1 Introduction

Agroforestry is the practice of growing trees and crops together on the same piece of land.Traditional or conventional agroforestry refers to the relatively simple but diverse form of agro-silvicultural, silvo-pastoral, and agro-silvo-pastoral combinations(Hsiung, 1996), and has long been practiced all over the world and is still one of the most important farming systems in China and other developing countries, not only for higher crop yields and better rural economy, but also for inexpensive establishment of tree plantations and forest management(Huang, 1982; Nair et al., 1986; Nair, 1987).Poplars are the major tree component of this traditional agroforestry system throughout the south temperate central area of China which includes all or portions of Jiangsu, Anhui, Zhejiang, Hubei, Henan, Shandong, and Shanxi provinces, an area of roughly 600 000 km²(Xu et al., 1988; Fang et al., 1999).Poplar-crop interplantation is the most popular pattern in the areas.Poplar may either be planted at final crop spacing, or be heavily thinned and pruned to achieve a final stand density which is much lower than typically experienced in monoculture forestry, and agricultural crops are taken from within the alleys which are created(Fang et al., 1997).Integrated production systems of this kind can offer a number of advantages when it is compared to traditional monoculture forest planting.Experience from the United Kingdom suggested that poplar agroforestry can be a more profitable option for a cereal farmer than monoculture, provided that the correct cultivars are used, and the downward trend in agricultural prices continues(Thomas et al., 1992; Newman, 1996).Results from China also showed that poplar-crop interplantation can produce more products and gain much greater economic benefits compared to the monoculture agriculture(Xu et al., 1989).However, lack of experimental data or basic information on production of the interplanting system is a serious gap in our knowledge, and consequently, a stable, optimized poplar-crop interplantation pattern is hard to achieve.In order to develop a poplar-crop interplanting pattern which is economically viable, environmentally sound, technically workable, and socially compatible, some new poplar-crop interplanting patterns were designed and established using the principle of edge effects in 1992.This paper mainly tests the difference of these new patterns on ecological benefits, site productivity and economic benefits, and assesses the potential viability of these new patterns.

2 Materials and Methods 2.1 Site description

The study area was located at Hanyuan Forest Farm, Baoying County, Jiangsu Province, P.R.China(33°08′N, 119°19′E).The area has a warm temperate climate with an average growing season of 229 frost-free days.Annual precipitation is approx.964 mm and the radiant intensity on average 494.04 kJ·cm^-2.Mean annual temperature is 14.3℃ and average temperatures are 0.4℃ in January and 27.6℃ in July.The soil is clay loam of moderate fertility, with organic matter content of about 8.43 g·kg^-1, a pH value of 7.4, total nitrogen content of 0.64 g·kg^-1, available P and K contents of 1.764 mg·kg^-1 and 67.574mg·kg^-1 respectively.The water table is about 1.0 m and the bulk density of soil to 1.0 m is 1.35 g·cm^-3.

2.2 Experimental design and tree establishment

The trial was established in 1992 by hand planting one-year-old seedings of clone NL-80351, a hybrid clone of Populus deltoides Bartr.cv.`LUX'×P.deltoides Bartr.cv.`Havard'.Six spacings were designed in the experiment with narrow-wide spacing pattern, i.e.Ⅰ:(3×3)×20 m, Ⅱ:(3×3)×30 m, Ⅲ:(3×3)×40 m, Ⅳ:(4×4)×20 m, Ⅴ:(4×4)×30 m, and Ⅵ:(4×4)×40 m, and the block arrangement was made at random with two replications.The total area of the experiment was about 5.0 hm² and was divided into 12 plots.The size of plots was differnet from 2 300 to 4 400 m², and the plots were arranged in south-north direction.

2.3 Assessments

Trees:Tree height, diameter at 1.3 m(DBH) and crown size were measured at the end of the growing season, and any tree damage or unusual conditions were noted.The mean-tree technique was used to assess the above-ground biomass of the poplar.This technique involves destructive sampling of trees that best represent the mean size of a plantation and uses the number of trees in the plantation to expand mean-tree values to an area basis.The selection of sample tree was based on the multiple characteristics, i.e.the mean DBH of the plantation and on the means of total height and crown dimensions.Each sample tree was cut at ground level in August or September and was divided into three components:stems, branches and foliage, Fresh weights of all components were determined in the field, and samples were collected for moisture and chemical analysis.

