4. Advances in poplar nursery techniques and planting.

Background: tradtionally poplar plantations are established using a track-type heavy-duty tractor (of at least 60 HP) equipped with a trenching plough. In the trench (25-30 cm deep and about 1.50 m wide) planting pits (0.50 m in diameter and 0.30-0.40 m deep) are dug manually, 15 cm deep. One tractor with a trencher requieres at least 20 people to keep up with the pit-digging. Rooted one-year planting stock is used. The trench is left open (supposedly to collect rainwater, or to catch wind-blown sand that covers the plants and avoids dehydratation, but due to its' opening-up, further dehydratation of the soil seems likely) but closes up gradually due to sand that gets trapped in the trench. This method of semi-mechanised tree planting has been used since the middle of 1970's. The tree-planting season is in Spring from middle March to middle April. The afforestation efficiency is 2800 saplings per day. Survival rate depends on rain fall, normally 60% to 70%, but much less in dry years. If the planting is for productive plantations irrigation is provided, normally four times during the first yer of establishment.

The Project has been advocating the use of poplar deep-planting techniques, using unrooted stock, to replace the traditional planting techniques. During the 12 years of the project, 67 comparative afforestation trials on poplar deep-planting were established, covering an area of almost 47 ha, and a total of 128 ha of pilot plantations.

Deep-planting of un-rooted cuttings offers several important advantages:

• planting stock used in deep-planting (unrooted cuttings) is harvested on stoolbeds in nursery that stay in production for up to seven years

• deep-planting puts the planting stock deeper in the soil, and thus in more humid conditions

• as no trench or pit is opened, soil does not dry out

Results show that deep-planting returns survival rates of 90%, after four years of continuing drought conditions in the region (200 mm of annual precipitation). This is 20% better than the survival obtained with the traditional technique. Furthermore, initial growth increases 81% compared to the traditional technique. These results are obtained without applying irrigation, while traditional technique irrigates up to four times after planting.

The technique, suitable for poplars and willows) is now rapidly being accepted and has been used outside the Project areas in Liaoning Province, Jilin Province and Inner Mongolia.

The deep-planting technique is not suitable for clay and saline or alkaline soils, but very suited in sandy soils.

Achievements by the Project

• Introduction of poplar deep-planting techniques, and improvement in survival and growth, without application of irrigation, even under extreme drought conditions

• Establishment of Permanent Sapling Stands of selected clones, capable of providing quality planting stock over up to 7-8 years, before re-establishement of the nursery is needed

• Fine-tuning of treatments for un-rooted stock used in deep-planting, allowing both spring and autumn planting, and determination of optimum planting depth

• Development of a stock-production and planting system suitable for mechanisation (see chapter 5 "Advances in Mechanised Afforestation")

4. 1 Poplar nursery techniques

Quality nursery stock is a must for high survival rates and good growth after planting.

Traditional poplar nursery practice

Uses one-year-old poplar twigs, to produce in autumn short cuttings (12-13cm long and 0.7-1.0 cm thick at the thin end); these are tied in bundles and buried; the next spring the short cuttings are put in a flat seedbed at a spacing of 70 cm x 8 cm (175 000 units per ha). In the autumn of the same year, after dormancy setles in, the one-year old tree (of about 1.4 m tall and 1.0 cm in diameter at the collar) are dug out and out-planted in the field.

New poplar nursery pratices

Stoolbeds are established on flat, fertile (sandy-loamy texture, pH around 7), well-drained sites with availability of irrigation, and protected from strong winds by shelterbelts. The field is to be ploughed at a depth of 25-30 cm, a year ahead of establishment, followed by disc-harrowing. During preparation, 40 T of natural manure together with 150 kg of phosphate is applied per hectare; 225 kg per ha of nitrogenous fertilizer (urea) should be applied twice yearly. With the growth of the stock, these amounts are to be increased gradually. Apart from this, from the third year of harvesting onwards, 15 T of natural manure should be plowed in between the rows, just after thawn in early Spring..

Selected poplar (or willow) clones are established in the nursery stool beds by short cuttings (15 cm long) at densities of between 30 cm x 80 cm or 40 cm x 100 cm (from 25000 to 41000 per ha). It is important to strictly maintain the planting distances, as otherwise the stock-production will result heterogeneous in size. Different clones should be planted in different blocks to avoid confusion. At the end of the first year, the nursery stock probably will be still too weak to be used in afforestation, but from the second year onwards the unrooted cuttings may be harvested yearly in autumn for deep-planting. The stumps and roots are left in the stoolbed, to produce new saplings next spring. The procedure will be repeated over and over again and the stoolbed can be used for 6 to 8 years.

In order to maintain over the years the quality of the nursery stock, the soil fertility of the stoolbeds must be maintained. Application of fertilisation is necessary after removal of the harvest in autumn, to delay the declining fertility of the soil caused by frequent cutting.

