A participatory simulation to facilitate farmers adoption of livestock feeding systems based on conservation agriculture in the uplands of northern vietnam

H ow ever, certain factors that ensu re the su ccess of the p articip atory m ethod Mosse, 1994 have to be taken into accou nt w hen d esign-ing a m ethod that allow s farm ers to sim u

A Pa rtic ip a tory Sim ula tion to Fa c ilita te Fa rm ers’ Ad op tion of

Livestoc k Feed ing System s Ba sed on Conserva tion Ag ric ulture in the Up la nd s of Northern Vietna m

Ced ric Martin1,2, Jean-Christop he Castella2,3, H oang Lan Anh2, Yann Egu ienta1,2,4 and Tran Trong H ieu2

1

Centre d e Coop e´ration Internationale en Recherche Agronomiqu e p ou r le De´velopp em ent (CIRAD), Av Agropolis, 34398 Montpellier Ced ex 5, France;2Mou ntain Agrarian Systems Program m e, Vietnam Agricultural Science Institute (VASI), Thanh Tri, H anoi, Vietnam;3Institut d e Recherche p ou r le De´velopp em ent (IRD), 213 rue Lafayette, 75480 Paris Ced ex 10, France, and International Rice Research Institute (IRRI), DAPO 7777, Metro Manila, Philip pines;4Centre N ational d ’Etud es Agronom iques d es Re´gions Chau d es (CN EARC), 1101 Av Agropolis, BP 5098, 34033 Montpellier Ced ex 1, France

In th e m ou n tain areas of n orth ern Vietn am , exten sive

h u sban d ry of large-ru m in an ts is lim ited b y th e lack of

n atu ral forage Sin ce 2000, th e M ou n tain Agrarian

System Program h as b een con cern ed w ith th is issu e

an d h as evalu ated tech n ical altern atives aim ed at su

s-tain ab le in tegration of crop -livestock system s O n e

altern ative con sists of feed in g system s for

large-ru m in an ts b ased on crop p in g systems w ith p lan t cover.

D iffu sion of in formation ab ou t th e n ew system is su p

-p orted b y a set of in teractive com m u n ication tools for

u se b etw een farmers an d research ers A p articip atory

sim u lation m eth od w as d evelop ed th at com b in es a com

-p artm en tal m od el of th e village territory w ith five tech n

i-cal in n ovation s Farmers sim u late th e ad op tion of th e

in n ovation s th ey ch oose am on g food -forage crop p in g

system s in rotation or in association , an d u rea-treated

straw Th ey evalu ate th e q u an tity of forage th at can b e

grow n for th eir h erd u sin g th e chosen set of tech n iq u es.

Sim u lation s revealed farmers’ con strain ts an d ob

jec-tives in ad op tin g su stain ab le crop p in g systems on th e

h illsid es w h ile m ain tain in g a sm all an im al h u sban d ry

system Th e resu lts ob tain ed b y each farmer p rovid e a

u sefu l b asis for d iscu ssion ab ou t th e imp lemen tation

of in n ovation Th is p articip atory sim u lation m eth od

cou ld b e m ore w id ely u sed to facilitate th e d iffu sion of

in n ovation s su ch as in tegratin g livestock feed in g

system s w ith con servation crop p in g p ractices.

Keyw ord s conservation agricu ltu re, livestock

system s, m ou ntain areas, p articip atory sim u

-lation, Vietnam

Introd uc tion

Since 1997, the Mou ntain Agrarian System s

Program has been investigating land -u se

changes and p rioritising d evelop m ent issu es in Bac Kan, one of the p oorest p rovinces in northern Vietnam (Castella et al., 2003a) The d iagnostic

p hase revealed three d riving forces for local

d evelop m ent that are shared by m any other

u p land areas in northern Vietnam and also by Southeast Asia in general First, the villagers’ access to m arkets, ed ucation, health services and technical inform ation is of critical im p ortance

in p rovid ing d iversified local d evelop m ent

op p ortu nities The geograp hic d iversity of the land scap e is reflected in the w id e d iversity of socio-econom ic environm ents (Donovan et al., 1997) Second , the su ccessive land p olicies that have accom p anied the p rocess of agricu ltu ral

d ecollectivisation over the last d ecad e have been a m ajor d riving force of changes in land

