Bergville Annual Progress Report 2019 | Mahlathini Development Foundation

APPENDIX2: BERGVILLEANNUAL

PROGRESS REPORT

CA FarmerInnovation Programme for

smallholders in Bergville

Period: October 2018 – September2019

Farmer Centred Innovation in Conservation Agriculture

in upper catchment areas of the Drakensberg in the

Bergville region of KwaZulu-Natal

Compiled by:

Erna Kruger andPhumzile Ngcobo

September 2019

Project implemented by:

Mahlathini Development Foundation

Promoting collaborative, pro-poor agricultural innovation.

Contact:Erna Kruger (Founder and Coordinator)

Address: 2 Forresters Lane, Pietermaritzburg, 3201, KZN

Email: erna@mahlathiniorganics.co.za, info@mahlathini.org

Cell: 0828732289

Time of operation: 2003-2016

Legal status: NPC

BEE status: 4. –Certificate available.

Funded by:

3
IDENTIFICATION OF THE PROJECT .............................................................................................5
DESCRIPTION AND SELECTION OF STUDY AREAS.......................................................................................5
APPROACH AND METHODOLOGY................................................................................................5
KEY ACTIVITIES: OCTOBER 2018-SEPTEMBER 2019................................................................6
FINANCIAL REPORTING....................................................................................................................7
PROGRESS.............................................................................................................................................8
RESULTS ACHIEVED TO DATE....................................................................................................10
CA PRACTICE...................................................................................................................................... 12
Year 1(1st level) trial outlines..................................................................................................... 12
Year 2 (2nd level) trial outlines.................................................................................................... 13
Year 3 onwards (3rd level) trial outlines..................................................................................... 13
RAINFALL........................................................................................................................................14
RUNOFF ............................................................................................................................................. 15
Stulwane...................................................................................................................................... 16
Phumelele Hlongwane: Ezibomvini............................................................................................ 17
Ntombakhe Zikode: Eqeleni........................................................................................................ 18
Ndunwana; Boniwe Hlatswhayo ................................................................................................ 19
Conclusions.................................................................................................................................. 20
WATER HOLDING CAPACITY................................................................................................................ 20
Conclusions.................................................................................................................................. 22
GRAVIMETRIC WATER................................................................................................................22
Comparison of gravimetric water content results for two seasons (Phumelele Hlongwane –
Ezibomvini) ................................................................................................................................. 25
BULK DENSITY.............................................................................................................................28
SOIL HEALTH................................................................................................................................29
HANEY SOIL HEALTH TESTS PARAMETERS............................................................................................ 29
PLFA ANALYSIS.................................................................................................................................. 39
NITROGEN ......................................................................................................................................... 43
COMPARISON OF SH TEST RESULTS 2015-2018.................................................................................. 44
IMPLEMENTATION SNAPSHOTS ACROSS DIFFERENT VILLAGES...........................................46
LATE BEANS........................................................................................................................................ 47
COVER CROPS..................................................................................................................................... 48
FODDER PRODUCTION AND SUPPLEMENTATION....................................................................................... 49
YIELDS; 2018-2019 PLANTING SEASON .......................................................................................51
YIELDS IN INTERCROPPED AND ROTATED PLOTS..........................................................................53
BEAN YIELDS FOR THE 2018/19 SEASON............................................................................................. 55
FARMER CENTRES .......................................................................................................................55
Case study 1: Ezibomvini Farmers Centre ................................................................................55
Case study 2: Thamela Farmers centre....................................................................................56

Case study 3: Ndunwana Farmers centre.................................................................................57

Case Study 5: Stulwane Farmers centre...................................................................................58

Case study 6: Emabunzini Farmers centre...............................................................................58

VILLAGE SAVING AND LOAN ASSOCIATIONS..................................................................................60

STAKEHOLDER ENGAGEMENT.......................................................................................................63

MAIZE TRUST BOARD VISIT................................................................................................................... 63

ISSUES, COMMENTS AND SUGGESTIONS......................................................................................64

SUMMARY OF ANNUAL EXPENSES AS ON AUGUST 2019..............................................................65

Identification of the project

Description and selection of study areas

Work in the Bergville (KwaZulu-Natal) site continued with the 18village learning groups brought

on board in the 2017-2018 season. Attention has been given to consolidating and expanding the

learning groups within each village.The overall number of participants for the Bergville site is

now 348 smallholder farmers with between 1-6 years of experience under conservation

agriculture. This season 14 new participants came on board.