Crops:The economic yield of the wheat was measured at harvest time each year for each plot.The sampling positions were detemined according to the distances from the poplar trees and the sampling size was 100 cm×100 cm for each sampling position.However, the biomass production of the wheat was estimated by means of economic yield coefficient method(Hou, 1986), and light use efficiency in each poplar-crop interplanting pattern was evaluated by the biomass method(Hou, 1986).The concentrations of starch and protein in wheat seeds were measured by the anthrone colorimetric method and the coomasie brilliant blue method(He, 1985)respectively, and the thousand grain weight of wheat seeds was also determined for each interplanting pattern and agricultural monoculture.

Microclimate measurement:Four weather measurement systems were located at different distances from the poplar trees and the control site(agricultural monoculture).Field measurements were taken at various phenological phases of winter wheat(4 phases:elongation stage, heading stage, boot stage and ear-milking stage), including illumination intensity, total solar radiation, air temperature, and relative humidity.The measurements at each phenological phase lasted 2~5 consecutive days, and the observations were conducted from 6:30 to 18:30 each day at 2-hour intervals.

3 Results and Discussion 3.1 Temporal and spatial variations in microclimate

Variations in solar radiation:Light passing through a canopy of poplars experiences change in intensity because of the reflectance, absorbance, and transmittance. As shown in Tab. 1, the total solar radiation in the poplar-crop interplantation was lower than that in CK (agricultural monoculture), and the general trend for different measurement positions was center of the wide row (Center) > 3 m near the row west (West) > 3 m near the row east (East). Compared with the CK, the reduction of total solar radiation in the interplantation was from 3.9% to 36.2%, depending on the phenological phases—poplar tree development especially leaf development and measure positions.

Variations in illumination intensity: Photosynthesis is a light-dependent process in which the rate of photosynthetic fixation of both CO₂ and solar energy is largely dependent upon light intensity. The illumination intensity reaching the crop varied greatly according to the phenological phases and the positions of the crop in relation to the canopy of poplars (Tab. 2). The illumination intensity in the poplar-crop interplantation was lower than that in CK (agricultural monoculture), and the variation trend for different measurement positions was similar to total solar radiation. Compared with the CK, the reduction of illumination intensity in the tnterplantation was from 3.9% to 36.2%, which was related to the phenological phases and measure positions. Compensation points and saturation points varied considerablely among different species, among different individuals, among different parts of a single individual. For the individual winter wheat plant the compensation point is about 400 lx, and the saturation point is 30 000 lx. However, for a sward of winter wheat, the saturation point can reach 60 000 lx (Huang, 1984).

As seen in Tab. 2, only at the elongation stage was the illumination intensity higher than the saturation point (30 000 lx) in poplar-crop interplanting systems. Fig. 1 gave the daily variations in illumination intensity for different measure positions. Illumination intensity increased after sunrise, reached its maximum after noon, and then declined. Daily variation in illumination intensity showed that the highest illumination intensity was reached on the center of the wide row, followed by 5 m near the row east, then by 1.5 m near the row west or east. The lowest illumination intensity occurred on the 3 m near the row east, which was almost lower than the saturation point and will directly influence the growth and development of the winter wheat.

Air temperature and relative humidity: Measurements taken at 120 cm above the soil surface indicated that temperatures in the CK were about 1.0℃ higher than those in the poplar-crop interplantations (Fig. 2). However, the difference in the daily variations of air temperature among different interplanting patterns was slight. Overall there were no significant differences in air temperature between CK (agricultural monoculture) and poplar-crop interplantations. The relative humidity in the interplantations was similar to that in the CK at elongation stage, but 2.5% higher at heading stage, 3.6% higher at boot stage and 3.9% higher at ear-milking stage, respectively. Among three measure positions, the relative humidity at 3 m near the row east was the highest, hitting to 85% (four stages averaged), which was 3.7% higher than CK. However, the relative humidity at the other two measurement positions was similar, about 84% (four stages averaged) and 2.4% higher than CK(Tab. 3) Except during a rainstorm, and sometimes even then, the atmosphere is almost always dry enough to permit evaporation and transpiration to occur (Kimmins, 1987). The lower air temperature and higher relative humidity in the poplar-crop interplantations favor the growth and development of winter wheat, especially when a strong dry-warm wind invades this area (generally 3-year cycle) at the ear-milking stage, thus resulting in an increase of wheat yield and an improvement of wheat seed quality.