u se Land d istribu tion to ind ivid u al hou sehold s

p rofou nd ly m od ified land -u se system s both

in the valley bottom s and on the hillsid es (Castella & Dang, 2002; Jam ieson et al., 1998) Third , crop – livestock – forest interactions w ere com p letely transform ed by a changing econom ic context, and by accessibility and land tenu re

p olicies The extensive m anagem ent of

large-ru m inants (bu ffaloes and cattle) has becom e a

m ajor obstacle to agricu ltu ral intensification in both the low land s and the u p land s (Castella

et al., 2003b; H u sson et al., 2001a) The restriction

of the access bu ffaloes and cattle had to their trad itional forage resou rces resu lted in their straying into shru b- and forestland , cau sing p ro-blem s for forest regeneration and also for the anim als, as p oor d iet lead s to p oor p erform ance

In recent years m ajor environm ental concerns

118

have been raised abou t m ou ntain agricu ltu ral

system s, concerns that em erged after the

aboli-tion of the agricu ltu ral collectives in 1988

Althou gh ou r research p rogram m e cou ld not

go beyond the d ocu m entation of the im p act

of accessibility and land p olicies on land -u se

changes, w e d evelop ed a range of technical and

organisational innovations to overcom e som e of

the crop – livestock issu es that w e id entified

A d iagnostic stu d y carried ou t beforehand set

the stage for the d iffu sion of the innovations

A large nu mber of cropp ing p ractices based on

p lant cover and embracing the p rincip les of

con-servation agricu ltu re w ere d evelop ed as

alterna-tives to slash-and -bu rn p ractices and tested in

the real cond itions that the farm ers face (Altieri,

2002; H usson et al., 2001b; Up hoff, 2002) These

w ere d esigned as basic com p onents of m ore

com p lex cropp ing system s that farm ers can

com bine in d ifferent w ays d ep end ing on their

specific need s and objectives (e.g give p riority to

rice sufficiency, crop – livestock association, p ut

m ore emp hasis on forage for livestock, exp ansion

of p erennial crops, etc.) The m ost p rom ising

inno-vations thu s far have been, (1) d irect sow ing of

u pland rice or m aize in a m u lch of Brachiaria

(grass) or M ucuna (legum e), (2) crop – legum e

association (e.g cassava – Stylosanthes, orchard

-Arachis), (3) vegetative strips (natu ral vegetation,

p lanted grasses or legum es) along contour

lines, (4) a soil slow burning techniqu e that

boosts the fertility of d egrad ed soils by releasing

p hosp horu s in a form that the p lant can easily

u ptake, and (5) m ini-terraces to control erosion

on steep slopes In ad d ition to restru ctu ring soils,

controlling erosion and im p roving crop m

anage-m ent, these innovations can also p rovid e good

quality forage for livestock (Eguienta et al., 2002)

Farm ers w ho tested the p rop osed techniqu es in

their ow n field s w ithin the fram ew ork of the

p roject w ere very interested in these alternatives

to trad itional slash and bu rn p ractices They

su ggested im p rovem ents, and p rop osed som e

of their ow n innovations (e.g treatm ent of straw

w ith u rea) that w ere tested at a w id er scale in

the p roject H ow ever, the d iffu sion of these new

p ractices w as not an easy p rocess The sp ecific

constraints of the m ou ntain environm ent had to

be taken into accou nt in d esigning strategies for

the d issem ination of innovation In m ost cases it

is im p ossible for su bsistence farm ers to ad op t a

com p lete p ackage;stepwise adoption(i.e the

intro-d u ction of su ccessive com p onents of a given

crop p ing system s) is p referred The introd u ction

of any innovation is m anaged in a systemic perspective, as changes in the crop p ing p atterns

of sm all u p land farm s inevitably affect livestock and forest resou rce m anagem ent For exam p le, the d evelop m ent of cover crop s or vegetative strip s w ou ld lim it livestock access to fallow field s d u ring the w inter and w ou ld thu s requ ire farm ers (1) to fence their field s to avoid d am age

to soil conservation crop s by straying anim als, and (2) to com p ensate for the restriction

im p osed on trad itional forage resou rces by grow ing their ow n forage or p rovid ing access to other grazing areas The integrated com p onents

of farm ers’ livelihood system s thu s cannot be stu d ied or m od ified ind ep end ently from each other Lastly, soil and w ater conservation tech-niqu es enable the highly d iverse u p land