Approach and Methodology

The farmer centred innovation systems research process underpinning the programme, which is

based on working intensively with farmer learning groups and local facilitators in each of the

villages, has been continued and strengthened.

Within the learning groups farmer innovatorsvolunteer to set up and manage farmer-managed

adaptive trials as the ‘learning venues’ for the whole learning group. Farmer Field School

methodologies are used within the group to focus the learning on the actual growth and

development of the crops throughout the season. New ideas are tested against the ‘normal’

practise in the area as the controls. Farmers observe, analyse and assess what is happening in the

trials and discuss appropriate decisions and management practices. Small information provision

and discovery-learning (training)sessions are included in these workshops/ processes. These are

based also on the seasonality of the crop and the specific requests and questions from farmer

learning group participants.

Local facilitators are chosen from within and by members of the learning group to be a person

who has the required experience, knowledge and a willingness to support the other farmer

innovators in their implementation. Facilitators are only chosen and appointed where people

with the appropriate skill and personality exists. Local facilitators receive a stipend for a

maximum of 10 working days per month, for their support to the farmer innovators. They fill in

detailed timesheets outlining their activities against which they claim a monthly stipend.

Learning group members agree to a season long learning process and put forward the farmer

innovators to run the trials. Each prospective innovator is interviewed and visited and signs an

agreement with the GrainSA team regarding their contribution to the process. They undertake to

plant and manage the CA trials according to the processesand protocols introduced as well as a

control plot of the same size. For the latter, farmers provide their own inputs.

The adaptive trials are also used as a focus point for the broader community to engage through

local learning events and farmers’ days. Stakeholders and the broader economic, agricultural and

environmental communities are drawn into these processes and events. Through these events,

Innovation Platforms (IPs)are developed for cooperation, synergy between programmes and

development of appropriate and farmer-led processes for economic inclusion. These IPs also

provide a good opportunity to focus scientific and academic research on the ‘needs’ of the process.

In this season (2018-2019) the projecthascontinued to focus on the following elements of the

model, namely:

a) Support farmers who are in their 1st, 2nd , 3rd , 4th , 5th and 6th seasons,

b) Conscious inclusion of crop rotation to compare with inter cropping trials,

c) Inclusion of summer cover crops in the crop rotation trials,

d) Continuation with experimentation with winter cover crops, but planted in separate

plots rather than in-between maize,

e) Planting of late season beans,

f) More focussed introduction of lab-lab beans,

g) Fodder production and supplementation and,

h) Initiation of nodes for farmer centres that can offer tools, input packs and advice,

i) Support for existing VSLAs and initiation of new savings groups where requested,

j) Conscious inclusion of the local facilitators in the crop and progress monitoring

processes,

k) Further supply of tools (MBLI planters, animal drawn planters and knapsack

sprayers) to learning groups.

Key activities: October 2018-September 2019

For this season the focus has been on working with the Local facilitators to manage the processes

of distribution of the trial inputs, running the planting demonstration workshops, assisting their

learning group members with planting and monitoring of progress with planting and crop

growth, along with implementation of the co-funded process from LandCare. Support here has

been in the form primarily of seed and fertilizer and for hosting of farmers’ days.

Researcher-managed trial plotshave now been set upin Ezibomvini, Eqeleni, Ndunwana and

Mhlwazini to work on quantitative benchmarking of some of the visual CA indicators being used

in this process.This includes rain gauges, runoffplots, a weather station and gravimetric soil

sampling; with the intention of comparing water balances across control and CA trial plots.

In addition, a training session has been conducted for all field staff and interns in quantitative

measurements as well as initiating the new methodology for Visual Soil Assessments. A small

case study was conducted in Stulwane, alongside a MSc student from the University of Pretoria.

