but 3 of the 60 paper mills in the province are using wood as the process fuel, and while the timber consumption for woodpulp came to 400,000 m3 in 1978, the wood burned by the inefficient mills totaled 4 million m3, an amount sufficient to turn out an extra I million tons of paper annually (Xinhua, June 9, 1979, SWB 1037). Paper recycling, another obvious conservation measure, appears to be already fairly extensive in China although it probably doesn't match Japan's nearly 90 percent rate.

Obviously, even the most practicable conservation measures could not stop the deterioration of Chinese forests: higher investment in forestry, much better management of young and middle-aged forests, aggressive replanting of all commercially cut areas, extensive afforestation of barren wastelands, and effective measures against any further conversions of forests into grainfields and against illegal tree cutting are essential to any successful long-range policy. On all of these counts-save the conversion to grainfields, which was long officially sanctioned and promoted-one finds that the past attitudes, decrees, proclamations, laws, and orders issued by the central and provincial governments have been mostly correct, strict, and farsighted-but the actual results have been at best disappointing (e.g., only one-third of all the prodigious afforestation efforts have taken root, there are chronic shortages of fire-fighting equipment), and more often outright appalling (recall the "responsible officials" leading the masses in deforestation drives and the state and army units black-marketing wood).

Official government policy since the founding of the PRC has been that regeneration must follow commercial harvesting to provide for sustained yields, and generally the Ministry of Forestry appears to have favored the forest management, rather than the purely forest utilization, approach. As mentioned earlier, the illegal felling of trees in state or local forests has been strictly forbidden by repeated regulations starting in the 1950s, and in 1979 the State Council issued a new law that not only requires fines and the planting of three trees in place of the one destroyed but also demands the guarantee of the new saplings' survival (Xinhua, January 26, 1979, SWB 1024).

These new regulations also absolutely forbid any forest clearing to reclaim new farmland and call for punishment of leading forestry officials responsible for huge losses caused by major forest fires. A new urgent circular was published on December 5, 1980 (Appendix B), placing distribution of all timber products in forest zones under the unified management of the forest departments, forbidding all other organizations and individuals to fell or to purchase any trees, closing all free markets for lumber and wood products, and forbidding the railways to ship any wood without a forestry administration certificate (NCNA, December 6, 1980, SWB 113).

Past laws have been nearly as strict, yet the rates of deforestation were hardly affected by their issuance. As long as the need for lumber or fuel wood remains acute, as long as quick cash can be made on the sale of wood and wood products, false permits and certificates will be issued, payoffs will be accepted, and the black market will continue to thrive, as past Chinese experience shows all too clearly.

I would suggest that rather than in these punitive and proscriptive measures, necessary as they may be, hope lies elsewhere. Much has been written in the modern sinological literature about the social pressure traditionally operating in Chinese society; when this is reinforced by the coercive nature of the ruling totalitarian regime, one might expect some impressive achievements in the quest for the common good. Not a few Western admirers of Maoist truths subscribed to such ideas, but as the rich, and often tragic, reality of three decades has shown, there is a limit, and a rather low one, to the effectiveness, quality, sustainability, and real success of all such efforts.

Since 1978 there have been many changes clearly indicating that the Chinese understand this and are willing to change some of their longstanding ways of managing basic affairs. Herein lies the hope, because people cannot be forced to grow trees. Foresters promoting new planting schemes all around the poor world have learned this lesson very well indeed (Arnold 1979). The high failure rate of Chinese plantings clearly demonstrates that assuring the saplings'survival is the key to success-and clearly the best way to sustain interest in the tiresome but necessary weeding, watering, thinning, and soil cultivation, as well as the subsequent protection of an established tree before maturity, is to let the people who planted the trees own them.

For long years, of course, this has been one of the clear signs of destructive bourgeois mentality, an unhealthy longing for despised private ownership. This ideological obduracy would be, by itself, almost comic, if it had not brought so much suffering. The best example I know of this is from Shaanxi's Yan'an-Yulin region, a dry and severely eroded loess area (see section 2.2. 1) drastically short of fuel wood, where the private planting of trees was repeatedly prohibited, most recently by a provincial administration order in October 1979, as a clear sign of "promoting capitalism" (Jiang 1979). Yan'an, of course, was Mao's headquarters during the Sino-Japanese War, and three decades later this was the Maoist reward to the wartime revolutionary base: the peasants forced to burn dried animal dung or to dig grass roots so that a backyard willow would not turn them into Wall Street monsters.

The reversal of this irrational policy came, finally, in the spring of 1980, when a joint Chinese Communist Party (CCP) Central Committee-State Council directive issued on March 5 stated, among other things, that all the trees planted by the villagers in a communal effort belong to them rather than to the state, and that the trees the peasants are encouraged to plant in any suitable place, either on their private plots or on land specially allocated to them, will belong to their families. The critical rider: this policy is to remain unchanged for a long time, presumably for decades.

After this directive was issued, one province after another adopted new regulations promoting private woodlots. In Heilongjiang, China's foremost wood producing province, where 2 million out of 3.65 million peasant households are now short of fuel, the new policy is summarized by the slogan "Whoever afforests the land owns the trees." Forestry licenses will be issued to those families whose plantings have a survival rate of 80 percent; households that fail to meet this target will have to try again to get the license. Each household can get 3 mu (0.2 ha), or more if available, of land once it earns the license that guarantees, based on the new measure, the ownership of trees (Heilongjiang provincial broadcast, October 13, 1980, JPRS 76079). The authorities expect that 666,000 ha of new woodlots will be producing in three to five years.

On the opposite extreme of the country in Guizhou, the allocation of hilly land to peasants for fuel wood, timber, or oil-tree lots is similar: between one-fifth and one-third of a ha will be granted to each household in the form of a long-term utilization license enabling the peasants to use or to sell the wood as they wish. This measure is expected to increase the share of the province's treed area from 14 to 20 percent in just four to five years (NCNA, June 21, 1980, SWB 1089).

The combined contribution of private woodlots can be enormous. Assuming that two-thirds of China's 170 million peasant households would set up woodlots averaging 3 mu (this appears to be the typical size, with the extremes ranging from one to 10 mu), the aggregate area would cover nearly 23 million ha-the equivalent of about four-fifths of the total land afforested in successive waves of mass-planting campaigns during three decades since 1949! Yet these plantings could take root and be growing in a matter of just a few years with, as a vicechairman of the State Agricultural Commission stresses, excellent quality and a high survival rate owing to assiduous care, and without any state investment (Zhang Pinghua 1981).

Private woodlots are undoubtedly the best step toward rapid afforestation of hills, mountains, and wastelands adjacent to settlements; they would be a major contribution to a sustainable household energy supply and relieve the pressure on illegal cutting. But they must be augmented by state and collective afforestation on larger scales in principal forest regions as well as by effective measures and incentives against conversion to farmland.

Large-scale afforestation continues to be promoted in a variety of ways. There are extensive plans for reforestation in Heilongjiang; a short-term plan to plant some 700,000 ha of fast-growing, high-yielding trees in the Jiangnan to obtain useful timber in 10-15 years; a grand scheme for a system of shetterbelts in the three northern and areas (see section 2.3); and a nationwide goal of planting nearly 67 million ha (one billion mu) by the year 2000, when China's afforested area should reach 20 percent of the country's territory, with 30 percent being the eventual long-term goal.

The most recent claims of work actually completed indicate a rate of progress more than sufficient to achieve the end-of-the-century goal, but as with so many current encouraging changes, a wait of several years, or even a decade, is necessary to perceive the real trend. While China's environment certainly needs the new huge forest cover planned for the year 2000, 1 have deep doubts about the effective fulfillment of such a task; half of the goal properly achieved would have to be considered an enormous success.

Whatever the actual extent of the afforestation effort, it will rely critically on two very different approaches: modern aerial seeding without which the revegetation of inaccessible mountain slopes would be extremely difficult, and mass-participation "voluntary" drives. Aerial seeding was an important part of earlier large-scale afforestation in 458 counties in 22 provinces: in the two decades before 1979, slightly over 11.3 million ha was seeded from the air with a claimed average survival rate of about 40 percent, accounting for about one-sixth of the total area afforested since 1949 (NCNA, March 12, 1980, SWB 1090). Most of the air-seeded areas are in warm and rainy southern provinces (above all Guangdong, Sichuan, Guizhou, and Guangxi), and of the 4.5 million ha surviving by 1980, grown forests accounted for one-third of the total (1.5 million ha), young trees for about two-fifths, and seedlings for the rest. The air-seeded area is expected to be around 6.7 million ha by 1985, and experiments are to extefid the technique to northern China (NCNA, May 5, 1980, SWB 1083).

Aerial seeding may present considerable technical and environmental challenges, especially in the north, but to me the new "voluntary" tree-planting campaigns are a much more dubious undertaking. Initiated in 1981, they are anything but voluntary: everybody between eleven and sixty years for males and eleven and fifty-five years for females is to take part, and the trees so planted will be collectively owned (see Appendix C). March 12 of every year is the official day to begin the annual plantings, and on its eve in 1982, 10,000 people attended a rally at the Great Hall in Beijing; and the next day nearly all the top Chinese leaders, including Hu Yaobang, Deng Xiaoping and Zhao Ziyang, joined the masses in planting. All of this may be a valuable public relations exercise, but when the city dwellers leave the saplings on the suburban hills (each one is supposed to plant three to five of them) who will take care of them? Old habits die hard: the new private woodlot regulations are as sensible as this continuation of Maoist-type campaigns preceded by exhortation rallies is dubious. To ensure that forest lands will not again be converted to farm fields will require more than exhortations and regulations. Encouragingly, new provincial polices in Yunnan include not only the allocation of some state mountain forests to village and individual care and absolute prohibition of forest conversion to new farmland but also efforts to settle slash-and-burn farmers in permanent areas and, most importantly, the option to exchange forest, livestock, and local products for grain from the state supplies, the arrangement for sufficient grain for food-deficient areas having no products to exchange, and the forbiddance of the establishment of new fields (Xinhua, August 20, 1980, SWB 1108).

The Sichuan provincial government allocated 4 million yuan as rewards for communal efforts to protect the woodlands, especially in mountainous Garzê, Aba, and Liangshan prefectures, which contain four-fifths of Sichuan's forest land (Beijing home broadcast, October 20, 1980, SWB 1110). Also, the strict ban on the conversion of forest to farmland must be complemented by no less strict and effective measures to prevent further losses of arable land (see section 2.5); otherwise, the already difficult task of intensifying farm output will become impossible.