environ-m ent to be u sed to ad vantage In the face of the

d iversity of situ ations and system s, a wide range

of solutionsis requ ired Farm ers can choose from

am ong the op tions those that best fit their ow n

p articu lar circu m stances and need s

The above constraints in the d iffu sion of inno-vation call for an integrated m ethod ological

ap p roach to farm ers’ livelihood system s and

m ore sp ecifically to crop – livestock m anagem ent

In su ch a d iverse natu ral and hu m an

environ-m ent, only a p articip atory ap p roach is able to

id entify farm ers’ need s in a tim e effective

m anner (Castella et al., 2003b; N eef, 2004; Pretty, 1995) H ow ever, certain factors that ensu re the

su ccess of the p articip atory m ethod (Mosse, 1994) have to be taken into accou nt w hen d esign-ing a m ethod that allow s farm ers to sim u late the

ad op tion of innovation The p articip atory sim u -lation ap p roach d iffers from a ‘rap id ’ ap p raisal that cannot resu lt in a fu ll u nd erstand ing of the com p lex system s involved nor establish a relationship of tru st betw een the local p op u lation and ou tsid ers Fu rtherm ore, a p articip atory

ap p roach can accom m od ate d ifferent d egrees of

p articip ation by p articip ants in a given session, and also d ifferent p ercep tions of the ap p roach

by p articip ants (Moser, 1991) The facilitators of the m eetings m u st have good technical know

l-ed ge as w ell as good com m u nication skills to enable all the p articip ants to exp ress them selves Facilitators also need a good know led ge of the local context to be able to interp ret the

infor-m ation generated by the ap p raisal, p articu larly the socio-p olitical asp ects, in ord er to avoid m is-interp retation (Sayer & Cam p bell, 2004; Shanks

& Bu i, 2001) The tools (i.e tables, m ap s, figu res)

u sed to generate this inform ation m ay be an

obstacle to active p articip ation if they are not easy

to u nd erstand (Mosse, 1994) In ou r case, the

sim u lation w as not carried ou t in ‘virgin’ territory

bu t in a village w here p roject staff (inclu d ing one

native of the village) had been exp erim enting

innovations w ith local farm ers for the p ast three

years This p rocess had resu lted in a good

relation-ship betw een local stakehold ers and researchers

and the latter had acqu ired a good know led ge

of the local context

In this p ap er w e highlight the p rocess of

inno-vation d iffu sion that engaged both scientists and

local stakehold ers in d esigning a p latform for

social learning for the m anagem ent of natu ral

resources (Ro¨ling & Jiggins, 1998) The scientists

facilitated the d iffusion of innovation by p rovid ing

all the inform ation need ed to enable local p eople

to m ake their ow n evaluation of crop–livestock

issues, to collectively d iscover d ifferent p ossible

solutions and to ad ap t them to the sp ecific issues

they had d efined together To start the interactive

com m unication p rocess w e d esigned a grap hic

language to be u sed by scientists and local

stake-hold ers The grap hic language incorp orated tw o

m ajor assu m ptions d erived from p revious stu d

-ies in the same area: (1) the village entity is the

relevant sp atial and social u nit for com m

unity-based m anagem ent of natural resou rces (Castella

& Dang, 2002), and (2) the spatio-temp oral

d im ension of crop – livestock interactions is a key

factor to take into accou nt w hen d esigning or

introd u cing technical innovations (Castella et al.,

2002a) The aim of this p ap er is to rep ort on our

exp erience u sing and evaluating the new com

-m unication tool as a sup port for the d iffu sion of

sustainable technical innovations The m ethod

itself is d iscu ssed in d etail in tw o other p ap ers

(Castella et al., 2002a, 2002b)

The Suc c essive Sta g es in the

Pa rtic ip a tory Sim ula tion of

Innova tion Ad op tion

We p resented our grap hic m od els to rep

resenta-tives of the local com m u nity to check that w e had

in fact d esigned a ‘comm on spatial langu age’

betw een researchers and farm ers The com m on

spatial langu age is intend ed to p rovid e a concrete

sup p ort for (1) the p articip atory valid ation of

local inform ation related to spatial m anagem ent

of natu ral resources, and (2) the introd u ction of

technical innovations to imp rove feed ing systems

for large ru m inants Sixteen farm ers attend ed the

first session held on 18 and 19 October, 2001 at thePhieng LiengPeople’s Com m ittee The p