Soil health samples have been taken for 12 participants across five villages, along with 24 soil

fertility samples for new participants (Ezinyonyane, Emahlathini, Emadakaneni, Ndunwane,

Ezibomvini, Stulwane) and 44 repeat samples for existing participants to build a body of

information about the soil fertility and soil health status of the CA trial participants.

Two stakeholder innovation platform event/ farmers’ daywere conducted in collaboration with

KZNDARD and LandCare; Stulwane, November 2018and Emahlathini, March2019. Crossvistis

were conducted for smallholder farmers from Growing Nations in Lesotho (March 2019) and for

the Maize trust board members (May 2019).

The fifteen(15) VSLA’s (Village savings and Loan Associations) have continued.The group

members (258) have saved R880 000 in the last 13month cycle, but very little of these savings

have gone towards the procurement of inputs, as expected.

Progress for the farmer centre in Ezibomvini has again been monitored. Three new farmer

centres have beeninitiated. Small business development training has been offered to the VSLA

members, with a focus on agriculture.

Financial reporting

Below is a summary of the key result areas and budgets provided under the 2018/19project

cycle.

Table 1: Bergville SFIP budget outline for 2018-2019

Bergville Milestones: Farmer Centred Innovation in CA. October 2018-September 2019

Milestones/

Outputs

Key activities

Outcomes/ deliverables

Budgets

Capital Equipment

R -

Farmer

experimentation

Bergville

Administration and

sundries

Travel,accommodation, admin,

publications, monitoring and

evaluation

R 131 160,00

Farmer centred

innovation

systems

Farmer experimentation,

researcher managed

experimentation, savings groups,

farmer centres

R 565 811,00

Innovation

platforms

Stakeholder meetings, platform

building and events

R 24 000,00

TOTAL: Oct 2018-Sept 2019

R 720 971,00

Expenditure by MDF has followed the key activities above. The inputs subsidies paid in by

farmers has amounted to R19 400 for the year. These monies have been ploughed back into the

community by providingrotating loans to the start-up farmer centres, as well as materials for the

fodder supplementation experimentation process initiated.

The table below outlines expenditure on inputs for the 12-month period of this project.

Date

Inputs

Amount

Pd for by

grainSA

Farmer's

payments

2018/10/22

TWK Agri Winterton

R65 330, 30

2019/10/18

AGT Foods; cover crops

R16 175,90

R6 900,00

2019/10/18

AGT Foods; cover crops

R1 932,00

R12 500,00

Sub-total

R 0,00

R 83 438,20

R19 400,00

Total

R64038,20

Progress

The project is now operational across 18villages in the Bergville area, with a total of 348 learning

group participants and 207 farmer-level trials.

The basic experimental design was followed for all 1st year participants and most of the 2nd year

participants as well. Variations have included crop rotation,intercropping, summer and winter

cover crop mixes, planting of lab-lab beans,late season planting of beans andfodder crops for the

3rd-6th year participants.

The table below outlines activities related to objectives and key indicatorsfor the period of

October 2018 -September 2019.

Table 2:SUMMARY OFPROGRESS (OCTOBER 2018 -SEPTEMBER 2019) RELATEDTOOBJECTIVES AND KEYACTIVITIES

Objectives

Key activities

Summary of progress

% completion and comment

1. Document

lessons

learned

Documentation for

learning and

awareness raising:

Farmer Field School

methodology and process

reports

Farmer level learning

materials; manuals –

isiZulu, English (re-print)

Project reports (monthly,

6 monthly and yearly).

Articles and promotional

material to engage

stakeholders in the

broader environment.

- To be done at end of season

- No reprint done

- Monthly reports (Oct2018-Sept

2019) interim and annual report

- Not done

LEGEND

BERGVILLE

SITES:

Ezibomvini

Stulwane

Eqeleni

Emabunzini

Emadakaneni

Emafefetheni

Emahlathini

Mhlwazini

Ndunwana

Ngoba

Vimbukhalo

Magangangozi

Emazimbeni

EMangweni (2)

Nsuka

Okhombe

Thamela

Potshini

Ntabamhlope

Sharing of information

through various

innovation platforms and

processes; including the

internet, social and

networking platforms and

conferences

-Stulwane and Emahlathini

farmers days, Growing Nations

and MT board cross visits, MDF

Website updated,

(100% completion)

Final report

- 6 monthly interim

reports

- Interim and annual reports

finalised.