Deforestation surely causes more ecosystemic, economic, and everyday human problems than any other form of environmental degradation in China. The trends of the recent past have been outright frightening-yet the very gravity of the situation presents hope and opportunity. When a nationwide conference on forestry is told that "according to the estimate based on the actual annual rate of reduction, by the end of this century there will be no trees to harvest" (Anonymous 1979a), the assessment is undoubtedly accurate, but it is equally clear that such a reality is truly unthinkable and that the trend cannot be allowed to continue. By a combination of efforts, none of them easy or yielding rapid results, the trend has to be reversed if the very physical foundations of China are to be preserved. And as hundreds of millions of Chinese are poor peasants living on the vast river and coastal plains where there are no true forests and no space left by intensively cropped fields, canals, houses, and roads to establish them, the planting of trees on any available spot in the lowland farming regions-a prodigious agroforestry effortwill have to be a part of the solution.

2.1.5 Agroforestry

Strictly defined, agroforestry, or agrisilviculture, is the practice of growing crops and trees together- crops for food, feed, or cash income and trees for one or more of these reasons (i.e., trees themselves may be crops), for fuel wood, as well as for environmental (soil, water) benefit. Here I intend to use the term in the broad sense to embrace all practices of tree growing in predominantly farming and virtually deforested areas, that is, not only true intercropping but also all tree plantings in what the Chinese call the "four besides": beside water (streams, ponds), beside roads, beside fields, and beside houses.

But first I will devote a few paragraphs to agrisilvicultural practices as narrowly defined. These have many understandable appeals: ecosystemic, agronomic, silvicultural, economic, nutritional (Budowski 1978; Weaver 1979). Land competition is lessened or eliminated outright by the joint production, while the trees create a desirable microclimate and provide nutrients for the crops (through litter in all cases, more significantly through nitrogen fixation when leguminous species or modulated trees are planted) as well as a continuous or cyclical harvest of food, fodder, fuel, timber, medicinal ingredients, and extractives. The integration of trees and pastures is also possible, an interesting way of controlling weeds, fertilizing grass, diversifying nutrition, and increasing land productivity.

The Chinese have traditionally been among the most proficient practitioners of agroforestry, and today they frequently interplant peanuts and corn and small grains with Chinese dates (Ziziphus jujuba) in the North, and the fast-growing Cunninghamia lanceolata and the oilbearing tea oil tree (7hea oleosa) in a variety of agroforestry combinations in the South (FAO 1978). The potential for the substantial expansion of these worthwhile practices is very large, especially in the South and when 7hea oleosa is involved. This small slow-growing tree appears to be an outstanding agrisilvicultural choice in warm (annual mean 15-22'C), moist (700-2,400 mm), and hilly (above 800 m above sea level) locations, precisely the conditions throughout most of the South.

In deep soils the smallish tree will grow for more than 100 years, and its seeds yield good edible oil (also easily storable); its long flowering period (five months) is excellent for beekeeping; its dense wood makes fine farming tools, and its pruning (necessary for maximum crown) provides some fuel wood. Intercropping with soybeans, peanuts, sweet potatoes, and in tea plantations is common, and the oil yields of well-managed stands are over 100 kg per ha, plus over 200 kg of oil cake, an excellent animal feed. And, of course, the tree's preference for hilly locations makes it an outstanding choice for reducing slope erosion.

But even the most extensive tree intercropping imaginable would give neither sufficient protection to rural China's deteriorating environment nor enough fuel wood to its households, especially throughout the country's densely populated farming plains. There the planting of suitable species in all available locations is imperative, and despite some impressive progress, much more can be done, as shown by the statistics for the large Jiangsu-Shandong-Henan plain (Hao 198 1). The plain's farmland that is networked with trees now totals 4.8 million ha, but this still represents only about 43 percent of the total suitable for such interplanting. About 1.2 million ha of farmland is now intercropped with tung oil trees (Aleurites fordii), but there is the potential for at least double this amount.

Over 3 billion trees have been planted on the plain since 1949 in the "four besides," but because this land and other scattered uncultivated patches amount to only about 15 percent of the provinces' farmland (even in the Shanghai suburbs the share is 10 percent) and because 200 trees could be planted for each mu (that is, up to 3,000 per ha), there is the potential for nearly tripling the amount or, again, certainly for at least doubling it. The trees most suitable for these plantings are various fast-growing poplars (Populus), willows (Salix), foxgloves (Pautownia), Sassafras, and locusts (Robinia pseudoacacia). Small groves, rows, or belts of these trees have proved their environmental worth in China: measurements within sheltered areas show wind speeds reduced by 30-50 percent, humidity raised by 10-30 percent, and evaporation reduced by 14-40 percent compared with the open plains; these beneficial changes can help to raise grain yields by up to 25 percent (NCNA, October 9, 1980, SWB 1105).

In addition to the usual advantages of more fuel wood (especially with fast-growing coppicing species), the leaves of most of the trees just mentioned can be used as feed for hogs, sheep, and rabbits, an important consideration in intensively farmed regions where little or no land can be set aside for feed crops. Another desirable option is to grow oil-yielding, nut-bearing, or fruit trees. As already detailed, in the South it is tea oil trees, currently grown on about 2.7 million ha (Lin Zi 1980) and suitable for a severalfold expansion; in the North the tung oil tree is the species of choice, planted usually with a density of three or four per mu. TUng oil trees, whose valuable industrial oil is an ingredient in some 2,000 different products, are now also enjoying a renaissance in Sichuan and other southern provinces (Xie Zijun 1980). The olive tree (Olea europaea) has been spreading quite rapidly since its introduction in 1964.

Fruit trees favored in "four besides" plantings are apples (Malus), Chinese chestnuts (Castanea mollissima), walnuts (Juglans mandshurica), persimmons (Diospyros kaki), and Chinese dates (Ziziphus jujuba). The dates are especially popular, with dry yields of up to 250 kg per ha. In the South the preferred trees are lichees (Litchi chinensis). Fruit trees are particularly desirable for enlivening the overwhelmingly vegetarian diet (see section 6. 1. 1), and their direct financial benefit can be surprisingly large: a village in Qianxi County in eastern Hebei found that each yuan invested in chestnut-growing returned over 100 times more than the money earned from general field farming (Luo 1981). Chinese dates and lichees are also fine export items.

2.2 Erosion

Deforestation invariably accelerates erosion. While in heavily forested regions virtually no raindrops fall directly on the ground -and even if they did they would be intercepted by the accumulated layer of organic debris-it is the impact of the drops on bare devegetated soil (and not, as it might intuitively seem, the runoff velocity of water) that detaches the largest quantities of soil particles and causes the heaviest soil losses. The greatest water erosion results from intensive summer storms, while the most severe wind erosion problems arise in and semiarid regions often swept by strong seasonal winds. And if the soils subjected to such storms and winds are light and easily erodible, the stage is set for a virtually uncontrollable natural hazard.

Unfortunately, a combination of all three preconditions is present in a large area of Northwest China-on the Loess Plateau, which extends over more than half a million km2, a region that has served as a classic paradigm of massive erosion in the world's geomorphological and pedological literature.

 

2.2.1 Loess erosion

The origins of this phenomenon are still debated, but it appears certain that the genesis of the problem goes back as far as 1.2 million years ago, when huge amounts of fine yellow particles, composed mainly of quartz, feldspar, and mica, were lifted by the northwesterly winds from the deserts of Mongo!ia and Xinjiang and deposited in the area bordered by the Lüliang Shan in the east, the Qin Ling divide in the south, the Ordos Desert in the north and northwest, and the mountain ranges of Qinghai in the west.

There are many other loess areas elsewhere in Asia and on other continents, but this is the world's largest aeolian accumulation of fine grained (0.05-0.005 mm in diameter) sand: the Loess Plateau (Huangtu Gaoyuan-Yellow Soil Plateau) covers some 530,000 km2 in 200 counties in eastern Qinghai, south-central Gansu, southern Ningxia, central and northern Shaanxi, southwestern Nei Monggol, and western Shanxi. Other toess deposits in China are in Liaoning and Jilin in the Northeast and in Xinjiang on the fringes of huge intermountain basins, but the layers are not too extensive (they total only some 50,000 km2), and they are relatively thin (less than 50 m). In contrast, the heart of the Loess Plateau has deposits 100-200 m thick, and large parts between Baiyu Shan and Xifeng and east of Lanzhou are covered by deposits more than 200 m thick, in places as much as 300 m thick (Wang and Zhang 1980).

Surficial features of these thick loess deposits largely correspond to the underlying palaeotopography. Where there were ancient rolling hills there are now liang (elongated mounds) and mao (round mounds); where the flat bedrock was overlain there arose yuan (high table-like plains with abruptly descending edges); and on the terraces of different heights produced by faults, loess layers form flat taiyuan. Naturally, there are many transitional forms created by denudation. Characteristic erosional landforms, shaped largely by running water from summer thunderstorms, are countless loess gullies, gorges, walls, columns, caves, funnels, and grooves (Wang and Zhang 1980). The appearance of heavily eroded loess landscapes is wild and depressing: one is awed by the power of natural processes as well as by the magnitude of the task required to bring these forces under control.

Yet this region, now predominantly desolate and wasted, harbors in its southern part, in the valley of the lower Wei He (Wei River), the core of ancient Chinese civilization: Banpo Village, a neolithic settlement, is just a few km east of Xi'an on the Wei He, Xi'an being the 3,000 year-old city that served intermittently as the capital of 11 dynasties for 1,100 years; some 30 km east of the city were buried the sculpture legions of Qin Shi Huang (259-2 10 BC), the first great unifier of China; about 150 km north of the city, in Huangling County, is the tomb of the Yellow Emperor, the legendary founder of the Chinese nation.

This long and conspicuous history paved the way for the vast environmental destruction of the region because, previously covered by forests and fine grasslands, it was stripped of its vegetation to build cities and imperial palaces, to heat houses, to cultivate crops. For centuries the erosion has been advancing. The tragedy of the recent past is that this advance has been, in general and in spite of much local effort to control it, dangerously accelerated. Severe erosion now exists on some 430,000 km2 of the plateau, or about four-fifths of its total (an area larger than Japan), and 280,000 km2 are particularly seriously affected (NCNA, April 17, 1980, SWB 1080).

As long as vegetation, and above all trees, protects the ground from the direct impact of raindrops and from wind erosion, topsoil losses in the loess region are not unusually high. As soon as the protective plant cover is stripped away, however, the extremely porous and easily erodible loess layers are washed or blown away. Water erosion is by far the most crippling; although the region is relatively dry (average annual rainfall is just 200-400 mm while evaporation is up to 1,400 mm), over two-thirds of all precipitation comes down in heavy summer thunderstorms when the high kinetic energy of raindrops easily dislodges the loosely accumulated loess particles.

The annual losses of topsoil are now over 10,000 tons per km2 in Yulin and Yan'an prefectures in Shaanxi and in LUliang and Xinxian prefectures in Shanxi, and they average 4,000-5,000 tons per km2 a year throughout most of the region, whose most eroded parts are now just a lifeless maze of deeply cut narrow and steep-sided gullies. The main reason for such staggering losses has been the widespread and indiscriminate application of the Maoist "grain-first" policy and the resultant conversion of remaining thin forests (a mere 3 percent of the plateau still has some forest cover) and good grasslands into fields.