artici-p ants w ere (1) reartici-p resentative (men and w om en)

of the three m ain household typ es classified in a hou sehold typ ology m ad e in Phieng Liengin 2000 (Eguienta et al., 2002); (2) know led geable and influ ential p eople w ithin the com m u nity; and (3) farm ers w ho w ere familiar w ith the p roject activi-ties and innovations through p reviou s

involve-m ent in p roject exp eriinvolve-ments

At the beginning of the m eeting, the d ecrease

in natu ral forage resou rces and the im p ortance

of natu ral resou rce m anagem ent w ere high-lighted At the first session, a 3D m od el of the village (Castella et al., 2002b) w as u sed as a visu al su p p ort to show the farm ers the m ain land scap e featu res and the d ifferent land -u se classes We established a link betw een the m ain land m arks on the 3D m od el and their rep resen-tation on a p ap er-based sp atial grap hic m od el Once the p articip ants w ere com fortable w ith the

sp atial grap hic m od el that rep resented land u se

in their village in 2001, w e u sed the sam e kind

of grap hic rep resentations to show the changes

in land u se that their village had u nd ergone over the last 40 years We w ere thu s able to show how cu rrent land u se and related issu es

w ere a p rod u ct of the p ast (Castella et al., 2002b)

At the end of the first session, a blank calend ar (Figu re 1) w as d isp layed so the p articip ants cou ld rep resent the d istribu tion of their activities over a

p eriod of one year by d istingu ishing the relative contribu tion of m en and w om en and the location

of the d ifferent activities in the village territory Eight activities w ere inclu d ed : irrigated rice,

u p land crop s, orchard and gard en, anim al

hu sband ry, gathering of tim ber and non-tim ber forest p rod u cts, and off-farm activities For each activity and gend er, the corresp ond ing w orking

p eriod w as m arked on the calend ar Inform ation abou t annu al d istribu tion of labou r betw een activities and betw een m en and w om en is very

im p ortant becau se lack of labou r can be a m ajor constraint in the ad op tion of innovations

Interesting inform ation also cam e to light abou t interactions betw een crop and livestock related activities, show ing that the anim als relied exclu -sively on natu ral resou rces or crop resid u es for forage:

After each rice h arvest, th e ru m in an ts graze in the rice field s (N ovem ber – Decem ber and briefly in May – Ju ne)

Th e on ly forage resou rce available in th e crop

-p ing area d u ring the cro-p -p ing -p eriod is

veg-etation bord ering the field s and this is not

su fficient

Farm ers h ave to carefu lly m an age th eir stock of

rice straw, w hich is d istribu ted to the w orking

bu ffaloes d u ring the second rice crop p ing

season

At the second session, the sp atial m od el

corre-sp ond ing to cu rrent land u se w as first p resented

to the p articip ants and the sp atial com p artm ent

m od el w as then introd u ced After the d ifferent

com p artm ents had been d escribed , a p ap er

cop y of the grap hic m od el of the village w as d

is-tribu ted to each p articip ant w ho w as invited to

locate his or her ow n land resou rces in the d

iffer-ent com p artm iffer-ents N ext, five innovations w ere

p rop osed together w ith their feed ing cap acity

for large ru m inants Pap er cop ies of the com p

art-m ental art-m od el w ere u sed as su p p orts for a p

artici-p atory sim u lation of innovation ad oartici-p tion The

farm ers w ere then invited to choose am ong the

p rop osed innovations the one they w ou ld like

to ad op t, to locate it on the com p artm ental

m od el, and w ith the assistance of the researchers,

to calcu late the feed ing cap acity of their herd The

p articip ants w ere alread y fam iliar w ith these new crop p ing system s as they had been tested by som e of them in their ow n village They w ere free to p rop ose their ow n innovations w ith the corresp ond ing technical p aram eters