The preparation of the un-rooted cuttings requires special care and attention to assure that cutting edges are neat and that the bark is not thorn off.

Picture 8 Stoolbed

Stock-treatments for deep-planting technique

Stock treatments refer to storage of cuttings, age of cuttings (one or two-year old), Spring or Autumn planting, and soaking of cuttings prior planting. It has been demonstrated that the right combination of treatments is essential for good survival and growth after planting.

- Cuttings storage methods:

Cuttings used for Spring-planting should be harvested in Autumn and stored for use in Spring.

The results obtained by the Project suggest that the most economical way to do this is to bury the cuttings in a horizontal position 0.5 m deep beneath the ground. (Among the other tried storage methods were: deep-burying in horizontal position well below 1m deep, at 1m50 and at 1m; burying in vertical position with bigger diameter down and covered with 0.50 m of soil; same as before, but with tops downwards pointing.

- Soaking of cuttings before planting:

Several trials have bben established to test the influence of soaking and of the duration of soaking Results show that for autumn planting cuttings should be soaked for 24 hours in water. For Spring planting, cuttings should be buried over winter (see section above), but not necessarily soaked. However soaking for 24 hours in water is recommended also.

- Age of cuttings:

In general it can be stated that 2-year old cuttings (N+2) are performing slightly better than 1-year old cuttings (survival rate of respectively 85% and 78%), maintaining all other parameters equal.

- Use of basal and upper part of long cuttings:

An un-rooted sapling of good quality, can be cut into two viable cuttings (basal and upper part), at the condition that the diameter at the small end is no less than 0.5 cm. This multiplies the production of the stoolbeds by a factor of 2, and reduces stock costs almost by half.

Nursery stock (long cuttings), produced on quality stoolbeds of age N+1 or N+2, reaches about 3m high, while the lenght of cuttings to be used in MDP-planting is 80cm. Therefore, in order to reduce afforestation costs, a long cutting can produce two 80cm long cuttings for MDP- planting, discarting the thinner and under-lignified top-end. Several trials have been set up in Naiman and Tongyu in the autumn of 1997, to test the viability of the use of basal and upper parts of the cuttings in MDP-planting.

Results of the trials show that use of the lower part or the upper part of the saplings has little effect on the survival rate if the saplings are well-lignified ones (survival rates in both cases well over 90%). That is, an un-rooted sapling can be cut into two 80cm cuttings for the use of planting with MDP technique. However, the thinner-end diameter of cuttings should be over 0.5cm, and 1.0cm is the best.

- Spring and Autumn planting:

Cuttings, are harvested form the stoolbeds in the nursery in autumn and should be soaked for 24 hours in water before planting in the same autumn. Cuttings intended to be used for spring planting, should be buried in wet sand in order to protect against dehydration over winter.

For deep-planting only the use of clones (poplar and willow) with a good rooting capacity are recommended, such as: P. Heilin #1; P. nigra × P. deltoides cv. "Shanhaiguan"; P. nigra × P. pyramidalis cv. Bailin #2; P. xiaozhuannica cv. Faku #1; P. simonii × P.nigra #14.

4. 2 Deep-planting techniques

If the groundwater table of the site to be planted is at 2.5-5 m, the planting depth should be 1.2 to 1.4 m. If the groundwater table is lower, the planting depth should increase accordingly.

During establishment of plantations using deep-planting techniques in dry zones, the above surface parts of the cuttings should be cut down at the soil surface to leave as little as possible of the planted cutting exposed to dehydratation.

The Project has developed two kinds of deep-planting techniques, one based on the auger-principle and one based on the sub-soiling principle. Both techniques have been mechanised, and each has its own merits and application fields (see also Chapter 5 "Advances in Mechanised Afforestation"). The first technique is called "Mechanised Auger Planting" or MAP, and the latter is called "Medium Depth Planting", or MDP.

Mechanised Auger Planting

By mechanical means (an auger driven by an engine, and may be hand-held, tractor- or trailer-mounted; the Project has been experimenting with all possible systems) a hole (10 - 15 cm in diameter) is opened in the soil up to the requiered depth.

Trial results on auger depth established and evaluated over the years, have shown that under normal circunstances the depth of augering has but little effect on survival rates (97-100%) of deep-planting, as long as depth is 1m20 or over (trials with augering up to 3m have been established). When the augering depth is in the range of 0.7 - 1.2 m, the positive correlation between depth of planting and survival becomes obvious. For example, in trial 1NM1995AF08, the survival-rate of trees established at an augering depth of 0.7 m and 1.2m is 52 and 88 % respecitively.

However, growth in tree height and diameter is strongly related to augering-depth: for example, height and diameter growth of six-year old trees, established by augering at 3m deep, was 42 % and 45 % superior to that of trees of the same age established by augering at 1m20.