The grandiose Huang He basin control plan published in 1955 envisaged a "comprehensive solution" of the erosion problem by large scale afforestation, grassing, terracing, and dam building. Instead, the Chinese are now admitting that the ecosystemic destruction on the loess highlands has been especially severe during the past 30 years (NCNA, March 7, 1979, SWB 1026).

In their pioneering, unorthodox Renmin ribao article (which opened a lively debate on the future developmental strategies for the region) Tong and Bao (1978) gave some shocking examples of the conversions and their effect on the livelihood of the peasants. For example, in Guyuan County in Ningxia the average grain share per capita was 410 kg in 1949, enough to give an adequate if not rich nutritional supply, and in the mid-1960s peasants were able to sell over 4,300 tons of edible oil and more than 3,000 head of cattle to the state each year. Then came the destruction of previously lucrative cattle breeding (in 13 communes of the county all of the grassland was converted to grainfields) and the cutting down of local forests (20 percent of the total area is now lost).

Cultivated land increased by about 82,000 ha, but after the first few years of good harvests the yields on the land unsuitable for cultivation plummeted to a mere 250 kg/ha. By 1977, per capita grain availability shrank to a mere 100 kg, a clear starvation level, only 400 tons of oil and 800 head of cattle were sold, and the cash income stood at a pitiful 19 yuan per capita. Similarly, in Maolin Village (Yan'an Prefecture, Shaanxi) each person could be provided with 700 kg of grain in 1943, a very comfortable supply; by 1977 the harvest was down to 250 kg per capita-and many localities have shared the same fate. Tong and Bao's (1978) startling conclusion: "What merits our attention is the fact that to this day the production levels and living standards of the masses in quite a number of places are still lower than during the early postliberation period or during the War of Resistance against Japan."

Few admissions among the recent Chinese revelations are so damning about the consequences of Maoist developmental policies.

Regionwide statistics show that the per ha grain yield is now only 1,275 kg (compared with the nationwide average of about 4,300 kg/ ha), with a considerable portion of the land yielding only 225-375 kg per ha and, consequently, with food rations below subsistence levels not even 150 kg per capita in 1977 The effects of the rapid erosion do not end with ruined fields and starvation harvests. Local precipitation is lowered as the ground moisture, and hence evaporation, declines, and unpredictable frosts become more common, making the selection of proper crop varieties for sustainable farming more difficult.

And eventually the tiny loess particles end up in the region's streams. The main rivers draining the Loess Plateau-the Wuding He, the Yan He, the Luo He, the Huan Jiang, the Jing He, and the Wei He- annually contribute hundreds of millions of tons of sand and silt to the Huang He, making it the siltiest river among the world's major streams; from the Wuding He basin alone 300 million tons enter the Huang He each year (Huang Yongshi 198 1). Before the river enters the loess land it carries, on the average, just 2 kg of silt per m3 of water; when it leaves the region, after its confluence with the Wei He near Laotongguan in Shaanxi, there are, on the average, about 35 kg of silt in every m3, and the highest recorded content was 651 kg/m3 (several Shaanxi rivers during peak stormflows carried slurries with more than 1.5 tons of silt per m3!).

This process, elevating the Huang He's riverbed in its lower course and posing a permanent threat to the large populous areas of the North China Plain, has been greatly accelerated by the recent environmental destruction on the Loess Plateau. In the early 1950s the Huang He annually carried about 1.3 billion tons of silt through the Sanmenxia gorge (between Shanxi and Henan, just before the river enters the North China Plain). By the late 1970s this staggering mass rose to 1.6 billion tons, roughly a 25 percent increase in less than three decades (Tong and Bao 1978). This accelerated silting of the Huang He is one of the most dangerous and, at the same time, most intractable environmental problems facing China, and it has few equals, even in global terms.

Effective control of the river will be impossible without adequate control of the silt's source, that is, without a major environmental rehabilitation of the severely eroded loess region, which now contributes at least 1.1 billion tons to the river's silt load. The story of the Sanmenxia dam is a costly lesson that illustrates this fact (Smil 1979a).

2.2.2 The Sanmenxia miscalculation

The grandiose plan for the complex utilization of the Huang He, approved in July 1955, had as its centerpiece the construction of 46 dams on the river to impound water for the generation of 110 billion kWh of electricity a year, to expand sevenfold the irrigated area in the basin, and to extend the navigable sections more than twenty fold (Berezina 1959). The dam at the gorge of the Three Gates Sanmenxia, located in Henan approximately 120 km downstream from the rectangular bend near Laotongguan-was to be the largest and most important project of the original Huang He cascade.

Designs prepared with Soviet aid called for a 110 m-high and 839 m long concrete gravity dam to create a 3,500 km2 reservoir and to retain as much as 36 billion m3 of water, one-and-a-half times the average annual volume of flow at the site. The project was to control 98 percent of the river's annual runoff, to cut the heaviest summer flood flow from 37,000 to between 6,000 and 8,000 m2 per second, to provide irrigation for 2.6 million ha, and to allow the installation of 1,100 megawatts (MW) of generating capacity. The cost was a staggering sum for the China of the 1950s: approximately US$(1957) 700 million.

The planners were, of course, aware of the danger of rapid silting, but they thought that it could be controlled by a variety of measures. The control plan foresaw the mass construction of 215,000 works to protect the heads of gullies, 633,000 silt check dams, and 79,000 silt precipitation dams as well as extensive afforestation, grassing, and terracing. The intent of the combined effect of all these measures was to extend the life of the reservoir to at least 50 to 70 years. As the experts confidently concluded, any "difficulties that may arise in power generation, irrigation, and navigation as a result of the silting up of the reservoir ... will be comparatively easy to deal with" (Teng 1955).

This turned out to be-probably owing to both the faulty initial appraisal and the subsequently much higher rates of silting-an astonishing and potentially extremely dangerous miscalculation. Silting of the reservoir greatly exceeded the original projection; most of the incoming mud and sand was being retained in the lake, and this accumulation became especially worrisome as the deposits started to extend rapidly upstream to the Wei He above Laotongguan, elevating the inlet channel and seriously endangering the densely populated agricultural plain and the city of Xi'an, China's ancient capital and now her eleventh-largest urban area.

Of the reservoir's total storage (7.7 billion m3), 4.55 billion m3, or 59 percent, was silted up between 1958 and 1973; at such rates the reservoir would have been completely filled in just 24 years even without storing floodwaters every year. On two occasions when flood flows were intercepted by the darn as much as 90 percent and more of the silt was retained: between September 1960 and March 1962, 1.645 billion m3 of silt entered the storage, and 1.529 billion m3 (93 percent) was deposited; and between July and October 1964, of 2.724 billion m3 of silt carried by a heavy flood, 1.952 billion m3 (72 percent) stayed in the storage (Li Changzhe 1980). Clearly, with annual flood storage the reservoir's life would be just a few years.

Emergency adjustments were therefore necessary. Power generation had to be stopped because the lowest water intake for the turbines was still higher than the natural river level at the Wei He confluence; but the removal of the turbogenerators and abandonment of water storage did not solve the problem because spillway intakes were too high and the silt continued to accumulate. The only possible solution was to open large outlets at the bottom of the dam, lower the steel penstocks, and turn them into silt discharging tubes. This piercing of the dam tripled the discharge capacity, thus appreciably helping to move the silt downstream. Compared with the average annual storage capacity loss of 540 million m3 in 1960-1970 the current value is about 10 million m3, but all three key roles of the reservoir-storage of irrigation water, power generation, and above all, guarding against floods and prevention of damage to dikes downstream in Henan and Shandong-have been substantially compromised by the reconstruction. Because between July and October the floodwaters carry more than 80 percent of the annual silt load passing through Sanmenxia, summer flood impoundments had to be cut drastically to minimize silting, and water is now stored only between flood seasons when the river carries some 40 percent of its average annual flow but with only 10-20 percent of its silt load. Simply put, when the dam is needed most it can be now used least.

This miscarriage of Sanmenxia's mission is clear when winter and summer LANDSAT images of the reservoir are compared (Smil 1979a). On the one hand, winter (and spring) images show the gorge segment of the reservoir between the dam and the Huang He bend filled with relatively clear water, swelling in the least confined place to a width of more than 6 km and covering as much as approximately 250 km2; on the other hand, during the peak flood season the reservoir below Laotongguan shrinks to a narrow ribbon of heavily silted water with an area as small as 90 km2 .

And Sanmenxia's predicament is repeated, on a smaller scale but often at faster rates, in reservoirs throughout the loess region. Some reports on impoundments in the Wei He basin mention losses of 20 percent of total storage capacity in just three years after completion, and a Shaanxi conference on antiflood work repeatedly issued warnings on blocked channels and silted reservoirs impairing flood drainage and control; provincewide, 512 million tons of sediment is deposited in reservoirs annually, corresponding to a staggering 15.3 percent loss of the total storage capacity (Li Chaobo 1980). The economic cost of turning Sanmenxia and hundreds of other reservoirs into hazardous storers of silt rather than useful water control projects is, needless to say, enormous. Sanmenxia now has an installed capacity of just 250 megawatts, instead of the planned 1,100 MW, and even so it must be idle most of the time, and its annual storage cost is estimated to be Y 30 million (Li Changzhe 1980).

The reconstructed version of Sanmenxia reservoir does little to eliminate the silting problem in Henan and Shandong where the Huang He, confined by dikes, flows 3 to 5 (in places up to 10) m above the surrounding countryside. Of the 1.6 billion tons of silt the river carries when it enters the North China Plain, some 400 million tons are deposited annually in its lower course through the two provinces, continuously raising the riverbed (the current rate is about 12 cm a year) and creating a major flood hazard. The ancient task of repairing and strengthening the dikes must continue. The magnitude of this effort without which the North China Plain would become uninhabitable can be illustrated by work done in Shandong between 1958 and 1976 (NCNA, March 18, 1977, SWB 921). More than 3M million m3 of earth and almost 7.3 million m3 of stone-virtuary all of it dug or cut, lifted, moved, and placed by human hands alone-went into the dikes. The volume of earth and stone that went into raising and strengthening 1,800 km of the Huang He dikes in Henan and Shandong during the 30 years between 1949 and 1979 totaled 480 million m3, a result of annual water and spring reinforcement carried out by 300,000-400,000 people (Henan provincial broadcast, January 30, 1980, SWB 1072).

Sanmenxia's failure and the continued hazardous elevation of the Huang He's lower course can be alleviated only by controlling the source of the problem-by moderating the erosion rates on the Loess Plateau.