Mod elling Sp a tia l Ma na g em ent of

Na tura l Resourc es a t the Villa g e Level a nd Rela ted Livestoc k

Ma na g em ent Issues

The c om p a rtm enta l m od el

The d ifferent land -u se typ es in Phieng Lieng

village territory w ere taken from a village scale GIS to create a com p artm ental m od el (Castella

et al., 2002a), w here each kind of land -u se w as rep resented p rop ortionally This com m u nication Figu re 1 Sp atio-tem p oral d im ensions of the sim u lation

tool w as p resented to the farm ers attend ing the

sim u lation p rogressively in d ifferent versions

First, the sp atial m od el corresp ond ing to the

cu rrent land u se (first level of abstraction) that

had been show n to the farm ers at the p reviou s

session w as d isp layed on a p ap er board Then

the d ifferent elem ents of the sp atial m od el

(Castella et al., 2002b) w ere id entified and p

ro-gressively d raw n on a new schem atic m od el in

ord er to lead the farm ers to a second level of

abstraction: the com p artm ental m od el

Once the schem atic m od el w as d raw n, a

colou red version w as p resented w ith p rop ortions

of land u se areas that show ed the real village

situ ation (Figu re 1):

th e low lan d crop area m ain ly corresp on d ed to

1 – 2 cycles of irrigated rice, som etim es m aize

(hot rainy season) and vegetables (cold d ry

season);

th e gard en area w as d ed icated to vegetables

and fru it trees;

th e u p lan d crop area corresp on d ed to rain fed

rice, m aize, cassava and p lantations;

th e term ‘forest’ in clu d ed p oor an d secon d ary

forests as w ell as w ood y regrow th

Finally, each p articip ant received a sheet of p ap er

show ing the village m od el on w hich they cou ld

p osition their ow n crop and anim al resou rces

along w ith the innovations they had chosen

Sp a tio-tem p ora l c onstra ints linked to

a nim a l husb a nd ry

Anim al hu sband ry issu es exist in the tw o

d im ensions of sp ace and tim e At the beginning

of the 1990s, the coop erative herd w as d istribu ted

to ind ivid u als (one head/p erson) w ho each

looked after their ow n anim al The forage

resou rces that form ed the basis of this d evelop

-m ent w ere collective p astu res and the forest

H ow ever, these areas w ere not controlled and

natu ral resou rces grad u ally becam e insu fficient

in term s of feed ing cap acity, cau sing tw o

seriou s p roblem s that w ere inter-related :

lack of forage d u rin g th e w in ter (cold d ry

season): forep art from a lim ited qu antity of

d ried rice straw, natu ral vegetation w as the

only feed ing resou rce available and its

grow th w as near zero The resu lt w as w

eaken-ing anim als lead eaken-ing to a red u ction in fertility,

abortion, d ecreased resistance to d isease and

low w ork p erform ance at the beginning of

the rainy season (first cycle rice p lou ghing/ harrow ing in March);

in tra- an d in ter-village con flicts: d u rin g th e

w inter, starving anim als broke fences and

d am aged crop s w ithin the village and d u ring the rest of the year they grazed in the neigh-bou ring village’s m ead ow

To clearly illu strate the p roblem s cau sed by roam ing anim als, the m ovem ent of the village herd in the d ifferent com p artm ents w as illu s-trated on the m od el by arrow s (Figu re 2) The

ad vantages and constraints of d ifferent solu tions for ind ivid u al and/or collective m anagem ent of the livestock herd w ere then introd u ced and

d iscu ssed : closin g off access to oth er village territories

w ou ld be very d ifficu lt as it w ou ld requ ire fencing large areas in the u p p er p art of the village w atershed ;

exten d in g grazin g in th e collective area: th e land cu rrently u sed cou ld only feed 12 bu ffa-loes w hereas there w ere in fact m ore than 100

in Phieng Lieng.Increasing the am ou nt of m

ea-d ow lanea-d w ou lea-d requ ire collective m

anage-m ent or ind ivid u al u se of p lots in the collective area, neither of w hich w as con-sid ered feasible by the farm ers at the tim e; im p lem en tin g feed ing system s for

large-ru m inants by com bining innovative crop p ing system s in sp ace and tim e on ind ivid u ally

ow ned land : this w as the solu tion chosen by the farm ers w ho agreed to sim u late the resu lts of su ch system s

Results of the Pa rtic ip a tory Sim ula tion

Inp uts–outp uts

There w ere tw o sim u lation inp u ts:

ad op ted in n ovation : each in n ovation h as an associated yield and p eriod (w inter or

su m m er);

su rface area: 2000 m2 w as chosen as the u nit

su rface area, as it is the size of the average p lot The ou tp u t is exp ressed as the nu m ber of equ iv-alent bu ffalo (equ bu ) p er u nit area