Based on the above results, the Project developed a mechanised trailer-mounted auger, that operates up to depths of 1m30. A un-rooted cutting, of about 1 m 40 long is introduced in the hole, and the soil is compacted around the cutting. Good soil-compaction is of atmost importance, as the existence of underground air-pockets will favour the drying-out of the cuttings and impede root-formation.

Cuttings for MAP planting are normally cut as long as the depth of the augered hole (N+1 or N+2, unrooted), so that after planting the cutting is (almost) completely covered with soil. When longer cuttings are planted, the above-ground part should be pruned-off at ground level, to avoid dehydratation of the cutting. This has a dramatic effect on survival rates: see Table 4.2.1.

Also, it is essential to assure a good compaction of the soil around the planted cuttings, as air-pockets will impede root-formation and increase dehydratation.

Table 4.2.1 : Comparison of survival rates fo MAP planting of un-rooted cuttings:
pruned above groud level and left standing

Location	Planting time	Clone	Age stock	Type of cuttings	Survival rate (%)
Tongyu	October 1996(normal rainfall)	P. xiaohei x P. p15A cl.	5+2	- cut off above ground - long, not cut-off	100 56
Tongyu	October 1998(dry after very wet summer)	P. xiaohei x P. p15A cl.	6+2	- cut off above ground - long, not cut-off	90 15
Tongyu	October 1999(dry year)	P. xiaohei x P. p15A cl.	5+2	- cut off above ground - long, not cut-off	90 11
			2+1	- cut off above ground - long, not cut-off	60 2
Naiman	October 2000(dry year)	P.xiaozhuannica cv. Faku #1	5+1	- cut off above ground - long, not cut-off	57 5

Advantages of the MAP:

• Trailer-mounted MAP can be used with low-powered tractors, accessible to the local farmer house-holds

• little disturbance is caused to the existing vegetation, so no increase in wind-erosion is caused

• existing humidity in the soil is preserved as only a small hole is openend and immediately covered again

• good survival rates and growth, even under unfavorable circunstances (extreme drought)

• easy operatiopn by a 3-man crew (tractor operator, auger operator, planter-compacter)

• planting speeds of up to 100-120 holes per hour

• High survival rates with clones that have no rooting difficulties, such as Heilin #1 and Baicheng #4 (survival nearly 100%).

Photo 9: MAP

Medium Depth Planter

This machine is based on a modified sub-soiler blade (90 cm tall and 6 cm wide ), connected to the three-point linkage of a 80 HP track-type tractor (Dongfanghong 802), and provided with operator seats, storage space for cuttings and compaction wheels. The sub-soiler blade is lowered into the soil and a 5cm wide and 80cm deep furrow is made. The planter siting on the machine, sticks N+1 or N+2 un-rooted cuttings (2-3 cm thick and 80 cm long) into the loosened soil, immediately after passage of the sub-soiler blade. The the packing wheel compact the soil again around the planted cutting.

Advantages of the MDP:

• very fast plantation establishment: one machine with 1 tractor operator and 2 planters can plant up to a maximum of 1000-1500 trees per hour, with an average of 800-1000, at a cost of 0.05RMB per tree, compared to a cost of traditional trench plantation is

0.21RMB per tree, and by Mechanized Auger Planter of 0.42 RMB per tree, including machine and labour costs), excluding stock.

• Good conservation of soil humidity and high survival rates (normally over 95 %).

• Medium Depth Planting is easy to operate and is very efficient.

Photo 10: MDP

Cuttings for MDP planting are normally 80 cm long (N+1 or N+2, unrooted), so that after planting the cutting is (almost) completely covered with soil. When longer cuttings are planted, the above-ground part should be pruned-off at ground level, to avoid dehydratation of the cutting. This has a dramatic effect on survival rates: see Table 4.2.1 (for MAP planting, similar results were obtained for MDP)

Also, it is essential to assure a good compaction of the soil around the planted cuttings, as air-pockets will impede root-formation and increase dehydratation.

4.3. Costs

The new techniques introduced and fine-tuned by the Project not only assure better survival and growth, but show also lower costs.

Nursery production costs

The following analysis compares quality (Table 4.3.1), costs and revenue for nursery production and afforestation for traditional planting and MDP (Table 4.3.2).