 

2.2.3 Controlling loess erosion

The seriousness of the environmental degradation on the Loess Plateau and its dangerous downstream consequences have led to increased scientific and public policy concerns and to proposals for viable long range strategies. Since 1973, repeated high-level meetings have put forward plans to ameliorate the destruction by extensive application of traditional remedies: planting trees, planting grass, and terracing. A gathering in Xi'an in March 1979 proposed to partition the whole plateau into four development zones, each stressing a different land use but still retaining a considerable measure of field cultivation even outside the specialized agricultural region.

This region, concentrated on grains, cotton, and edible oil production, would include northern Shanxi, western Gansu, and central Shaanxi as far north as southern Yan'an Municipality. The forestry and animal husbandry region would cover the hills and gullies of southern Gansu, southern Ningxia, and northwestern Shaanxi north of Yan'an; here the planting of trees and grazing herbages would have priority, with some grain cultivation along the rivers and on the terraced riverbanks. Central Gansu, southern Ningxia, northwestern Shanxi, northern Yulin Prefecture, and some parts of Nei Monggol would specialize in animal husbandry while also planting windbreaks and grazing forests as barriers against further erosion. Finally, the forestry region would embrace higher-lying areas in the counties around Ziwu, Long, Huanglong (Shaanxi); Lao, Jiao, Liupan (Gansu and Ningxia); and the Lüliang Mountains (Shanxi) (Xinhua, March 20, 1979, SRT 1026).

Tong Dalin and Bao Tong (1978) were the first influential decision makers to argue persuasively for a more radical and ecologically appropriate approach: in the long run most of the area is not suitable for field crops, and hence an appreciable portion of the farmland should be reconverted to pastures and forests that would not only bring higher economic returns but also safeguard the Huang He's lower course against further rapid rising.

Tong Dalin (1980) singled out the region's large fossil fuel (especially coal) and hydroenergy resources and its great potential for animal grazing as the two essential advantages, and large-scale planting of pasture plants (above all perennial legumes), shrubs, forests, and fruit trees and construction of dams and water conservancy projects as the key preconditions for success in alleviating the plateau's current rnisery. Sun Hua of the Northwest Agricultural College, quoted by Tong (1980), argued that the plateau with its countless hills and slopes could support very extensive orchards, including red dates and grapes in Ningxia and northern Shaanxi, olives and citrus fruit in the southern part of Shaanxi, and apples in Gansu. Planting of vigorous bush fruit trees and vines, such as raspberries, gooseberries, and pomegranates, should also be encouraged.

Tong and Bao (1978) illustrate the benefits of a grazing-forestry orientation with two cases, one historical, the other recent. They note that although in the course of the more than 2, 100 years from the Qin dynasty to 1950, 973 major dike breaches occurred in the lower course of the Huang He, there were only two major breaches during the 580 years from Wang Mang (AD 9-23) to the beginnings of the Sui dynasty (581-618), a time when the Loess Plateau was primarily grazing area with a few crop fields.

Their contemporary example cites the benefits brought about by changes in Gaoxigou Village of Mizhi County in northern Shaanxi, whose farmland used to yield starvation levels of around 200 kg of grain per ha. From the late 1950s on they cut the cropland by two-thirds (increasing the total yield substantially) and concentrated on planting grasses and trees. A record rainstorm in the summer of 1978 (which brought 300 mm, or as much as the total precipitation in 1977) washed away the topsoil of new fields into reservoirs in neighboring villages, but Gaoxigou emerged unscathed. Although the authors may be indulging in some hyperbole when they describe "crystal-clear-water streams" around the village, there is little doubt that very substantial microand meso-scale changes can be effected by protective plantings.

The Gaoxigou example is wholly consistent with the measurements made by the Northwest Water and Soil Conservancy Bureau in 1973 when the total rainfall in Shaanxi averaged 346 mm: the exposed fallow land lost 6,750 kg of soil per ha, and farmland erosion reached about half of that value (3,570 kg/ha)-but on the grasslands only 93 kg was lost, and the water erosion in forested area carried away a mere 60 kg of topsoil-two orders of magnitude less than on the fallow land (of course, during prolonged summer droughts, much cropland turns into de facto barren land as the plants wither away!).

Afforestation may often be difficult, however, owing as much to the harshness of the area as to the lack of experience (see section 2. 1. I for more details). Many more people in the area know how to grow grazing herbages-different grasses, alfalfa, sweet clover-and that is why Shi Shan (1980) of the Support Agriculture Office of the Chinese Academy of Sciences made a proposal for a major conversion of crop fields to pastures. He estimates that 15 kg of seed and a month of work are needed to plant one ha of new pasture, which should be, from the third year of growth, sufficient to feed 15 sheep or 600-750 rabbits. After 8 years the pastures have to be reseeded.

The arguments advanced by Tong, Bao, and Shi have not been universally accepted. Stubborn advocates of the grain-first policy maintained that only the land in excess of 2 or 3 mu of grainfields per capita capable of providing 400-500 kg/capita of food grain should eventually be converted to pastures or afforested. Simply put, as long as there is no local grain self-sufficiency, it would be impossible to develop animal husbandry or to afforest. What a fine example of putting abstract, ideologically influenced guidelines ahead of simple economic realities and inescapable environmental imperatives! Naturally, in the short run the conversion may (but not at all necessarily) mean the loss of some foodgrain-but in return the country would gain valuable forest and animal products, much needed wood and meat, plus, a no less critical consideration, downstream protection and alleviation of the Huang He's silting.

By 1980 the rational approaches respecting natural conditions and pointing to inevitable specialization finally gained the lead. They were supported by hundreds of local studies and scientific reports, and in April 1980 a Xi'an symposium, organized by the State Scientific and Technological Commission and the Chinese Academy of Sciences, reached, after an emotional debate, a consensus on turning the eroded loess areas into pastures and woodlands (Xinhua, April 16, 1980, SWB 1080). Fourteen counties were selected as pilot areas for implementation of extensive environmental conservation programs to be executed with the state's financial and technical help. One of the major innovations of this program is aerial sowing of forage grasses, which is to be performed yearly over thousands of ha of eroded slopes.

New studies of the erosion process and river silting should be also helpful in the control effort. For example, Chen Yongzong (1976) discovered that statistically the most important determinant of the erosion rate is the slope gradient rather than the intensity of rainfall or the depth of runoff. In practical terms this means that before grassing or planting young trees an effort should be made to change the slope gradient to greatly lessen the erosion. Qian Ning, after analyzing the records of 103 Huang He floods between 1952 and 1960 and between 1969 and 1978, concluded that most of the silt consisting of coarse particles originated in two areas totaling 100,000 km2 along the river's course through Shaanxi, Nei Monggol, and Shanxi . This finding allows the concentration of the anti-erosion effort to be more specific and more highly effective (NCNA, March 29, 1980, SWR 1077).

As with the giant northern tree shelterbelt, at least 5 to 10 years will be needed before the pace and the success of the conversion and control program can be appraised: its direction is undoubtedly correct; its execution will depend not only on the state's sustained financial and professional aid but equally as much on the maintenance of the newly introduced private incentives, which make the conversion to woodland and grazing very appealing from the monetary viewpoint alone. Another campaign condemning a private woodlot on a slope beyond a peasant's house, two scores of rabbits in cages in his courtyards, and a dozen sheep in a pasture as the "last vestiges of capitalism" and the bold new program will go the way of Shaanxi's topsoil in a July thunderstorm.

As long as the Huang He receives large amounts of silt from the Loess Plateau, new ways will also have to be found to control the changes in the river's delta in the Bo Hai. Currently the sediment adds an average of 23 km2 annually, and during the years 1855-1972 the coastal zone (there is no sharp delimitation between the silty shallows and the low-lying sedimentation deposits) advanced seaward at a rate of 150-420 in a year (Figure 10). Lengthening of the estuary decreases flow velocity and aggravates silting, and the addition of new land thus necessitates extending the protective dikes about 40 km and digging new channels to control the water and silt movement (NCNA, March 29, 1980, SWB 1077).

Obviously, there are no easy, shortcut solutions to China's loess erosion and Huang He silting. Filling countless gullies, building small dams on streams to check the silt flow, leveling fields, and terracing slopes are essential steps in cutting down the rapid soil erosion, and the Chinese are both experienced and skillful in these endeavors. The conversion of fields to grazing should also prove relatively uncomplicated, although grassing on steeper slopes will be more difficult to accomplish. Afforestation is the critical step-and the most difficult: survival in the and climate with pronounced hot-cold fluctuations and months-long dry spells has been appallingly low, and a great deal of research and testing will have to be done on the selection, planting, and management of the most suitable tree and shrub species. I believe that loess erosion is one of the most intractable environmental problems, even in global terms, and it will demand much greater and deeper attention than it has been given since 1949. Such attention, costly and only slowly rewarding, is absolutely necessary not only to ameliorate the environmental degradation in the region itself, and thus to improve the meager livelihood of its nearly 50 million inhabitants, but also to protect the low-lying plain in Henan and Shandong from the excessive silt accumulation that is still raising the river's yellow waters above land inhabited and tilled by more than 50 million peasants.

 

2.2.4 Spreading erosion

Serious erosion and silting is far from limited to the Loess Plateau and the middle and lower Huang He: it now blankets 1.5 million km2, or about 15 percent of all China (Ma and Chang 1980). In the Huang He's upper course, the Longyangxia hydroelectric project in Qinghai, China's second largest hydro powerplant, is now threatened by rapid wind erosion (Li Wei 198 1). Conversion of grasslands and hilly forest land to grainfields raised erosion rates from Heilongjiang in the northeast to Yunnan in the southwest. But certainly the most disquieting recent news regarding erosion outside the Loess Plateau has been the disclosures of the magnitude of the problem in the Chang Jiang basin.

There, large-scale deforestation, especially during the 1970's, has been responsible for rapid topsoil losses (in Jiangxi as much as 4 cm a year), a decline in crop yields (in parts of Hunan cereal harvests are down to just 750 kg per ha), and widespread stream and reservoir silting. The following details on the basin's erosion and silting problems are taken from recent papers by Guo Tingfu (1980), Li Changzhe (1980), Liu Haifeng (1980), Wang and Chen (1981), Huang Yongshi (1981), and Wang Ganmei (1981).

Serious soil erosion now affects 360,000 km2 in the Chang Jiang basin, or about 20 percent of its 1.8 million km2 area, with 2.4 billion tons of topsoil eroded every year, 50 percent more than in the Huang He watershed. In the Jinsha Jiang section of the basin in Yunnan, erosion rose from 130 million tons in 1958 to 290 million tons in 1974 (Wang and Zhou 1981). Sichuan and Guizhou formerly had low erosion rates, but the recent increases resulting from deforestation have been rapid. In the basins of three Sichuan rivers-Jialing Jiang, Tho Jiang, and Fu Jiang-the erosion is extremely serious, with annual losses of over 250 million tons, the equivalent of a 12-cm-thick topsoil layer over 100,00

ha of farmland. Water erosion in the hilly and mountainous terrain of the deforested regions along the river's upper course is compounded by gravitational erosion (mud and rock flow frequently stripping the thin soils to bedrock). In Guizhou many locations now have barren rock patches on badly eroded slopes; erosion cones and deposits cover the foothills and farmlands and make the cultivation of previously good soils impossible. In this sense erosion in the mountainous region of the Chang Jiang basin is much worse than on the Loess Plateau, where even severe erosion still usually leaves thick layers of loess for eventual revegetation.