Fou r p aram eters w ere inclu d ed : p rod u ction p eriod , exp ressed in d ays (PProd ); con su m p tion rate for a given forage, ran gin g from 1 to 10 (CR);

d aily th eoretical m ain ten an ce n eed s (MN ) for

1 equ bu , exp ressed in kg DM/equ bu/d ay

The average w eight of a bu ffalo is 300 kg and

its d aily food requ irem ent is 2.5 kg d ry m atter

(DM)/100 kg In ord er to take other anim als

into accou nt, a cow is estim ated as 2/3 equ bu

and a goat 1/10 equ bu ;

p oten tial yield , exp ressed in kg/h a (PY) For a

2000 m2p lot, PY

CR/(PProd

MN ) ¼ n equ bu

Orig in of the innova tions a nd the m od el

p a ra m eters

The crop p ing system s w ith p lant cover sim u

l-taneou sly enable su stainable agricu ltu re on

slop ing land and im p roved anim al feed ing

These techniqu es w ere first tested u nd er

con-trolled cond itions on exp erim ental p lots hired

from the farm ers Once valid ated in local

con-d itions, they w ere im p lem entecon-d by farm ers on

their ow n p lots w ith the assistance of technicians

w ho collected the d ata and feed back from

farm ers abou t the su ccess they had and the

d ifficu lties they encou ntered w hile im p lem enting the new techniqu es

At this p oint, five typ es of innovations had been selected in consu ltation w ith farm ers, each one sp ecific to an agro-ecological u nit (or com p artm ent of the village land scap e)

Food -forage crop rotation (e.g th ree years of

Brachiara sp then m aize) allow s feed ing of

N ovem ber

Food -forage crop association : – Brachiarasp interlined w ith m aize allow s feed ing of half bu ffalo/u nit area from

Au gu st to N ovem ber;

– Stylosanthes guyanensis– cassava or orchard (p erennial system , one year settlem ent and w eed ing requ ired ) allow s feed ing of one bu ffalo/u nit su rface from March to

N ovem ber;

orchard soil covered by Arachis pintoı¨ (p eren-nial system , one year settlem ent and w eed ing requ ired ) allow s feed ing of one bu ffalo/u nit area from March to N ovem ber;

Figu re 2 Discu ssion abou t a collective livestock m anagem ent p lan

u rea treated straw allow s feed in g of:

– one bu ffalo/u nit area from Decem ber to

March for m aize straw ;

– tw o bu ffaloes/u nit area, sam e p eriod for

rice straw

Win ter oats in th e irrigated rice field s allow s

feed ing of one bu ffalo/u nit area from

Decem ber to March

Valu es for p aram eters (Table 1) w ere estim ated

from exp erim ental resu lts obtained in the Mou

n-tain Agrarian System s Program from 2001 to 2003

and from the literatu re (FAO) In ord er to be

realistic and to highlight the constraints of

innovative crop p ing system s, reference yield s

corresp ond to m inim u m valu es obtained w ith a

low level of inp u ts on p oor soils Du ring the

d iscu ssion, inform ation abou t com p u ted yield s

w as com p lem ented by inform ation abou t forage

valu es d istingu ishing betw een p rotein and

energetic forages (Figu re 3) and consid ering the

biop hysical effects of the cover crop s:

grasses w ith a d eep an d stron g root system ,

su ch as Brachiaria sp , p lay a key role in soil

d ecom p action and in the im p rovem ent of

p orosity; they also p lay the role of ‘biological

p u m p ’ by recycling m inerals that are not

accessible to associated or rotated crop s;

legu m es, su ch as Stylosanthesand Arachissp p ,

im p rove chem ical fertility of the soil by fixing

atm osp heric nitrogen;

u rea treatm en t of rice an d m aize straw are

alternatives to bu rning straw that help red u ce

CO2 em ission w hile p rovid ing good qu ality

feed for livestock;

w in ter oats crop p in g in p ad d y field is a p

oss-ible first step tow ard m ore com p lex crop p ing

system s associated w ith a living p lant cover

A su bstantial increase in the soil organic

carbon content in the 10 cm top soil layer in

no-tillage soils com p ared w ith soils u nd er natu ral vegetation and long-term conventional tillage (Se´guy et al., 2003) can occur d u e to high crop-resid u e inp ut and lack of soil d

is-tu rbance Moreover, m icro and m acro-fauna

p opu lations, abund ance, d iversity and soil bio-logical activity d ecrease d uring trad itional cu lti-vation system s Conversely, the soil stru ctu re is imp roved w hen a vegetal cover w ith a strong root system is p erm anently m aintained on the soil (H ussonet al., 2003; Up hoff, 2002)