(100% completion)

2Increase

focus and

efficiency of

CA systems,

scale out

sustainable

farming

systems

scenarios

and build

social

platforms

Farmer centred

innovations

systems research

Scale out using

information

systems approach.

1st , 2nd , 3rd , 4th , 5th and

6th level experimentation

Develop and manage

PM&E framework; –

weekly and monthly M&E

visits

Innovation platform

events- cross visits,

conferences, workshops,

meetings, farmers’ days

Action planning with

innovation platform

events; Major planning

event for experiments

Bi-annual steering

committee meetings

- Undertaken for 19 villages

(100% completion)

- New VSA methodology, staff

training in Quantitative

measurements, pendragon e-

survey for crop monitoring

- Stulwane and Emahlathini

farmers days, Growing Nations

and MT board cross visits

- Small business Dev training for 4

learning groups

- Soil fertility and soil health

learning sessions for 2-3 groups

- Future fodder experimentation

planned in association with AGT

Foods and Cedara Soil Science

dept.

-Mycotoxin research planned in

association with the ARC

(Potchefstroom)

-Planned for Sept 2019

(100% completion)

A performance dashboard is indicated below. This provides a snapshot of performance according

to suggested numbers and outputs in the proposal.

Table 3: PERFORMANCE DASHBOARD; SEPTEMBER2019

Outputs

Proposed (March 2018)

Actual (Feb 2019)

Number of areas of operation

Number of villages active

No of 1st level farmer experiments

No of 2nd level farmer experiments

No of 3rd level experiments

120

No of 4th level experiments

No of 5th- 6th level experiments

No of local facilitators

No of direct beneficiaries

350

207

VSLAs

Participatory monitoring and

evaluation process (farmer level)

Yes

The extremely dry conditions during the planting season (October-December 2018), linked to

difficulties with providing inputs due to lack of supply through the LandCare process led to a

smaller no of farmer level trials being undertaken. The drought continued through January 2019

and further reduced planting of late season beans and winter cover crops. It also meant that the

2-row planter procured for expansion of CAimplementation and for farmer level

experimentation was not used this season.

Results achieved to date

The framework for scaling out implementation included: Continuation with existing farmer

experimentation options for 1st, 2nd and 3rd level participants and mentoring and monitoring for

4th and 5th yearparticipants. This includes intercropping, crop rotation, late season planting of

beans and combinations of summer and winter cover crop mixes.

The table below outlines the villages, numbers of participants and experimentation processes for

the present learning groups in the Bergville area.

Table 4: ACTIVITIES AND NUMBERS OF FARMERS INVOLVED,PER VILLAGE FOROCTOBER2018-SEPTEMBER 2019.

BERGVILLE

Year started with CA

2018

/19

COMMENTS

Villages

2013

2014

2015

2016

2017

2018

Total

Trials

Emabunzini

Intercropping with

hand hoes and

MBLI planters;

Maize, beans,

cowpeas

Emangweni-

Engodini

1st and 2nd level

experimentation;

intercropping

Emangweni-

Emaqeleni

intercropping 1st

level

experimentation;

intercropping

Eqeleni

1st, 2nd and 3rd level

experimentation;

MBLI’s hand hoes

and animal drawn

planters;

intercropping crop

rotation summer

and winter cover

crops, late season

beans

Ezimbovini

1st, 2nd and 3rd level

experimentation;

MBLI’s hand hoes

and animal drawn

planters;

intercropping crop

rotation summer

and winter cover

crops, late season

beans

Magangangozi

1st and 2nd level

experimentation;

intercropping

Mhlwazini

1st, 2nd and 3rd level

experimentation;

MBLI’s hand hoes,

intercropping crop

rotation summer

and winter cover

crops, late season

beans

Ngoba

1st, 2nd and 3rd level

experimentation;