Most of the eroded soil is deposited in the rivers and lakes, continuously elevating their beds, blocking their outlets, lowering their storage capacity, and increasing the dangers of severe flooding. Dongting Hu in Hunan, a major regulating water body in the basin, lost over 1,600 km2 to heavy silting, and it now measures only little more than 2,700 lan2; Poyang Hu, China's largest freshwater lake, in Jiangxi, now contains 29 percent more sediment than it did in the 1950s.

In flood-prone Hubei Province 178 reservoirs in Huanggang District lose 6 million m3 of storage capacity each year to silting, many small reservoirs had to be abandoned after just two or three years, and some larger hydroelectric stations had to be taken partially out of operation. A recent survey of 33 large and medium-size storages in the basin revealed that 16 are already more than half filled with silt, and the average useful life will not surpass 13 years. Even Danjiangkou, China's most voluminous reservoir on the Han Shui in Hubei, is silting rapidly: 1 15 million tons of mud and sand are now entering the storage each year, and in just over a decade it lost 580 million in3, or oneseventh of its total capacity. And there are fears that the useful life of the Gezhouba hydroproject, China's largest power station, now under construction on the Chang Jiang in Hubei, will be considerably shortened (see section 3.2.1 for details).

Silt content of seven other major rivers in the basin increased by 23 percent in the years 1966-1975 compared with the 1956-1965 level, and shallow waters have made navigation impossible on 977 km in Jiangxi Province since 1957, and on 6,455 km of 214 rivers in Hubei since 1960. During the more frequent flooding, waters carrying sand and pebbles deposit this debris on the farmland or scour the fields. In Xishui County near Wuhan, one-seventh of all arable land has been thus affected. In terms of fertility losses, the topsoil currently eroded in the Chang Jiang basin contains nitrogen equivalent to the output of 50 medium-size (500,000 tons a year) fertilizer factories. And there is a mounting direct human and economic toll as the floods, aggravated by the losses of water retention in forests and of storage capacity in lakes, reservoirs and rivers, appear to become ever more destructive. There can be little doubt that the devastating Sichuan flooding in the summer of 1981, in which 15 million people were affected and when the peak flow recorded in Chongqing was the highest since 1949, was intensified by the massive deforestation and erosion in the Chang Jiang basin (see also section 2.1.3).

Of course, the Chang Jiang is still much less turbid than the Huang He: normal suspended loads of the two rivers are, respectively, just around I kg/m3 and over 30 kg/m3, but this large difference is owing to the Chang Jiang's much larger water flow (see section 3.2. 1). In absolute terms the difference is now only 2.5-fold (640 vs 1,600 million tons), and the rapid increases of the Chang Jiang's silt load have led many Chinese scientists to dramatize the worsening situation by stating that the river is becoming a second Huang He; even the premier, Zhao Ziyang, has picked up the phrase in his exhortations for accelerated afforestation.

Afforestation provides the only practical solution, and interesting comparisons published by Li Changzhe (1980) show how effective it can be. During a torrential rain in August 1975 parts of southern Henan received 8001,000 mm in just three days, and, on the one hand, two reservoirs in the area with only 20 percent forest cover were breached, exacerbating the downstream flooding; on the other hand, two nearby reservoirs, in whose basins 80 percent of the land was forest-covered, survived the flood undamaged. Long-term comparison of silting rates shows that in the former case 20 cm of silt a year (and up to 40) are deposited on the bottom; in the latter case the new layer is mere 1.5 mm, a 133-fold difference!

Another outstanding illustration of the anti-erosion effectiveness of forests is Fengman reservoir on the Songhua Jiang (Sungari River) in Liaoning. Surrounded by forests, in 27 years it lost only 142 million m3, or a mere 1.3 percent of its 10.78 billion m3 storage capacity; at that rate, therefore, even after 1,000 years it would be only half silted. In this case, however, an argument can be made that the Changbai Shan surrounding the reservoir are covered with natural climax forests (though cut over) and that new plantings would be far from so effective. Generally this is true in the short term, but in just a few decades such plantings can afford excellent anti-erosion protection, especially in the south where a variety of fast-growing trees can be used. In the 23 years since the construction of Ting Jiang reservoir in Fujian, 85 percent of the barren land surrounding the storage was afforested, and the annual silting rate is now a mere fraction of a percent of the total volume (Li Changzhe 1980).

Urgent and extensive action is needed in virtually every region of the country to moderate the rapid erosion rates. Nationwide figures are quite alarming: topsoil loss is now at least 5 billion tons per year, and with this mass go millions of tons of principal macronutrients, nitrogen, phosphorus, and potassium, as well as precious rnicronutrients -China's annual synthetic fertilizer production provides less soil nutrition than that lost to erosion! And if only larger reservoirs (over one million m3 storage capacity) are counted, China has recently been adding 260 million m3 of new water storage each year-but 80 million m3 per year have been lost to silting (Qian Ning et al. 1980).

 

2.3 Desertification

In and lands everywhere, erosion is one of the principal causes of desertification, a severe environmental degradation affecting globally nearly 50 million km2 on all continents (about 35 percent of this total is in Asia) and directly influencing the lives of more than 600 million people. In the 1970s there was concentrated research and public attention when this problem was brought into worldwide focus by the Sahelian drought of 1968-1974, and culminated in the United Nations' Desertification Conference in 1977 (United Nations 1977). The general consensus blames poor land management rather than climatic changes as the principal cause of desertification, and the southerly advance of China's large interior deserts provides clear support for this view.

China's deserts, extending over an area of 1, 095,000 km2 (about II. 4 percent of the total territory), are overwhelmingly concentrated in the Northwest and North, west of 106'E (Figure II, also Appendix A.4.c, A.5.a). Takla Makan, with 327,000 @m3, by far the largest, is made up mostly of huge composite dunes usually 100-150 in high. Nationwide, the sandy deserts, including the true gobis as well as sand dune areas in the desert steppe and steppe zones, constitute 59 percent of the total; the rest is largely gravel-covered (Department of Desert Research 1978). Most Chinese deserts receive less than 200 in of precipitation annually, while evaporation typically ranges between 3,500 and 4,000 mm, and aridity indexes, generally above 4.0, reach as high as 20-60 in the Tarim Basin. Only in the smaller deserts and sandy lands of the Northeast does the summer monsoon bring up to 450 mm of rain and keep the aridity index between 1.5 and 4.0. The survival of grasslands bordering these deserts is thus precarious even without human interference, and anthropogenically induced desertification of northern China is an ancient phenomenon, with large parts of Mu Us Shamo in Nei Monggol and nearly all of Horqin sandy land in Jilin as the two most extensive examples of the advance. Both areas used to be fairly rich grasslands, the former with rainfall in excess of 400 mm and with marshes and thickets in sheltered areas. Gradual reclamation of grasslands for fields, tree felling, and overgrazing destroyed the stabilizing plant cover and the sands began to shift: a 60 km wide belt along the Great Wall between Shaanxi and Nei Monggol and the large sandy stretches in the bend of the Xiliao He (Xiliao River) have been so desertified over the past 200 years.

The very same processes have accelerated since 1949: like spreading erosion, most of China's current desertification can be clearly ascribed to human actions, above all to widespread and indiscriminate reclamation for grainfields as well as to frequent overgrazing and tree felling. Between 1949 and 1980 about 65,000 km2 (an area roughly equivalent to twice the size of Belgium and Luxembourg) in 207 counties in II provinces was desertified, and 91 percent of this large loss was caused by improper land use; an additional 160,000 lun2 is in danger of desertification (Ma and Chang 1980; Wang Lichao 1981; Wen and Liu 1981). The country's extensive northern and northwestern grasslands bordering the large interior deserts of Mongolia, Ningxia, Gansu, and Xinjiang have suffered most heavily. Nei Monggol offers the most drastic examples. Of the region's 86 banners and counties, 66 are affected by desertification, and windblown sand now encircles 33,000 km2 of farms and grasslands (Ma Wenyuan 1980). In the Ih Ju league 12,000 km2 of grasslands turned sandy between 1957 and 1972, and during the late 1970s more than 2,000 km2 became sandy each year. Forty percent of the Ju Ud league's 47,000 km2 of grasslands has become sandy. Similar widespread losses have occurred in Ulanqab and Xilin Gol leagues.

Still, the losses may seem minor compared with the nearly 2.87 million km2 of true grasslands (of which 2.2 million km' could be used for pasture) in the semi-arid northern regions. However, only 15 percent of this total is abundant long-grass meadows, and while outright desertification destroyed only some 65,000 km2, according to one source about 460,000 km2 of pastures has qualitatively degenerated -and the statistics revealed at a 1979 national conference on grassland development show that more than one-quarter of China's usable grasslands, that is, at least 550,000 km2, has deteriorated owing mostly to desertification or alkalization (Xinhua, July 29, 1979, SWB 1044).

The effects are predictable. First, the advancing desertification cuts down both the quantity and quality of the grass available for grazing and drastically lowers the amount of hay for winter feeding. Consequently, in the northwestern part of China as much as 2 ha of pasture are needed to support a single sheep (Ma Wenyuan 1980), and large numbers of cattle die not only of cold but of plain hunger, a harsh reality best illustrated by the fact that animals dying as a result of chronic spring shortages of fodder far surpass the number of cattle delivered to the state by herdsmen (Xinhua, July 29, 1979, S" 1044). The resulting decline of cattle herding in Nei Monggol (although not all of the drop can be ascribed to desertification!) is shown by the published growth rates: while between 1949 and 1958 animal husbandry output rose by an average 9.2 percent a year, the rate dropped to 2.9 percent between 1959 and 1969, and average animal production decreases of 0.7 percent prevailed between 1970 and 1978 (Xinhua, December 3, 1978, SWB 1018).

Second, when the formerly grass-protected, water-retaining surfaces become barren, the frequency of droughts increases and the strong anticyclonic winds cause more frequent sandstorms, further expanding the desert area. Beijing's Meteorological Observatory recorded an average of three sandstorm days a year in the early 1950s (Watts 1969); during the 1960s the number rose to 17; in the years 1971-78 the mean frequency was 20.5 days annually (Anonymous 1979b); and between 1974 and 1980 it reached 26 days (NCNA, July 10, 1981, JPRS 78623).

Third, the reclaimed fields have suffered from extensive erosion and evaporation, with grain yields quickly deteriorating. Thus, in spite of Nei Monggol's large-scale reclamation of arable land-each person now has 0. 355 ha compared with the nationwide average of 0. 106 ha -per capita grain availability declined considerably in comparison with the mid-1950s, and the region's average cereal yield is now less than 1,500 kg per ha, a mere one-third of the all-China figure (Hu 1981).