Im p lem enta tion of the sim ula tion

Po sitio ning fa rm ers’ reso urc es o n the m o d el

Farm ers d rew lines on the grap hic m od el to show the land they ow ned or cou ld u se in each com p artm ent, i.e the lim its of their ind ivid u al land u se system (Figu re 4) A table w as d istribu -ted so each farm er cou ld note d ow n the size of his bu ffalo herd

Figu re 3 Forage valu es of som e grasses and legu m es tested (H u ssonet al., 2003)

Tab le 1 Valu es for p aram eters of the innovations u sed for the sim u lation

Innova t ion P ot ent ia l yield

(kg /ha )

P roduct ion period (da ys)

Consumpt ion ra t e N equ.bu

Pre senta tio n o f the inno va tio ns

The innovations w ere p resented on a table

(Figu re 5) giving p otential ou tp u ts (equ bu/u nit

area) w ithin the p rod u ction p eriod along

w ith sp ecific characteristics (settlem ent tim e,

fer-tilisation and labou r requ irem ents, etc.), and

illu strated w ith p hotos taken d u ring p reviou s

exp erim ents in their village

Cho ic e o f inno va tio ns b y the fa rm ers a nd eva lua tio n o f the results

Coloured squares corresp ond ing to the d ifferent innovations w ere exp lained and then d istribu ted to the farm ers w ho chose som e of them and p laced them in the relevant com p artments on their

sche-m atic sche-m od el A brief d esche-m onstration showed thesche-m how to calcu late the nu mber of buffaloes fed Figu re 4 Positioning farm ers’ resou rces and innovations on the com p artm ental m od el

d uring the d ifferent p eriod s of the year The result

w as then transferred to the buffalo table In

ad d ition to the d em onstration, it w as exp lained

that:

th e feed in g cap acity lin ked w ith th e in n ovation

is only estim ated and in p ractice it m ay be

slightly d ifferent;

on ly th e sim p lest an d ch eap est tech n iqu es

had been p resented , and these w ou ld not be

su fficient to feed all the bu ffaloes even if they

w ere ap p lied to the w hole village territory;

at th is stage, th e calcu lation d id n ot take labou r

requ irem ents into accou nt

The farm ers w ere then invited to m ake their own

calcu lations based on their ow n sp ecific situation,

w ith the assistance of the p roject staff, and to enter

the result on their ow n table, enabling them to

see if their w hole herd cou ld or could not be fed

from their ind ivid u al p lots This triggered a

lively d iscussion (Figure 5) At the end of the

session they w ere remind ed that this w as only

the first step in solving their p roblem , and they

w ere encouraged to contact the team if they

w ished to imp lem ent the p roposed innovations

From Sim ula tion to Ac tion

Und ersta nd ing a nd d isc ussing the m od el a nd

p roposed innova tions

Follow -u p visits (tw o sessions totalling five

d ays, betw een the end of N ovem ber and the beginning of Decem ber) w ere m ad e to Phieng Liengfarm ers w ho had attend ed the sim u lation The aim w as to m aintain the farm ers’ interest in innovations initiated by the sim u lation, to check their u nd erstand ing of the m od el, and to get their p oints of view abou t the feasibility of the innovations The su rveys w ere cond u cted as

op en interview s to:

listen to th e farm er ’s com m en ts abou t th e

m eeting and p articip atory sim u lation;

ch eck the farm er ’s u nd erstand ing of the

m od el Before d iscu ssing the innovations the farm er w as interested in, he w as asked to

d escribe the three m ain com p artm ents of the

m od el, the location of his resou rces and the innovations he had chosen, in ord er to check his u nd erstand ing of the m od el and to clarify any p oints that had not p reviou sly been m ad e clear;

Figu re 5 Discu ssions betqeen farm ers and researchers abou t the sim u lation in term s of feed ing cap acity

d iscu ss th e in n ovation s ch osen (con strain ts,

p lans, etc.)

