ดินพังทลายและไหลบ่าผิวครอบคลุมพืชแตกต่างกันและการไล่ระดับสีลาด: เขตทดลองในจังหวัดมณฑลส่านซี จีนตอนใต้พังทลายของดินเป็นกระบวนงานที่เก็บวัสดุดีเป็นสารอาหารจากดินแดนลาด (Polyakov และ Lal, 2004 และ Thomas et al., 1999) ดัง ที่ดินกลายเป็น ขอบเขตอุดมสมบูรณ์น้อย หรือดีหมด นอกจากนี้ ความหลากหลายของพันธุ์พืชและสัตว์ป่าค่อย ๆ หายไป ในที่สุด การคุกคามความมั่นคงของระบบนิเวศทั้งหมด (Cotler และ Larrocea Ortega, 2006, Dregne, 1987 และ Pimentel และ Kounang, 1998) การกัดเซาะดินขนาดใหญ่ขึ้น ด้วยเป็นหายนะทางธรรมชาติเช่นน้ำท่วม ดินถล่ม และเศษไหล (ซีเอ็ด al., 1997 และ Zou และ เร็น 2008) โดยไม่คำนึงถึงผลกระทบของการพังทลายของสิ่งแวดล้อมสิ่งแวดล้อม การพังทลายของดินนอกจากนี้ exacerbates ความยากจนของชุมชนชนบทเนื่องลดในผลผลิตดินThe present study is conducted within the frame of a Sino-German project in southern Shaanxi Province (El Kateb et al., 2009) that aims to develop a decision support system for a sustainable land use planning. Reliable research-based-information on the degree of erosion on different land use forms and slope gradients is lacking in the study area. Therefore, the present study lays emphasis on comparing soil loss and surface runoff on the most frequent vegetation covers at different slope steepness. However, the landscape in the study area is a mosaic of agriculture, horticulture, grass, and forest, which all, except the forestlands, appear in small to very small patches that do not exceed a quarter hectare. Small-sized erosion plots of 7 m2 in size were used to compare the runoff and soil loss on uniform slopes of the different combinations of the vegetation cover and slope gradient on the dominating soil (yellow-brown soil) in the study area. It should be underlined that the objective of the present study is to compare the surface runoff and soil loss at the plot level of the small size of 7 m2, but not at larger spatial scale. However, for comparison purposes, the hectare unit is used in the present study. In many studies (e.g., Braud et al., 2001, Descheemaeker et al., 2006, Mathys et al., 2005, Mohammad and Adam, 2010, Ngatunga et al., 1984 and Vacca et al., 2000) small field plots of a size ranging between 1 and 50 m2 were used to compare erosion on different soil types, vegetation covers or land use forms. The used small plots of 7 m2 in size proved to be appropriate for comparing soil erosion in different vegetation covers and slope gradients, as there was no indication that the gained results are not in agreement with similar studies dealing with soil erosion. Taking forestlands as an example, Hill and Peart (1998) determined for forestlands in southern China, based on extensive literature review, an average erosion rate of 50 kg ha− 1 a− 1. Zhang et al. (2011) estimated an erosion rate of 38 kg ha− 1 a− 1 in average for China's forestlands using USLE (Universal Soil Loss Equation) model. In the present study, an average soil loss of 28.5 ± 9.7 kg ha− 1 was recorded at the low and high forests over the observation period. The lower estimated value is attributable to the shorter observation period of the present study than that used in the above-mentioned literature.The study involved a sample size of 33 erosion plots. The variability was largely depending on the level of the study factors with higher variation on farmlands and steep slopes. In addition, the soil loss had higher variability than runoff. The logarithmic transformation was necessary to stabilise the variance and, thus, perform the ANOVA. There was no any evidence that the assumptions of the ANOVA were violated for any of the analyses conducted. The difference between the means of the levels of the vegetation cover was too large, so that the probability of failing to reject a false null hypothesis was too small. This was more pronounced for the soil loss than for the runoff. The power (the probability of rejecting a falls null hypothesis, cf. Kirk, 1995) of the test for the vegetation cover at a significance level of 0.05 was > 0.99 and 0.85, when analysing the soil loss and runoff, respectively. In comparison, the power of the test for the slope gradient was lower having values of 0.78 and 0.35 for the soil loss and runoff, respectively. It can be concluded that the sample size was sufficient to detect the differences between the means of the different levels of the vegetation cover. However, this was not always the case for the slope-gradient factor, particularly when considering the runoff. According to Kirk (1995) the estimated sample size for each of the combination of the vegetation cover and slope gradient in a completely randomised factorial design to achieve a power of 0.9 for detecting an effect of the slope gradient on the runoff, with magnitude in accord with Fig. 7, is six. For further field studies and with the interests of reducing costs, the sample size may best be weighted to include twice as many samples in farmlands (maize