Table 4.3.1 : Qualiy Comparison of Improved and Traditional Nursery Production

Class I: D₀=2.2cm; H=250cm Class II: D₀=2.0cm; H=200cm Class III: D₀2.0cm; H=200cm

D₀ collar-diameter D_1.3 diameter at 1.3 m H total heigt

Table 4.3.2 : Revenue generated by nursery stock in improved vs. traditional production
(in RMB Yuan/ha)

Table 4.3.2 is calculated based on the following retail prices:

Retail price for 1+1 rooted sapling (for traditional shelterbelt est) 0.15 yuan/tree

Retail price for 2+2 rooted sapling (for trad. productive pit-planting) 0.60 yuan/tree

Retail price for N+2 un-rooted cutting: 0.60 yuan/tree

Retail price for 2+1 un-rooted cuting: 0.40 yuan/tree

Retail price for N+1(with N=3) un-rooted cutting: 0.60 yuan/tree

Using upper and basal part of un-rooted cuttings, prices per unit are halved, i.e. for N+2 stock 0.30 yuan per tree

From Tables 4.3.1 it is clear that stock-quality from stoolbeds is better than that from traditional nurseries. The number of Class I stock in 2+1 and N+1 production is 6.5 and 7.6 times higher as that by traditional way, and in 2+2 and N+2 production it is 7.3 and 8.1 times higher as that by traditional production. The older the stoolbed becomes, the higher the diameters at collar-level and at 1.3 m and the height of the saplings also increases.

The benefits produced from 1 ha by N+2 resp. N+1 are twice to four times that of traditional production (Table 4.3.2).

Afforestation Costs

The following tables compare afforestation costs by MAP, MDP and by traditional planting for the establishment of productive plantations and shelterbelts.

Table 4.3.3 : Comparison of unitary afforestation costs
(in Yuan RMB)

Planting mode	Transport/handling (1)	Stock cost (age)	Planting cost	Total
MDP	0.06	0.30 (N+2)	0.05	0.41
MAP	0.18	0.60 (N+2)	0.42	1.20
Trad. large pit-planting (2)	0.25	0.60 (2+2)	0.51	1.36
Trad. mechanised planting.	0.12	0.15 (1+1)	0.21	0.48

(1) Including bundling, loading, un-loading, field storage, soaking.

(2) Manual digging of very large pits (80 cm x 80 cm x 80 cm), and planting of 2+2 rooted stock (cost similar to N+2 unrooted stock used for MAP planting, i.e. 0.60 yuan RMB).

Table 4.3.4 : Comparison of afforestation costs for productive plantations (per ha)

(1): in traditional planting for productive plantations 4 irrigations are applied; in MAP planting none.

(2): manual digging of very big pits (80 cm x 80 cm x 80 cm), and planting of 2+2 rooted stock (cost similar to N+2 unrooted stock used for MAP planting, i.e. 0.60 yuan RMB).

Table 4.3.5 : Comparisson of afforestation costs for shelterbelt establishment (per ha)

n.a.: not applicable

The numbers in the tables are self-explaining. Costs for establishment shelterbelts by MDP in comparison with traditional technique are 8% lower, but MDP is faster and gives a survival rate 20% higher than with the traditional technique. MDP technique is fast, does less damage to the existing vegetation and preserves soil moisture far better. MDP should be employed for the establishment of large-scale shelterbelt and protection forest in semi-shifting sandy lands.

The cost for productive plantation establishment are even more convincing for the use of MAP (less than half) in comparison with the traditional large pit-planting. This is mainly due to the elimination of the need for irrigation with MAP planting. Survival and growth (diameter and heigth) are all superior and this is even more so under difficult circunstances, such as the recent 3-year long drougth.

5. Advances in Mechanised Afforestation

Background: One of the main goals of the project has been to develop effective and efficient means for mechanised afforestation, as the region is relatively sparcely populated and the peak of field-operations in afforestation coincides with those in agriculture. At the beginning of the Project and all over the First Phase different kinds of equipment for land preparation, agrcultural transport, deep-planting augers and nursery and irrigation equipment have been purchased locally and from abroad, together with a series of prime movers, and equipment for the installation of 3 complete workshops, including milling machines.

Achievements by the Project:

• Development of a trailered mechanised auger planter (MAP), suitable for operating with a small tractor (18HP)

• Development of the Medium Depth Planter (MDP)

• Development of a mechanised undercutter for pine nurseries

5. 1. Development of the Mechanised Auger Planter Characteristics of the imported augers

The trailer-mounted auger imported from Australia (KimSeed Type T290F) had to be pulled by a John Deer 4450 tractor, and has a depth reach of up to 3.0 meters. The diameter of the hole is 120 mm to 200 mm by mounting different augers. The auger's rotation and up and down movement is driven by two hydraulic motors. A two-stage hydraulic pump supplies the pressure in the hydraulic system. The pump is driven by the tractor's PTO shaft. The original purpose of the auger is for post establisment.

The Mobile Digger General (550 DIG-R-Mobile) on a self-contained trailer, imported from the USA, has a maximum depth of 1.7 meters. This machine has an 8 HP four stroke petrol engine to drive auger at a rotation speed of about 175 rpm. The diameter of the hole can be changed from 70 mm to 200 mm by using different augers. Two men are needed to operate the up- and down movement of the auger by rotating a manifold wheel that is engaged with a chain transfer system. The whole machine is trailer-mounted and can be trailed by any vehicle (no PTO needed).