3.2 Variations in wheat morphology, crop yield and quality

Wheat leaf size and leaf area: Light intensity plays an important role in determining the morphology of plants. Tab. 4 showed the variations in quantity of green leaves per plant, leaf length, leaf width, and leaf area per plant at different sampling positions and various phenological phases. Variance analysis indicated that at the elongation stage there were no significant differences (p≤0.05) in wheat size and leaf area among 7 sampling positions, because at that stage poplar leaves are still developing. However, at the boot phase (when poplar leaves develop into two-thirds of the full size) there existed significant differences (p≤0.05) in leaf area per plant and quantity of green leaves per plant among different sampling positions, though no significant differences existed in leaf length and leaf width. Within the positions of 3 m near the row east or west, both leaf area and quantity of green leaves decreased due to the shading of poplars. Compared with the position of center of the wide rows (10 m near the row east or west), leaf area and quantity of green leaves per plant within 3 m near poplar trees were about 34% and 31% lower than that at the center of the wide rows respectively, so that the net photosynthesis and crop growth will be affected profoundly.

Crop yield:Various poplar-crop interplantations had a great impact on poplar stocking volumes and wheat yields. As the planting density increased, the poplar volume increased while the wheat yield decreased. When the stand age of poplars was at age 4, the reduction in wheat production varied from a minimum of 1.3% to a maximum of 14.8% depending on the spacings and the growth rate of poplars. Compared to another poplar-crop interplantation pattern, a spacing was 4 m×10 m (Xu et al., 1989), which lowered the yield about 50% at the fourth year, these new poplar-crop interplantations only reduced the wheat yield by an average of 8.7%. However, the poplar volume in the stand with spacing 4 m×10 m (about 26.0 m³·hm^-2) was much higher than that of the new interplantations, 3 times that of the pattern Ⅵ, nearly 2 times that of the pattern Ⅲ and 1.2 times of the pattern Ⅰ. Fig. 3 illustrated the variations in wheat yield and 1 000 grain weight at different sampling positions. Variance analysis showed that there were significant differences (p≤0.05)in wheat yield and l 000 grain weight at the various sampling positions, and normal curve best described the distributions of wheat yield and 1 000 grain weight at different sampling positions. In the fifth year, both wheat yield and 1 000 grain weight at the CK were about 15.0%, 11.3% and 2.0% greater than at the sampling positions of 1.5, 3 and 5 m near the row east or west, but similar to the center of wide rows (10 m). From a land use policy point of view, agroforestry may be viewed as nothing more than a means of not producing food (Thomas et al., 1992). This is achieved in two ways. Firstly, the spatial area of land available for agricultural production is reduced due to the necessity for protective strips around the rows of trees and the requirement for fallow areas. Secondly, the productivity of remaining agricultural area is reduced as a result of competition for light, moisture and nutrients associated with the development of the trees.

Wheat seed quality:The thousand grain weight, content of starch and protein in the wheat seed are the important indexes for evaluating wheat quality. The influence of poplar-crop interplantations on content of starch and protein in the seeds was presented in Tab. 5. The most dramatic impact on wheat quality between poplar-crop interplantation and agricultural monoculture (CK) was the content of protein in wheat seeds.Comparea with CK, the protein content of the wheat seed in the interplantations was upgraded by an average of 18.3%, ranging from 15.0% to 20%. However, the difference in protein content of the wheat seeds was not significant (p≤0.05) for different sampling positions. The starch content of wheat seed was also improved through the poplar-crop interplantation. Compared to the CK, the increase in starch content varied from a minimum of 10.9% to a maximum of 15.0% which depend on the planting spacing, by an average of 12.1%. The results observed in this study highlighted the potential for the poplar-crop interplanting management system to improve crop quality, though the system reduced some crop yield because of the competition between crops and poplar for environmental resources.

3.3 Biomass productivity and light use efficiency

The amount of dry matter produced by a plant stand is linearly related to the amount of light energy intercepted by the foliage canopy (Cannell, 1989). As a whole, the aboveground biomass productivity and light use efficiency were in order of pattern Ⅰ > patternⅡ > pattern Ⅳ > pattern Ⅲ > pattern Ⅴ > pattern Ⅵ > CK (agricultural monoculture) for different land use way throughout the first 7 years. It seemed the more dense the spacing of poplar was, the greater the biomass productivity and light use efficiency were, For example, compared with the CK at the fourth year (Tab. 6), the increase in biomass productivity and light use efficiency varied from 15.1% to 38.7% and from 18.8% to 43.8% respectively, depending on the poplar spacings in spite of the fact that the aboveground biomass productivity of wheat was lowered by an average of 11.8% in the poplar-crop interplantations. However compared to the old poplar-crop interplantation with spacing being 4 m×10 m, in which the productivity and light use efficiency were 24.1 t·hm^-2 and 0.93% respectively (Cao et al., 1997), the above ground biomass productivity and light use efficiency in these new interplantations were about 50% lower at the fourth year. The results discussed above indicated that poplar-crop interplantations can make full use of solar energy and growth space above the ground, and on the other hand, the root systems of poplar grown in the soil with a depth range of 0.6~1.0 m and a horizontal range of 4.0~5.0 m at age 4 (Fang et al., 1997), which can also use the soil resources fully. Therefore, from the view of improving land productivity, poplar-crop interplantation management system was much better than the agriculture monoculture system.