and tea) and at steep slopes, due to their higher variability than the other levels of the study factors.Results of the analyses of variance gave no indication that horticulture (tea plantation with peanut as an intercrop) and agriculture (maize in a winter-wheat–summer-maize rotation) lands at slopes ≤ 30° differ in the runoff or soil loss. Tea plantations at slopes > 30° proved to be the most susceptible vegetation cover to erosion. Tea plantations in the study area are cultivated in rows. The space between each two rows in the tea plantations is as the same as the average height of the tea plants equalling around 40 cm. Under the tea, peanut was cultivated: Seeding took place over the last two weeks of April, while harvesting over the last two weeks of September by removal the entire plants by hand. There was no evidence that harvesting the intercrop affected erosion, as during and after the harvesting time only little rainfalls with low intensity < 5 mm h− 1 had been recorded. However, the intercrop cultivation was not dense and in average 65% of the surface soil remained bare between two rows of tea. Bare soil surface is exposed to rainfall drops that cause greater discharge and soil loss than soils covered with vegetation. A general conclusion that tea plantations induce the highest soil loss may be false, if the cultivation technique is not considered. The conclusion, on the other hand, is valid for the cultivation technique used in the study area, as large proportion of the surface is bare soil, even with peanut as intercrop. However, tea generates a significant income for the local community, although it is a relatively new cash-crop in the study area. To maintain tea cultivation in the study area, the technique used must be altered in order to improve production but, more significant, to reduce erosion and subsequently fulfil the objective of the soil and water conservation. Zehetner et al. (2008) suggested using terracing technique in tea plantations in order to manage that eroded materials can be trapped and downhill transportation can be reduced. Field (1970) and Stigter (1987) proved that organic mulch between rows in tea plantations prevents soil losses.Previous studies (Bakker et al., 2004 and Uri and Lewis, 1998) showed that agriculture activities generate high surface runoff and soil loss because soil tillage causes high erodibility. The present study dealt only with the main cultivated summer crop in the study area, which is maize in a winter-wheat–summer-maize rotation. The study did not consider other summer crops. It is, however, well recognized from Putthacharoen et al. (1998) and Basic et al. (2004) that the degree of erosion is depending on the cultivated crop. Results are, therefore, only valid for the studied maize crop. Nevertheless, introducing soil-friendly farming practices (Gaynor and Findlay, 1995 and Gould et al., 1989) that minimise tilling and thus reduce erodibility, may support the objective of soil and water conservation in the study area.Grasslands, which had been abandoned farmlands and had been left to natural succession, developed in few years to lands covered by grasses and herbs. In comparison to cultivation of maize and tea, grassland had less runoff and soil loss. Fullen (1998) found that grass intercepts the force of rain drops and accelerates infiltration. Ban of tillage in slope lands can considerably reduce runoff and soil losses as found by López-Bermúdez et al. (1998), Souchère et al. (2003), and Feng et al. (2010). However, results showed that over the observation period soil loss on grasslands was 19 times higher than forestlands. This argues for continuing the reforestation activities of grasslands to substantially reduce soil erosion.Analyses showed that the runoff and soil loss produced on both forest types were statistically insignificant. The forestland's runoff was not affected by the slope degree. There was, however, an indication that soil losses on the steep slope gradient were higher than on slight and moderate slopes. Dissmeyer and Foster (1984) found that the slope degree has an effect on the forestland's erosion and that the soil loss is much more responsive to changes in slope degree than the runoff. However, in the Three-Gorges-Reservoir area, Dong et al. (2006) did not find a significant effect of the slope gradient on the forestland's soil loss using 137Cs tracing method.Forestlands, in comparison to the other investigated vegetation covers, generated the least total soil loss (28.5 ± 9.7 kg ha− 1 in comparison to 553.7 ± 183.7 kg ha− 1 on grasslands) and runoff (24.6 ± 3.8 m3 ha− 1 in comparison to 29.8 ± 7.8 m3 ha− 1 on grasslands) over the entire observation period. The same held true for the rainfalls with high intensity (> 5 mm h− 1). However, the rainfalls with low intensity (> 2.75–≤ 5 mm h− 1) induced highe
การแปล กรุณารอสักครู่..