The Teson Cross B Hole Digger was imported from Italy, and is mounted on the three-point linkage of a 60 HP tractor; planting depth is up to 60 cm, and various auger diameters can be mounted. The tractor PTO drives the auger rotation by serial shaft and gear transfer. The up-and down movement of the auger is carried out by the operating tractor three-point linkage.

A Chinese-made earth auger TW-60 has similar structure and planting depth as the Italian one. The diameter of drilled hole is 300 mm to 400 mm by mounting different size of augers.

The main limitations of these imported augers are the high purchase costs, the access to spare parts and the sometimes cumbersome operation (American auger). Above all, most require powerfull tractors to operate, increasing operation costs and making the technique less accesible for farmer households.

Development of a Deep Planting Mechanised Auger

Reaerch results showed that a planting depth of 1m40 for poplar deep planting is giving good results and survival rates of up to 98%. A new prototype earth auger was designed and developed to allow planting of un-rooted cuttings at this depth.

The machine development went through 3 stages.

At the beginning, the first prototype had the auger side-mounted to an 18 Hp diesel engined tractor. This configuration made it easier for the tractor operator to control the working of the auger and operate the machine by pushing or pulling the hydraulic control valve levers. A hydraulic motor drives the auger at about 200 rpm and a second hydraulic motor moves the auger up and down. The hydraulic system is powered by a hydraulic pump, driven by the tractor's PTO shaft. Altough operating satisfactorily, it was difficult to dismount (or mount again) the auger when the tractor was requiered for other purposes.

The second prototype was a modification of the first. In order to solve the mounting and dismounting difficulty, the side-mounted auger was adapted to be mounted on the tractor's three-point linkage. The operator now had the inconvenience in turning and looking over his shoulder while operating the auger. Also, the auger seemed too heavy for rear-mounting on a small tractor.

(Plate 2 of Appendix 2 shows the second prototype)

The third prototype is a trailer-mounted auger. All components and parts of the auger are mounted on a trailer, and connected with a small sized tractor (18 HP) with PTO. Compared to the two previous prototypes, it can auger faster and is easier to operate and to mount and dismount to the tractor. The tractor supplies the power to the auger's hydraulic system by PTO and pulls the auger to its location. Two hydraulic pumps are used to supply pressure for auger rotation driving motor and for the up-and down movement (by chain-sprocket transfer) individually. This avoids both activities influencing each other. The hydraulic control valve is installed on the trailer.

(Plate 3 of Appendix 2 shows the third prototype).

5. 2. Development of the Medium depth planter

The aim was to develop a machine capable of large-scale afforestation and shelterbelt establishment for poplar (and willow), by planting un-rooted cuttings.

The departing idea was to use a ripper to loosen the soil in a narrow furrow up to a certain depth, and then insert manually the unrooted stock into the loosened soil.

To find the right ripper several equipments were tried out:

• A Single Tooth Ripper. Maximum usuable depth of the tooth is 0.9 to 1.0 metre, and it is 70 mm thick.

• A three teeth subsoiler also made by the Tongliao Forestry Machine Factory. The two side teeth were taken off to leave only the middle one as a single tooth ripper. The maximum working depth is 0.9 metres, and the tooth is 40 mm thick.

• A three teeth subsoiler, imported from the UK, that was treated similarly as the above. The tooth was 24 mm thick with a usuable working depth of 0.8 metres.

A limitation was the capacity of the available prime-movers: maximum power was a 80 HP diesel crawler tractor (DFH-802). The Single Tooth Ripper was not able to operate at the requiered depth with the available power (maximum depth was about 60 cm). Applying more power (by putting another tractor in front of the installed one), the ripper frame was bended. The local-madse subsoiler worked well up to a depth of 70 cm. The final choice was the UK-made subsoiler, that due to its lesser width had less resistance in the soil, and worked well at 80 cm deep.

The machine is mounted on a 70 HP crawler tractor by three-point linkage. The subsoiler is fixed at the center rear of the machine. The maximum working depth of the subsoiler to loosen up the soil is 80 cm. Three people are needed for tree planting. A pair of compacting wheels are hinged on the back of the machine to compress the soil around the planted cuttings after planting. The planting efficiency is between 800 to 1500 trees per hour, depending on site conditions.

The first prototype of the MDP had the poplar saplings inserted into the loosed soil by two crews, sitting one in front of the other on the machine: the sapling is inserted half-way into the soil by the first crew and insertion is then completed by the latter. Another crewmember hands the sapling to the first crew. Two main problems were detected with this first prototype:

• it is difficult and hard work to insert the poplar saplings 80 cm deep in the loosed soil when the tractor moves fast in first gear.

• the machine was not able to plant in areas with existing vegetation or debris, because of vegetation/debris piling up in front of the subsoiler.