All those interview ed exp ressed p ositive

reac-tions to the m eeting Most of the p articip ants

cor-rectly u nd erstood the m od el and the sim u lation

gam e This resu lt confirm ed their interest and

ability to concentrate d u ring the testing of the

m od el Farm ers’ com m ents, p lans and

con-straints abou t p ossible innovations and their

ap p licability are p resented in Table 2 The

general colu m ns d escribe id eas for w hich there

w as a consensu s in the com m u nity The sp ecific

colu m ns are based on ind ivid u al com m ents

m ad e by farm ers of each farm typ e classified by

Egu ienta et al.(2002)

Th e lack of forage in w in ter w as a m ajor con

-straint for m ost of the resp ond ents It has

nega-tive effect on p erform ance, rep rod u ction and

resistance to d iseases In 2000 there w as an ep

i-d em ic that lei-d to the i-d eath of m any bu ffaloes

Ap art from in d ivid u al activities, th e p roject

m ay assist the village in a collective activity

p rop osed by one farm er to grow m u cu na on

the collective p astu re area to regenerate the

soil and overcom e w eed s Mu cu na is easily

cleared and other crop s can then be grow n

The p roject cou ld thu s initiate a d ynam ic of

collective m anagem ent that cou ld p artly com

-p ensate for -p roblem s cau sed by ind ivid u al

im p lem entation

Man y resp on d en ts w ere w orried abou t th e fact

that forage crop p lanting tim e (March) m ay

coincid e w ith the sp ring season There is a

risk of a labou r shortage at this tim e and

som e hou sehold s w ou ld thu s not be able to

im p lem ent the innovations

Very few of th e farm ers in terview ed w ere in

ter-ested in rotating forage and food crop s They

ap p eared to consid er it a w aste of land and to

p refer associating these crop s, w hich highlights

the land satu ration situ ation

Som e farm ers fear th at u rea treated straw,

esp ecially m aize straw, cou ld be d angerou s

for their livestock

On ly a few farm ers w ere in terested in grow in g

w inter oats becau se the p ad d y field s are

u su ally not fenced d u ring the w inter This

absence of fencing is also a seriou s constraint

for innovation im p lem entation in the hillsid es

Most of the crop associations are p lanted in the

resid ential/gard en area w here su rveillance is

easier

Table 3 show s the d iversity of the p articip ants rep resented by the w id e range of agricu ltu ral and anim al resou rces Farm ers’ strategies

d u ring the sim u lation w ere influ enced by their available resou rces, cap abilities and objectives, resu lting in a w id e range of reactions and inno-vation choices (Figu re 6) The objectives of their anim al hu sband ry activities d iffered as a fu nction

of their ind ivid u al circu m stances These can range from ow ning a cou p le of bu ffaloes for land p rep aration in the p ad d ies, to d evelop ing a living cap ital, p rod u cing m eat or d iversifying sou rces of incom e All these p aram eters influ -enced the farm ers’ interest in the sim u lation, their choice of innovations, and their m otivation

to actu ally im p lem ent the innovations For instance, som e farm ers w ho d id not ow n large

ru m inants also p layed the gam e, either becau se they p lanned to bu y an anim al or becau se they

w ere interested in the soil fertility im p rovem ent

fu nction of the techniqu es, or in the feed ing valu e of the cover p lants for fish farm ing It can also be seen that som e constraints affect the

w hole com m u nity w hile others are sp ecific to som e hou sehold s: available labou r is a p roblem for som e farm ers, w hile for others it is the

d isp ersion of their p lots or their rem oteness that

p revents them from fencing

The m ain ou tcom es of this exp erim ent in p arti-cip atory sim u lation w ere first that farm ers w ere not p rep ared to collectively m anage forage resou rces at the village scale in ord er to feed their entire bu ffalo herd They p referred to rely

p rim arily on the anim al feed they cou ld

p rod u ce on their ind ivid u al p lots (Castella et al., 2002b) Second ly, crop p ing system s based on cover crop s w ou ld be d evelop ed on the hillsid es only throu gh a step w ise p rocess that w ou ld solve the p roblem of forage d eficit d u ring the

w inter before establishing cover crop s on the slop es and therefore avoid ing d am age by anim als to u p land field s N ew forage resou rces from the valley bottom w ou ld facilitate the

intro-d u ction of the conservation crop p ing system s on the hillsid es

Im p lem enta tion of innova tions

The follow ing w inter season (2001), tw o inno-vations w ere p rop osed to the farm ers: u rea treated rice straw and w inter oats The aim of these tw o new com p onents in the livestock feed ing system s w as to low er anim al p ressu re

on the new cover crop s on the hillsid es as w ell