A long overdue restitution is now under way: further conversion of grasslands into fields has been banned; the orthodox, and unachievable, requirements of local grain self-sufficiency have been dropped; animal husbandry enjoys its former priority, and an effort is being made to rationalize its operations; private woodlots on desert fringes are encouraged; and new research centers have been set up to study the techniques of controlling the drifting sands. One of these new centers, at Dongsheng in the Ih Ju league, will concentrate on the Mu Us Shamo and Hobq Shamo in the Ordos Bend, while the Dengkou station will work on the Ulan Buh Shamo west of the Huang He. The Desert Research Society of the Institute of Geography was established in October 1980 (Hou Renzhi, honorary president, Zhu Zhenda, president), and several scholarly journals are now being published.

Research and experiments conducted in many of China's desert areas have selected the most suitable trees and established many effective sand-controlling techniques (Department of Desert Research 1978). For planting inside or on the fringes of oases as well as for shelterbelts, the following trees and shrubs have the most widespread utility: various poplars (Populus cupidata, bolleana, diversifolia, simonii, nigra), willows, (Salix nwtsudana, cheilophila, j7avida, microstachya), tamarisks (Tamarix chinensis, ramosissima), saksaul (Hatoxylon ammodendron), and Russian olive (Eleagnus angustifolia). Approaches found useful for controlling the shifting sands include a variety of fringe shelterbelts, combined tree-shrub-grass structures, and the establishment of sheltered enclosures (so called "grassy kulums") to protect pastures.

By far the most important recent development in China's attempts to control desertification is the start of work on a huge forest shelterbelt system which is to eventually span some 4,000 km. The project was approved by the State Council in 1978, and it is intended to be a key strategic measure to check advancing desertification and to improve the conditions for farming and animal husbandry. The shelterbelt is to extend over three northern areas (hence its Chinese name san bei): the Northwest, the northern part of North China, and the western part of the Northeast. A total of 324 counties (or banners) in Xinjiang, Qinghai, Gansu, Ningxia, Nei Monggol, north Shaanxi, northwest Shanxi, the Bashang area of northern Hebei, and in the western parts of Liaoning, Jilin, and Heilongjiang will be involved. Their combined area is about 2.6 million km2 (nearly one-quarter of China's territory), of which almost half is deserts, and more than 200 counties in the combined area with 13.3 million ha of farmland and pastures are being damaged by sand. The livelihood of 44 million peasants in the area is very poor, with low grain rations, insufficient fuel, and miserable housing.

The first stage of what the Chinese call the "great green wall" is to be completed by 1985. About 5.3 million ha of trees are supposed to be planted to increase the forest coverage of the desert fringe areas from the present 4 to about 10 percent and to give what the Chinese term "initial protection" to 13 million ha of farmland and pastures (Xinhua, June 1, 1979, S" 1041). Existing forests and shelterbelts will be incorporated into the project which, the Ministry of Forestry urges, should be carried out through measures "suitable to local conditions" without "practicing formalism or demanding uniformity in everything" (Xinhua, June 7, 1979, S" 1041).

These are encouraging guidelines insofar as in the past large afforestation campaigns were too often implemented in a simplistic manner by planting one tree species regardless of its suitability for local conditions. The planting of bushes and grasses is also an important part of the effort, as is the strengthened protection of existing vegetation to be enforced by special units of forest police and by new forestry courts set up in the san bei area (Xinhua, June 17, 1980, SWB I 100). Even in the best of circumstances such a grandiose undertaking would not be easy to manage and to bring to a reasonably effective conclusion. To bring about visible positive results in just a matter of years in the harsh environment where even the existing trees have difficulty surviving would rank among the most notable successes in environmental improvement anywhere. Initial claims of sapling cultivation and tree planting have been most encouraging, but only by the end of this decade will we be able to judge the extent of the accomplishment.

 

2.4 Reclamation of lakes

The grain-first policy destroyed, above all, many forests and grasslands, but it also seriously affected the country's aquacultural resources. Fresh-water fishing and breeding of aquatic species are ancient Chinese traditions, with the first comprehensive instructions, by Fan Li, preserved from the year 473 BC, and with a wide variety of practices (Gu 1975; FAO 1977a; Ryther 1979; Pritchard 1980). Over 500 fish species reside in China's fresh waters, and although at least 200 are suitable for human consumption, the catch is dominated by four cyprinid species (family fish or Chinese carps): grass (Ctenopharyngodon idella), black (Mylopharyngodon piceus), silver (Hypophthalmichthys molitrix), and bighead (Aristichthys nobilis) carps. Other carps, including the golden (Carassius auratus) and the common carp (Cyprinus carpio) and shads, perches, breams, and tilapia are also widely bred, as are several crustaceans and water plants, above all lotus (Nelumbium speciosum), water chestnut (Trapa natans), and prickly water lily (Euryule ferox).

These traditions and tastes were carried on after the establishment of the PRC, and in the late 1950s no less than two-fifths of the nation's aquatic harvest came from inland waters, with 70 percent supplied by natural catches and 30 percent by aquaculture (Cong 1979). Since then the irrational Maoist policy of "planting crops in the middle of lakes and on the tops of mountains" has drastically reduced the size of many of those natural water bodies that were best suited to intensive aquaculture. There have been other damaging effects: blocking of migration routes by dams and locks while failing to provide special passages or installing improper fish ladders (in the Chang Jiang basin, over 50 large lakes have had their outlets to the river cut); growing urban, industrial, and farm pollution (see section 4.2); as well as appalling overfishing, including dynamiting and poisoning. Most crippling, however, has been the massive reclamation of lakes and ponds.

Examples from all around the country abound. In Northern Jiangxi's Poyang Hu, China's largest fresh-water lake (originally 565,000 ha), 331 embankments were built during the 1970s to reclaim close to 90,000 ha of fields, and the area devoted to aquaculture shrank by half, to just 26,000 ha; Dongting Hu in northern Hunan was, with 435,000 ha, China's second largest fresh-water lake, but land reclamation reduced its surface to only 282,000 ha (see also Appendix A.5.b); halfmoon-shaped Tai Hu in Jiangsu was diminished by more than 10 percent to 213,000 ha (Cong 1979). Even in and Nei Monggol where any water surface is precious, Ulansuhai Nur was reduced to 22,000 ha, or only one-third its original area (Tian and Liu 1979). Shanghai Municipality lost 30 percent of its fish breeding area, and in Zhejiang Province nearly 3,000 ha of fish ponds were turned into fields. But Hubei, the proverbial "province of a thousand lakes, " has been affected most of all: of its 1,056 lakes larger than 1,000 mu (66.6 ha) less than 400 remain, and the total provincial lake water surface dropped by 75 percent between 1949 and 1978 (Hu and Tian 1981). Major lakes that completely or largely disappeared in eastern Hubei include Chen Hu, Diaocha Hu, Xi Hu, and Bei Hu; Hong Hu was reduced to 40,000 ha, or half its original size, and much of the remainder is so shallow that people can wade across (Figures 12 and 13). For China as a whole incomplete statistics for 1949-1978 show a loss of over 20 million mu, or at least 1.33 million ha, of waters suitable for inland fishery (Xinhua, March 17, 1979, S" 1026).

The effects have been predictable. The natural inland catch in 1978 was only half of that in 1954, and it constituted only 30 percent of the total fresh-water harvest (compared with 70 percent in the 1950s). In spite of the massive extension of reservoirs (for hundreds of large and tens of thousands of small hydrostations as well as for irrigation) and pond breeding, the total fresh-water harvest was down from 40 percent of the aggregate aquatic output in 1959 to just 23 percent in 1978 (Xinhua, April 11, 1979, S" 1030). In some leading fresh-water fishery provinces and counties the decline was even more precipitous.

Guangdong's annual catch went from 20,000 tons in 1966 to just 10,000 tons in 1978; the output of Hong Hu County in Hubei fell 60 percent between 1975 and 1978 (NCNA, December 24, 1979, SWB 1064). Fresh fish, a traditional delicacy enlivening an overwhelmingly vegetarian diet, has become scarce even in the locations previously famous for their rich catches. In Wuxi Municipality on the shores of Tai Hu, in the very midst of Jiangsu's "rice and fish country," annual consumption of fresh-water fish amounted to just 4.5 kg per capita per year (Xinhua, March 1, 1978, SWW 979). In 1979 in the Zhu Jiang (Pearl River) delta, and in Chang Jiang lakeland Xinhua correspondents often heard people saying that "it is now very difficult to get fish to eat" (NCNA, December 24, 1979, SWB 1064). In Hubei the provincial annual average of fresh fish consumption per capita stood at only 2.5 kg in 1979 (Hubei provincial broadcast, April 2, 1979, SWB 1030). And China's total fresh-water fish output of 1.11 million tons in 1979 translated into a nationwide average of a mere 1.1 kg per capita (Xinhua, November 14, 1980, SKB I II 0)-this in a country with a chronically precarious food balance where fish offers perhaps the most appropriate opportunity to increase the consumption of essential high quality protein.

Naturally, not only fish catches were lost: the harvests of water plants for food, waterweeds for fertilizer and feed, and reeds for various manufactures have also declined substantially. For example, in Nei Monggol's Ulansuhai Nur reed cuttings plummeted from 2025,000 tons annually to a mere 3,400 tons (Tian and Liu 1979). And, most ironically, after freshwater breeding, spawning, and feeding grounds were destroyed truly en masse, a large part of the reclaimed land surrounding the lakes is now reported to be wasted and empty, unsuitable for cropping after a few harvests. But the water surfaces are gone, and with them their moderating influence on local climate, above all on the extension of the frostfree period (a critical consideration in all colder double-cropping regions of populous eastern China), and their floodwater retention capabilities. And it is also beyond any doubt that the recent floods in the Chang Jiang valley were intensified by the severe loss of lake storages there.

The costly irrationality of all this is striking. On the one hand, there is the waste of much heavy labor, investment, and scarce materials (iron, cement) for massive reclamation of land from waters that could yield valuable protein, regulate water supplies, and moderate local climate; on the other hand, there is the cultivation of rice or wheat in the newly reclaimed land with often very low yields, and not infrequently for just a short time. On the one hand, there is the destruction of natural water surfaces and freshwater fisheries; on the other hand, there are continuing mass campaigns to build more water reservoirs, to dig up new fish ponds, and to set up new aquatic breeding centers. Indeed, it is the perfect image of towering inconsistencies and pervasive mismanagement of natural resources so characteristic of China for the past few decades.