On the second prototype, to overcome these main problems, two devices were adapted:

• two back-pointing metal wings (80 cm long and 30 cm wide; separation at widest end about 15 cm) were welded on the rear edge of the subsoiler blade. These wings keep the soil "open" after passage of the subsoiler to allow the operator to drop a sapling in, before reclosure.

• a cutting blade made of hardened steel sheet (2.5 mm thick and about 150 mm wide) was bolted in front of the subsoiler under an angle of about 40 degrees. The angeled cutting blade will push aside and eventually cut roots and accumulated vegetation/debris, allowing free passage of the subsoiler.

Additionally, several other modifications were done:

• Two beams were attached to the machine towards the rear, on which the operators' seats are placed (changed from the front of frame to the rear) and at their end the mounting of the packing wheels.

• The seats for the operators that insert the cuttings into the soil are mounted on the beams by bolts, allowing adjustement according to operators' leg's length.

• The rubber tired compacting wheels were replaced by steel wheels to increase compaction.

• A distance tracking wheel was mounted to the left of the operators, to allow for easier determination of spacing. The wheel rings a bell every half meter.

• A plant counting device was installed at one of the operator's right hand. It not only allows the operator to register the number of trees planted, but also to support himself by the handle.

• To allow that the machine can operate and move along according to the ground surface, two depth limitation wheels were mounted on both side of the machine-frame.

With these modification in place, the machine is perfectly suitable for poplar planting in sandy areas. Average survival rates are around 80 %.

6. Advances in Agro-forestry

Land degradation results from a combination of inappropriate and over-use of the soil, and of over-grazing, in an attempt to produce more than the site can produce in a sustainable way. By introducing alternative crops, particularly cash crops, or other methods of generating income that provide incentives for the further stabilization of unproductive land, the reliance on unsustainable practices can be reduced. One way to promote diversification of rural production is the integration between forestry, agriculture and animal-husbandry on the same or adyacent areas that are interchanged over time.

Models using shrubs as fences and hedgerows around and in pastureland, in combiantion with shelterbelts, to reduce wind erosion and as a means to implement controlled rotational grazing are under development. Stubble-management of agricultural crops plays an important role in the reduction of wind erosion. New fodder species have been introduced and tried out. Introduction of reduced tillage improves soil-fertility and reduces wind-erosion. Local herdsmen and farmers have been consulted and involved in the design, planning, preparation and implementation of the agroforestry models.

Achievements of the Project:

• introduction of stubble-management

• introduction of reduced tillage

• implementation of hedgerows

• fodder species trials

• trials on fencing-off

• sprinkler irrigation trials in sandy land

• reduction of wind erosion in experimental area by 3/4 compared with surrounding land

6. 1 Description

The site

The main agro-forestry demonstration area is situated 13 km to the northwest of the Village of Molimiao (Fuju Country, Tongliao). The area is an approximate square of 1400m long, and an area of 80 ha. The eastern part (33 ha) is comparatively flat and homogeneous; the central part (20 ha) is undulating sandy dune-land and the western part (27 ha) is flat pasture land. Due to over-grazing and over-cultivation four blow-out areas have been formed in the eastern part (with an average erosion depth of 4.1 cm, registered between March and May 99, before fencing off). During a field check from March to May, 1999, it was found that over 80% of the total area was severely affected by wind erosion and shifting sands.

The underground watertable is shallow, at 2.6-3.5m in the flat areas. The soil is sandy, in a layer between 30 to70 cm thick. Underneath, the sandy silt alluvial deposits by the West Liao River, have a relatively better nutrient content: organic matter 0.3-1.3%; total N: 0.04-0.15%, total P₂O₅ : 0.03-0.18 %; total K₂O : 2%.

Originally, vegetation was sandy grassland. The composition of vegetation is reduced to a few species, the plants are short and small and the quantity of palatable plants is very reduced. Non-palatable species, such as Tribulus terrestris, Xanthium stumarium, Messerachmidia sibirica have greatly increased their presence. The main plant community contains Cleistogenes chinensis, Atraphaxis manshurica, Trigonella doushinsdyi, Aneurdepidium dasystachys, Bassia dasypyyla and Artemisia siversiana.Inside the fenced-off area, the grassland vegetation is regenerating naturally, and diversity is increasing.

Division of the area and management

The area was divided in three parts according to the soil types and the demonstration objectives, as follows.

The eastern area was retained for development of sustainable agriculture trials: 4 plots are reserved for trials under direct supervision by the Project and 11 plots are farmed by peasant households, in co-ordination with the Project. Farmers that cultivated these plots are encouraged to apply technical recommendations given by the Project, such as use of reduced tillage techniques and stubble-management.

The central area is reserved for sand-fixation and afforestation: forest belts and plantations in block have been established for sand-fixation, with poplar, Pinus sylvestris var. mongolica and shrubs, such as Lespedeza.