3.4 Economic assessment in different poplar-crop interplanting patterns

The poplar-crop interplantation is a traditional agroforestry practice, which is relatively simple but produce diverse outputs. Apparently the vertical rather than horizontal arrangement of production makes more effective use of a limited area of land. The most important factors impacting on relative profitability of agroforestry management system are current prices of crops and wood. With the entering of the WTO (World Trade Organization) and the implementation of "Natural Forest Protection Project" in China, the current price of wheat is down to 1.5 yuan·kg^-1, and the price of poplar wood is up to 450 yuan·m^-3 (averaged in different sizes). As shown in Tab. 7, the profits of the interplantations per unit were 14% to 49% more than that of the agricultural monoculture. It was easy to find that in the first 7 years, the most economic profits in poplar-crop interplantation management systems came from the wheat production. However, with the increasing of poplar stand age, the wheat yield dropped gradually year by year and the profits of the crop became smaller and smaller in the interplantation. For example, in the first four years, the wheat yield was about 4.4 t·hm^-2a^-1 in the interplanting pattern Ⅰ (Fang et al., 1997), but only 4.0 t·hm^-2a^-1 in the first seven years. On the contrary, the volume of poplar will increase rapidly with the increase of its age and the proportion of profits produced by poplar on the total profits also rise. For instance, in the first four years, the poplar wooer production was 5.38 m³·hm^-2a^-1 in the interplanting pattern Ⅰ (Fang et al., 1997), but reached 10.46 m³·hm^-2a^-1 in the fast seven years. Cao el al (1997) concluding that economic benefits of forest-crop complex was related to poplar clone, crop species, planting density and tree ages. Tab. 7 also indicated that the economic benefits in the poplar-crop interplantations were closely related to the interplanting patterns (planting density). Compared with CK, pattern Ⅰ showed the highest economic benefits, and pattern Ⅵ did not show poplar obvious benefits, suggesting that within a certain planting density of poplars, the denser the planting density of trees was, the higher the economic benefits were.

4 Conclusions

This research was initiated to determine the difference of six poplar-crop interplanting patterns on ecological benefits, site productivity and economic benefits. Based upon this research, the following conclusions are offered.

(1) Poplar-crop interplantations modify the microclimate, by decreasing solar radiation, lowering air temperature, and enhancing relative humidity. Therefore, Poplar-crop interplantations have a great impact on plant morphology, wheat yields and crop quality.

(2) Variance analysis showed that there were significant differences (p≤0.05) in wheat yield and 1 000 grain weight at the various sampling positions, and normal curve best described the distributions of wheat yield and 1 000 grain weight at different sampling positions. The most dramatic impact on wheat quality between poplar-crop interplantation and agricultural monoculture (CK) was the content of protein.

(3) The aboveground biomass productivity, light use efficiency and economic benefits were in order of pattern Ⅰ > pattern Ⅱ≥pattern Ⅳ > pattern Ⅲ > pattern Ⅴ > pattern Ⅵ > CK (agricultural monoculture) for different land use way throughout the first 7 years, indicating that poplar-crop interplantations can make full use of solar energy and growth space above the ground, and soil resources. From the view of improving land productivity and profitability, poplar-crop interplantation management system was much superior to the agriculture monoculture system.

(4) Generally, the rotation of poplars in the southern area of China is about 10~12 years, thereby the further studies are required in order to determine the effects of these new poplar-crop interplanting patterns on the ecological benefits, site productivity and economic benefits. However, on the basis of this research, we suggest that some of six new poplar-crop interplanting patterns, such as pattern Ⅰ, Ⅱ and Ⅳ, can be applied in such agricultural areas as the southern plain in North China, especially in the old course of the Yellow River, the Jianghan, Dongting Lake, and Poyang Lake plains, and in the lower and middle reaches of the Yangtze River.

Acknowledgements

We acknowledge the financial support of the Foundation of Jiangsu Province for Research Science and Technology, and Chinese National Foundation for Research Science and Technology. Numerous people contributed to this research, including Dr. Shixin Lu, Mr. Louzhong Tang, Mr. Guibin Wang, Mr. Zhaoyuan Song and Xuefu Ding.