Nevertheless, given the proper incentives and conditions, the potential for China's aquaculture is undoubtedly impressive. The country's

lakes, rivers, reservoirs, and ponds cover 26.7 million ha (lakes alone account for more than 6 million ha), of which at least 5 million ha could be used for fish breeding but only 48 percent of which is so used (Xinhua, April 15, 1979, SWB 1033). The opportunities are especially favorable in more than 80,000 reservoirs, but only less than two-thirds of their 2 million ha is now being used for aquaculture, and the average yield has been very low, a mere 85 kg per ha (Xiao Peng 1981). In contrast, the average yield in lakes is about 140 kg per ha, and in 7,400 ha of China's extensive fish ponds it was about 1,000 kg per ha in 1979 (NCNA, January 4, 1980, SWB 1066). Yields at Hubei's Hong Hu used to be around 4,000 kg, Tilapia bred in southern lakes could bring up to 18,000 kg, and raising fish in nets cast into reservoirs and lakes can provide (of course, only from small areas) harvests as high as 70 tons per ha (NCNA, December 24, 1979, SWB 1064). Careful management, scientific approaches, and priority attention to pond cultures could thus bring impressive nutritional benefits, with up to 5 million tons of fresh fish produced annually (Cong 1981).

What is being done to stop the further destruction of inland water surfaces? For one thing, new regulations governing the breeding and protection of aquatic resources, issued on February 10, 1979, by the State Council, do not explicitly forbid any new conversions but prescribe that "in reclaiming land from sea or lakes, it is necessary to make overall and systematic arrangements to ensure that aquatic resources are not damaged" (Xinhua, March 27, 1979, SWB 1030). Perhaps even more important is the disappearance of bureaucratic pressure to convert to grainfields, as most of the reclamation arose from the central directives rather than from the wishes of local peasants. As a result, in some areas much of the reclaimed land was reflooded, and with the return to diversified farming most villages traditionally specializing in aquaculture have made a strong comeback.

Incomplete statistics for 1980 show an increase of 200,000 ha in fresh-water fishing grounds and about an 8 percent growth in total catch (NCNA, November 15, 1980, SWB I I 10). In Hubei the increases were stronger: 16 percent more fish caught, 18 percent more fish fry bred (Hubei provincial broadcast, September 18, 1980, SWB 1110). For 1981, catches were expected to be at least 10 percent higher than in 1980, and in most provinces, peasants are now allowed to dig fish ponds on their private plots, an opportunity seized by 1.2 million southern households on 24,000 ha by the end of 1981 (NCNA, December 26, 1981, SWB 1168). Should the current rural development policies continue, environmental damage caused by the reclamation of lakes and the destruction of fresh-water aquaculture may be substantially reversed in a relatively short period of time.

The same should apply to reclamation of coastal areas where grainfields displaced beaches traditionally used for the breeding of marine species. Between 1959 and 1978 nearly 70,000 ha of such beaches were turned into fields, incurring the loss of thousands of tons of fish, shrimp, oysters, and mussels. As with fresh-water production, the loss was not balanced by grain production gains: it took years to wash the reclaimed fields clean of salts, and former fishermen could not farm well. Even when they were relatively successful, their incomes suffered: for example, the average annual output value of one mu of oysters is equivalent to the gain from two outstanding crops of rice on two mu of land; for seaweed or mussel breeding the ratio is 1:5 (Yang Haiqun 1979).

Again, such conversions, including a massive plan in Fujian that would have displaced 100,000 fishermen in 13 coastal counties dependent on mussel breeding, were largely halted after 1978, and labor intensive mariculture is now also promoted as an excellent means of earning foreign exchange: after all, one ton of prawns can buy 56 tons of wheat, and ope ton of dried oysters can buy 44 tons of wheat on the foreign market. Such comparisons make the grain-first policy and the demands for local grain self-sufficiency patently losing propositions!

2.5 Losses and deterioration of cultivated land

Preceding sections have described how the understandable Chinese preoccupation with rising grain production found unfortunate and environmentally disruptive expression in the irrational mass land reclamation campaigns, leading to deforestation, filling of lakes, erosion, and desertification. There is now general agreement among Chinese farming experts that the price paid in terms of ecosystem degradation has been too high and that such efforts must not be repeated-especially in view of the fact that most of this reclaimed land is relatively unproductive, coming nowhere near the yields of old fertile farmlands. For example, Weng and colleagues (I 98 1) state that on the average 3-5 ha of such land still produces less than the output of one ha of established cropland.

In a country that has to feed nearly a quarter of mankind this alone would be a most regrettable result: destroying valuable ecosystems for a dubious low-quality, often just temporary, gain. What makes the situation truly tragic is that the destruction of natural ecosystems and the creation of poor quality cropland has been accompanied by almost incredibly large losses of prime arable iand -a diabolical combination of processes that ruins the irreplaceable natural heritage, negates the immense labor sacrifices of the peasants, and puts China farther away from the cherished goal of comfortable food self-sufficiency.

 

2.5.1 Losses of arable land

Given its physical endowment and large population, modern China never had the luxury of abundant farmland. In 1957, the last year of the 1950s for which official figures are available, China's cultivated area was 1. 677 billion mu, or 111. 8 million ha (State Statistical Bureau 1960); in per capita terms this translates to just 0. 172 ha. In spite of the official information blackout from the late 1950s on, much had been inferred about the country from a variety of fractionary sources, but nothing indicated the magnitude of the change that was revealed in the late 1970s.

The first clear admission of the loss, contained in a document on agricultural development issued by the Central Committee of the CCP '(1979), was that between 1957 and 1977 "more than 100 million mu," that is, some 7 million ha, of cultivated land was requisitioned for various capital construction projects. This loss alone implies that China's arable land in the late 1970s was no more than about 105 million ha, a decline of at least 6 percent in comparison with the late 1950s. On many occasions in 1979 and early 1980, however, a variety of Chinese sources, including the Soil and Fertilizer Institute of the Chinese Academy of Sciences (1979) and official Xinhua releases, were referring to just 1.5 billion mu of arable land nationwide, that is, only 100 million ha.

Finally, in April 1980 Zhang Zhenming (1980), writing in Renmin ribao, revealed the complete statistical dimensions of the losses and gains for the two decades, an account of astounding damage: between 1957 and 1977 China lost, to urban and rural construction and to natural disasters, 500 million mu or 33.33 million ha, an incredible 29.8 percent of the 1957 total. This is a truly frightening figure whose full import is not easily comprehended: the world's most populous nation with already scarce farmland losing nearly a third of its prime cropland in just one generation! The losses were partially made up by the previously mentioned reclamation campaigns: 21.33 million ha (320 million mu) were added in this way so the net loss was 12 million ha, or 10.73 percent of the 1957 total.

Slightly different figures were offered 18 months later by Yi Zhi (198 1) in Hongqi: a total decrease of 440 million mu (29.33 million ha) and reclamation of 260 million mu (17.33 million ha)-the resulting net loss of 12 million ha, however, is identical with Zhang's figure. Even when looked at in strictly quantitative terms-ignoring the inferior quality of most of the reclaimed land-it is an awesome drop, equivalent to the disappearance of all cultivated land from Sichuan (China's most populous province, with some 100 million people), Guangdong (60 million), and Guangxi (about 36 million) combined.

This farmland, capable of sustaining some 200 million people (the equivalent of both Germanies, the United Kingdom, and France put together!), is gone-while China's population grew during that period by about 300 million people. Consequently, per capita availability of farmland dropped from 0. 172 ha (2.59 mu) in 1957 to 0. 104 ha (I. 57 mu) in 1977-a 40 percent decline-and it is now no more abundant than in Bangladesh or Indonesia, and about 2.5 times less than in India or Brazil (FAO 1980). Naturally, in many provinces and localities the losses have been even more rapid. In Liaoning, with the massive expansion of heavy industrial concentrations and settlements, farmland available per capita shrunk by almost 60 percent, in Shaanxi by nearly the same amount (Yi 1981).

Perhaps most striking has been the loss of suburban farmland, traditionally planted mostly to several crops of vegetables (Figure 14). Beijing Municipality had 607,000 ha of such land in 1949, and only 427,000 ha by 1980, a 30 percent loss brought on by the state taking over two-thirds of all vegetable areas in close-by suburbs for new housing and factories instead of rebuilding the decaying city itself (Zhang Zhenming 1980). During the 1970s these farmland withdrawals proceeded at a rate of 20,000 mu (1,333 ha) a year, and should this trend continue for another decade, the equivalent of all of Haidian, the city's principal vegetable producing district, would be gone. Yet even now, in spite of priority deliveries for the capital, vegetable supply is strained because China has no long distance vegetable transport like North America or Europe, and suburban cultivation is thus decisive: while for each of the capital's 2.03 million inhabitants there were 0.3 ha of vegetable plots in 1949, today there are only 0.05 ha for each of the city's nearly 9 million people, and a decline of this magnitude is difficult to reverse even with intensified multicropping and higher yields.

Similarly, in Shanghai over 650 ha of vegetable plots were confiscated in 1979 and 1980, and in Wuhan some 2,000 ha, representing one-third of all suburban vegetable fields, were diverted to construction uses (Qui Yuan 1981). In all Xi'an suburbs, cultivated land is now less than 0.02 ha per capita, making the continued survival of farming families impossible (Anonymous 1981b).

Serious as the losses of suburban land are, they could be slowed or even reversed if the city and provincial authorities redirected growth from outward expansion to urban reconstruction. Controlling the waste of farmland in rural areas will prove to be much more difficult, however. The huge rural losses can be attributed to three main processes: the activities connected with farming, the indiscriminate expansion of small-scale industries, and the building of new houses. Surprisingly, the first category is a relatively minor contributor: new roads, irrigation canals, reservoirs, and grain storage and processing grounds usually do not eat away excessive areas. Tractor stations, however, have been taking too much land: where just 0. 33-0.40 ha would be entirely sufficient for a six-or seven-machine enterprise, up to 1.3 ha of land are taken (Feng and Li 1980).

But small rural industries and housing have been the main offenders. The frenzied construction of small factories to provide local self-sufficiency in many industrial products has been often hailed by Western admirers of Maoist China as a splendid contribution to a "small is beautiful" world. In reality, these enterprises, with primitive technologies, shoddy management, rampant duplication, prodigious waste of resources, and huge production inefficiencies, have been an overwhelming economic loss and burden to the countryside. Built in response to erratic decisions during dubious campaigns, they have taken over much flat, easily accessible prime farmland. Yi (1981) describes the process succinctly: "Why could an enterprise leader advance a proposal to use cultivated land by a snap of his fingers after he got out of a jeep which had just carried him to the countryside?"

Small industries are now deservedly out of favor, but the newfound prosperity of many rural areas and the greater room for individual initiative-those precious gains for which the Chinese peasants have waited so long-promise only further destruction of farmland by the peasants' pursuit of that universal symbol of well-being, a new family house. Between 1978 and 1980 more than two-fifths of the cropland taken out of cultivation in the southern provinces was used for new rural housing. Disaggregated figures for Hangzhou Prefecture in Zhejiang, a province with losses of 13 percent of arable land between 1957 and 1978, show that in 1979 new housing took away 90 percent of the lost farmland, state capital construction the rest (Wang Weizhang 1980). In some counties the demand for new houses affects up to 150 ha a year, and up to twice that area is taken by brickworks (Yi 1981). Every household in the North now occupies about 80-90 M2; in the South, 70-80 M2; but some new buildings,, yards, and walls often appropriate 400-500 M2 (Feng and Li 1980).