The western area is reserved for hedge-row agriculture: shrub hedges are planted at spacings of 30-40 m, and in between agriculture or fodder production.. Some of these fields are left for the natural growth of grasses and herbs and shrubs in order to regain soil fertility and harvest fodder.

Picture 1 Subdivision of the demonstration area of mixed plantation

The management of the area is aimed at integrating rural production, bringing together agriculture, animal husbandry and forestry, implemented in an as sustainable way as possible.

The main objectives are to provide income for the peasant households, to stabilise and revegetate shifting sands and reduce wind-erosion.

Involving the peasant households is essential as this assures feed-back to the Project from the direct stakeholders and the households get a hands-on training and demonstration of the newly introduced practices: 95% of the area is farmed directly by peasants.

The first step in the process is the establishment of a grid of shelterbelts, at a spacing of 200 m (N-S) x 100 m (W-E) apart, and fencing off of the whole area..

Basic agronomic measures taken are:

• crop rotation

• appropriate crops (f.i. no beans!), and if possible crops with an added value to increase economic benefit for the farmer (such as millet, special varieties of maize, some kinds of gourds and melon for seed production)

• zero or reduced tillage (as little soil disturbance as possible)

• stubble management: keeping the stubbles of the crops on the field over winter reduces erosion greatly (instead of the traditionnally applied method of leaving the fields "clean" for winter)

• additional techniques: late sowing, proper manuring and fertilizing, sprinkle irrigation

• hedgerows (in the western area): provide additional protection against wind-erosion, provide fodder and contribute to soil fertility

Some comparative trials of integrated land-management including forestry and farming, or forestry and fodder production were also established at Tongyu and Naiman. This has further enriched experimental results and extended the divulgation of the techniques.

6. 2 Results

First, it should be reminded again that the years 1999 through present were exceptionally dry years, and that the 2001-2002 winter was especially harsh. This has greatly affected progress.

Fencing off

Tthe restoration of natural vegetation is progressing very well. The example set by this relative green area in the middle of sand-blown fields has already inspired local farmers to start fencing off their own lands.

Stubble-management

Stubble-management that a few years ago was unheard off in this area, is now general practice. However, fields that had crops that leave lignified stubble such as maize, wheat, millet, and legumes, have to be cleared before sowing in spring.

Reduced tillage

This is a more complicated issue, as farmers do not have ready access to chemicals for weed eradication or direct sowing machines. Merely technically speaking, results are excellent.

Sprinkle irrigation

Same observation, but to a lesser extend, as above. Some farmers in the area have invested in pumps and sprinklers.

Sand-fixation, afforestation and shelterbelt establishment in farmland

In the central part 20 ha of semi-fixed sandy dunes have been planted with belts and plantations in block of Pinus sylvestris var. Mongolica(3 ha), poplar(4 ha), Lespedeza shrubs (2 ha) and grasses (10 ha). In the west, where land degeneration is very severe, alternating belts for agriculture, for fodder production and for grasses are established, separated by shrub-hedgerows. These alternating belts should be managed with crop rotation every 6 or 7 years: crops should be replaced by grasses, fodder by crops and grasses by fodder.

Introduction of fodder species

In 1998 14 kinds of fodder seeds were imported from Australia and sown experimentally in 2000. 13 species emerged. After winter, five species are surviving: Astragalus adsurgens, Melilotus officinalis, Medicago sativa, Pisum sativum, Zea mays and Secale. Additionally, provenances from Medicago sp. were imported in 2001 from Canada and are now being established.

Re-establishment of soil coverage

Soil coverage, constituted by one or several layers of vegetation (grasses, shrubs and trees) is the main measure to check the expansion of desertification in the Sandy Lands. Fenced-off areas have a remarkable fast rate of recovery, indicating the high resilience of the natural vegetation under aparently adverse conditions. The number of species increase sharply, plant associations become obvious, vegetation growth increases and wind erosion is controlled effectively.

In order to monitor soil erosion and plant differentiation inside and outside the fenced-off demonstration area, ten observation plots of 2m by 2 m were established on different sites in the demonstration area: 3 were set up in an area of 35 ha of farmland, 2 on the top of undulating sandy dunes, 2 more in waste land to the west of the demonstration area, and three control plots were established outside the fenced-off area, for comparison. Inside the plots, the number of individual plants, height, and cover rate of every plant were observed and recorded.

Results show that after fencing-off the soil coverage increased from 50% to 92%, the average height increased from 10 to 50 cm and the grass biomass yield increased from 0.25 kg/m² to 0.60 kg/m². In the demonstration area not only the amount and growth of plant species increased but also the regenration. Trigonella korshinskyi and Lespedeza hedysaroides increased in coverage from 30% to 70% on even undulating land.

When the area was turned over to the Project in 1998, before fencing-off, the vegetation existing in the eastern part, had a soil coverage of under 50 %, with short and small plants, due to (over-)grazing. Average height was 10 cm; main species are Atraphaxis manshurica, Artemisia sp., Chloris virgata, among others. In the western part, before fencing-off, plant heights were also low, with an average of 10 cm, and a soil coverage of 60%. Main species were Artemisia sp., Potentilla anserine, Setaria viridis, Digitaria sanguinalis, Lespedeza hedysaroides, Tribulus terrestris and others.

Presently, outside the fences, what is left are barren sands and moving sands. The amount of plant species outside the fences is being further reduced, and some species not-palatable for graziers increase. If the wind-erosion is continued un-checked, the plant cover will continue to reduce and even more serious desertification symptons will appear.

Table 6.2.1 illustrates the effects of fencing-off after 2 years.

Table 6.2.1 : Comparisson of results of fencing-off

However, inside the fenced-off area, the plant-community increased in growth and diversity. Leguminous plants such as Trigonella korshinskyi and Lespedeza hedysaroides recover rapidly when not over-grazed, and their coverage (60%) and frequencies (40%) become high.

After plowing these leguminous plants are replaced by mainly grasses (80%), such as Chloris virgata, Digitaria sanguinalis and Setaria viridis.

Since 2001, a grass species, harmful to agriculture and animal husbandry (Cenchrus calyculata), has invaded parts of the fenced-off area, mainly along the road-side, and rapidly expanding.

Reduction in wind-erosion and micro-climate improvement

It has been convincingly demonstrated that the integrated approach used inside the demonstation area has reduced wind-erosion significantly. The measures taken to reduce the impact of wind-erosion were: sowing of crops with resistant stubble and prohibition of red beans and buckwheat; maintaining stubble on the fields after harvest, avoidance of ploughing the land in autumn, late-sowing to avoid the strong winds of April and early May.

Measurments were taken in the evaluation plots on different sites from November 1999 to April 2000. Table 6.2.2 provides information on severity of wind-erosion under different kinds of cultivation.

Table 6.2.2 : Wind-erosion under different kinds of cultivation
(from November 1999 to April 2000)

Vegetatiuon/Cultivation type	Dunes with grasses	Dunes with burned grasses	Wheat with high-cut stubble	Maize with short-cut stubble	Maize with high-cut stubble	Red beans	Grazed grass-land	Enclosed grass-land (not grazed)
Depth of wind-erosion (cm)	0	2	0	1	0	7	1	0

From the table, the destructive effect of bean cultivation on wind-eroson is obvious. However, farmers prefer beans, above less destructive crops, due to tolerance to poor sites, easy cultivation, and security of income provided by this crop.

Table 6.2.3 indicates results of measurements of wind-speeds (at an height of 1m) on land were stubble was maintained in comparison with land were only cut-off stumps remained.

Table 6.2.3 : Windspeeds (at 1m above ground level) on two differently managed maiz fields (in m/s)
(observations done on April 15^th , 2000, at four diffent times of the day)

Keeping crop stubble not only protects the soil from wind erosion but also raises the temperature of the upper 30 cm by 0.3°C, and moisture content of the soil, between 30 and 60 cm deep is raised by 0.9% on average.

When a snow cover is present (as over January and February) the soil temperature effect of the crop stubble left standing, is not obvious, but in November and March without snow cover, it is. However, during the monitoring period, there was allways a difference in soil moisture content, but small.

Table 6.2.4 : Comparison of soil moisture content between land with wheat stubble left standing and land ploughed after harvest (in %)
(Observations over 1999-2000)

6. 3 Social and economic benefits Financial benefit gained by the peasant

Table 6.3.1 : Financial benefits obtained during operations by participating farmers
(Year 2000)

It is estimated that in the demonstration area (fenced-off) the crop production is 2-3 times higher than before and productive potential of the land may be up to 3 times higher than before. As an additional benefit, between 20 and 30 peasant have been familiarised with the newly introduced techniques.

Table 6.3.2 : Costs of fencing-off
(in Yuan RMB)

Item	Unit cost	Cost /1000 rm	Cost per ha (2)
Fence poles (concrete)	3.0 Yuan/rm (1)	3000
Fencing wire mesh	1.05 Yuan/rm	1050
Transport	1.0 Yuan/rm	1000
Labour	0.25 Yuan/rm	250
Total		5300	106

rm: running meter

(1): cost poles is 12 Yuan (without transport), placed at 4m intervals

Transport costs are 3 Yuan per pole, so 1 Yuan per rm

(2): Cost per ha is based on a conversion factor of 1 ha equivalent to 50 rm (Molimiao area is 80 ha, with 3700 rm of fences, giving a conversion factor of 46m to the ha)

Usefull lifetime of the fence is estimated at 10 years, with maintenance costs of 10% of initial cost yearly, giving a total cost of 11.66 Yuan per ha and per year.