The construction of lavish houses and country villas by the leading local Party bureaucrats sets an example for their subordinates, and in violation of repeated recent injunctions against it, the practice thrives. Zhang Zhenming's (1980) most incredible admission, coming as it did in the official Party newspaper, is the bleakest possible assessment of future prospects: "The state has, in fact, lost control over the use of land in villages. " This appraisal must be combined with the realization that the pent-up demand for better rural housing to replace the old homes will intensify over the next generation with the rising demand for new housing to accommodate the record numbers of newlyweds, whose families, even with strictest population controls, will add at least 200 million people to China's one billion by the year 2000. Clearly, some heavy taxes on the private requisition of farmland, and especially on suburban lots, should be enacted and enforced, and the state itself should set an example in conserving land in its new construction projects. The dilemma is well defined by Yi (1981): "The housing problem must be solved, but not at the expense of solving the food problem. "

The continuation of past trends for even just another two decades is unthinkable: with the population growing to 1.3 billion, per capita availability of farmland would slip to a mere 0.068 ha (680 M2, a square of 26 x 26 in), and a 30 percent yield increase from every cultivated field would be required just to maintain nutrition at its current barely sufficient level. Even if the annual losses were brought to just one-third the rate prevailing since the late 1950s (that is, to some 550,000 ha a year), and even if continuous reclamation could match past achievements and add another 20 million ha in two decades-both very optimistic assumptions-the net addition would be just 9 million ha, and even with a population of only 1.2 billion-again a highly optimistic assumption-the per capita farmland availability would still go down to about 0.091 ha.

 

2.5.2 Qualitative deterioration

High future production gains from every planted field, however, are totally unrealistic because the quantitative decrease of arable land has been accompanied by the widespread qualitative decline of soils throughout China, a process that imposes serious limitations on continuing yield increases. To begin with, 40 percent of the current 100 million ha under cultivation is on soils already belonging to the low-yield category: 20 percent are hillside fields with relatively thin soils prone to erosion, 8 percent are waterlogged lowlands and saline-alkaline fields, over 9 percent are sandy soils, and the remaining nearly 3 percent are poor paddy fields (Yi 1981). These poor soils typically yield no more than 0. 75-1.50 tons of grain per ha (100-200 jin/mu), and not infrequently the harvests fall below half a ton per ha (Soil and Fertilizer Institute of the Chinese Academy of Science 1979). The best illustration of how widely these poor soils are distributed is that fact that 16 of the country's 26 provinces and regions have grain yields lower than the national average (Hu 1981).

And, unfortunately, many recent cropping practices are seriously degrading the previously good or excellent soils. Careless irrigation is spreading the area of saline and alkaline soils throughout the Huabei Plain, and it creates bog or gley soils in many intensively cropped parts of Jiangnan: for example, no less than 40 percent of the cultivated area (1.2 million ha) in Hunan has been turned into bog soils whose lowered air content and lowered temperature slow down the absorption of nutrients and impede plant growth. Tai Hu lakeland in Jiangsu is also seriously affected, as are large areas in Guangdong. Continuous cropping of rice throughout Jiangnan has been responsible for the spreading formation of "blue asbestos mud," an approximately I 0-cm-thick clay layer about 20 cm below the soil surface that retains water and fertilizer, prevents root growth, and causes frequent rotting (Xi 1979). Crops grown in these degraded soils, shallow and deficient in organic matter, do not respond properly to higher (and costly) water and chemical fertilizer inputs; many southern paddies now yield annually just 2.253.0 tons per ha, and peasants talk about hybrid strains bringing misfortune because of their fields not being able to surpass the record yields set in the 1960s (Qiu 1979).

The continuous double- and even triple-cropping of rice, the improper application of synthetic fertilizer and lower quantities of organic fertilizers, and the failure to rotate wet and dry crops (especially decreased green manure and legume planting) have been greatly accelerating soil degradation. Rice multicropping has been promoted as a component of the simplistic Maoist policy of indiscriminate expansion of grain growing, and more irrigation and more chemical fertilizers were thought sufficient to bring repeatedly higher harvests. Proper crop rotation had been neglected. As a result, Jiangnan's area planted to green manures decreased from 8.4 million ha in 1972 to 7.7 million ha by 1977; Shandong's soybean fields shrank from 2 million ha in 1949 to 0.73 million ha by the niid-1970s (Soil and Fertilizer Institute of the Chinese Academy of Science 1979). The cultivated area of various edible beans has been steadily declining for years (Zheng Zhouje 1981). Fewer green manures and legume crops and the diminished application of organic fertilizers have recently caused a rapid decline of organic matter content in soils and further exacerbated phosphorus and potassium deficiencies, now one of Chinese agriculture's most intractable obstacles to additional yield increases.

Domestically produced and imported synthetic nitrogen fertilizers surpassed the nitrogen theoretically available in organic wastes in 1975-76, and the ammonia and urea output has since risen sharply with the construction of 13 large, modern plants imported from the United States, Western Europe, and Japan. Total synthetic nitrogen applications now average about 106 kg per ha of farmland or, taking into account the extensive Chinese multicropping (the index now stands at 150), nearly 70 kg per ha of sown land. This is a very respectable performance-the European average is just about 100 kg, the U.S. mean is some 55 kg, and the Soviets apply only around 35 kg of nitrogen per ha of farmland (FAO 198t)-yet the yield returns on applications have been diminishing nevertheless.

These decreases are not owing, as in some countries with very intensive fertilization, to the virtual saturation of plants' needs but rather to a widespread disregard of an essential ecological tenet, venerable Liebig's law of minimum limiting nutrients. In the early 1960s, the application of one kg of standard nitrogen fertilizer (20 percent nitrogen content) increased the production of rice by 4 to 5 kg, of wheat by 2 to 3 kg, and of cotton by 2 kg-while in recent years all of the returns have been a mere one kg or less (Liang, Lin, and Li 1980). At that time, the nitrogen/phosphorus application ratio was an appropriate 1:0.6, but since then the described expansion of nitrogenous fertilizer output has not been matched by proportionate increases in phosphorus applications, with the disparity being especially acute in the northern provinces. Widespread phosphorus deficiencies lead to phenomena well described in soil science literature: poor root development, reduced or no tillering, short stems, retarded blooming and fruition, poorly developed grains, overall low yields.

Phosphorus deficiencies have been noticeable for some time in some parts of China, but potassium shortages are fairly recent. Formerly, extensive recycling of crop residues and animal manures was able to meet all of the potassium demand in traditional farming, but with increased applications of nitrogen fertilizers, higher unit yields and higher multicropping, insufficient straw recycling, and the virtual absence of potassium fertilizers, shortages of this essential macronutrient have started to slow down yield increases, especially in southern China. Extensive burning of crop residues for fuel (see section 6.1.2 for details) takes away annually at least 3.2 and perhaps as much as 4.9 million tons of potassium, as well as around half a million tons of phosphorus and about 1.5 million tons of nitrogen. With intensified farming and higher nitrogen use, micronutrient deficiencies are also bound to spread, and many cereal crops, and above all rice, may be more susceptible to lodging unless more silica, previously returned with recycled straws, is applied to stiffen the straws.

The remedies are straightforward but costly because the current Chinese macronutrient disparities are considerable: instead of the desirable nitrogen:phosphorus:potassium (N:P:K) ratio of 1:0.6:0.2, current applications average 1:0.2:0.002 (Liang, Lin, and Li 1980). Large deposits of phosphorous ores have been discovered in the southwest and central-south part of China, but without first concentrating these ores through expensive production of compound fertilizers the poor Chinese railway network could not handle the voluminous northward shipments. Potassium fertilizer production is now a mere 30,000 tons, and as with most of the world's nations, China's verified reserves of suitable ores are not large. Prospecting for new sources together with appropriate utilization of fuel ashes and recycled crop residues are imperative.

Most regrettably, however, the use of organic fertilizer-be it green manuring or the recycling of plant, human, and animal wastes-has declined to such an extent that negative effects are already clearly evident (see also section 7.4). For example, Zhen Xiazheng (1982) cites numerous instances from Zhejiang Province where a sharp decline in the spreading of organic wastes and a heavy reliance on synthetic fertilizers has led to widespread micronutrient deficiencies and to related plant diseases.

Boron deficiency in rape, cotton, and wheat is marked by the plants flowering but not bearing seeds; potassium deficiency in cotton leads to spreading stem blight; and silica deficiency in rice causes easier stalk lodging. Crops fertilized by chemical materials also appear to have lower pest resistance, and outbreaks of stripe rust, leaf blight, and spike neck pests have become more common. While 80-90 percent of cotton fields used to be planted in rotation with green manures, only 15 percent now have their organic soil content so renewed, and this, combined with repeated heavy chemical fertilizing, has led to worsening soil crusting and a 50 percent drop in yields. As a peasant saying aptly sums it up: "How can you expect a good harvest if cotton is planted on hard-surfaced highways?"

About the impressive effects of green manures on soil structure, fertilizing, and subsequent crop yields there can be no doubt. Some persuasive illustrations have come from three years (1977-1979) of controlled experiments conducted by Xiao Shuxian (1980) in Guangxi. Compared with winter cultivation of wheat with double cropped rice, winter planting of Chinese vetch (Astragalus sinensis) increased the number of soil granules greater than 0.25 mm in diameter by nearly 50 percent and pushed the subsequent rice yields 50 percent higher than those following ordinarily fertilized wheat. Yields in wheat-rice-rice rotation could be, of course, increased by higher fertilizer applications, but the economic benefits are poor, and the demands for seasonal labor exertion are very high. And the soil structure continues to suffer.

The maintenance of good soil structure thus leads to a conclusion directly contradictory to the Chinese cropping policies of the recent past: "Only in those places where the double cropping of rice has a very high yield ... and where labor and fertilizer are sufficient can the triple cropping system of rice-rice-wheat ... be used" (Xiao Shuxian 1980). In all other cases an omission of green manure from the triple cropping cycle will cause the soil structure to deteriorate and soil fertility to decline.

Soils with higher organic content also retain moisture much better. Zhang Qinwen's (1981) calculations for Shanxi show that each ha of green manure plowed under can hold up to 1,950 m3 of water so the aggregate retention on 2.7 million ha of the province's dry fields would total 5.2 billion m3, sixfold the volume of all water stored in Shanxi's reservoirs! As for yield improvements, the incorporation of 15 tons of green manure in southern rice fields usually increases unhusked yields by over 20 percent (Lin Yin 1980).