MBLI’s hand hoes

and animal drawn

planters;

intercropping crop

rotation summer

and winter cover

crops, late season

beans

Nsuka-

Zwelisha

Intercropping with

hand hoes and

MBLI planters;

Maize, beans,

cowpeas

Okhombe

1st and 2nd level

experimentation;

intercropping

Potshini

3rd level

experimentation

Stulwane

1st, 2nd and 3rd level

experimentation;

MBLI’s hand hoes

and animal drawn

planters;

intercropping crop

rotation summer

and winter cover

crops, late season

beans, fodder

crops

Thamela

Intercropping with

hand hoes and

MBLI planters;

Maize, beans,

cowpeas

Thunzini

Intercropping with

hand hoes and

MBLI planters;

Maize, beans,

cowpeas

Vimbukhalo

1st and 2nd level

experimentation;

intercropping, crop

rotation, Lab-Lab,

SCC

Ndunwana

1st and 2nd level

experimentation;

intercropping

Emahlathini

Intercropping with

hand hoes and

MBLI planters;

Maize, beans,

cowpeas

Emazimbeni

Intercropping with

hand hoes and

MBLI planters;

Maize, beans,

cowpeas

Emafefetheni

1st and 2nd level

experimentation;

intercropping

Grand Total

125

348

207

15,2ha ha trials;

~4ha controls

348 Participants across 19villages have joined the CA experimentation process between 2013-

2018 and have been implementing the CA trials for between 1-5 seasons. This year 207 trials have

been planted of whom 14 participants started this season for the first time.

The level of planting of CA control plots-normal planting for participants-was very low;again

due to the very dry conditions and late planting dates for this season.

CA practice

After the 3rd year, the farmer experimentation protocols for each farmer participant is defined by

their own preferences, given that those farmers with more experience can now incorporate some

of their own learnings and preferences in the trials, but the 1st and 2nd level trial participants still

need to get used to the overall CA planting process and thus the close spacing intercropping trial

plots are ‘prescribed’ for them.

The protocols are outlined below:

Year 1(1st level) trial outlines

Experimental design is pre-defined by the researchteam (based on previous implementation in

the area in an action research process with smallholders). It includes a number of different

aspects:

•Intercropping of maize, beans and cowpeas

•Introduction of OPV and hybrid varieties for comparison (1 variety of maize and beans

respectively)

•Close spacing (based on Argentinean model)

•Mixture of basin and row planting models

•Use of no till planters (hand held and animal drawn)

•Use of micro-dosing of fertilizers based on a generic recommendation from local soil

samples

•Herbicides sprayed before and/or at planting

•Decis Forteor Kemprinused at planting and top dressing stage for cutworm and stalk

borer

•Planting of cover crops; winter mix in Autumn

Experimental design includes 2 treatments; planter type (2) and intercrop (2)

Year 2 (2nd level) trial outlines

Based on evaluation of experiment progress for year 1, this includes the addition of options that

farmers choose from. Farmers also take on spraying and plot layout themselves:

•A number of different OPV and hybrid varieties for maize

•A number of different options for legumes (including summer cover crops)

•Planting method of choice

•Comparison of single crop and inter cropping planting methods

•Use of specific soil sample results for fertilizer recommendations

•Early planting and

•Own choices.

Year 3 onwards (3rd level) trial outlines

Based on evaluation of the experimentation process to date this protocol includes issues of cost

benefit analysis, bulk buying for input supply, jointactions around storage, processing and

marketing. Farmers design their experiments for themselves to include some of the following

potential focus areas:

•Early planting; with options to deal with more weeds and increased stalk borer pressure.

•Herbicide mix to be used pre and at planting (Round up, Dual Gold, Gramoxone)

•A pest control programme to include dealing with CMR beetles

•Intercropping vs crop rotation options

•Spacing in single block plantings

•Use of composted manure for mulching and soil improvement in combination with

fertilizer, or singly.

•Soil sample results and specific fertilizer recommendations

•Planting of Dolichos and other climbing beans

•Summer and winter cover crops; crop mixes, planting dates, management systems,

planting methods (furrows vs scatter)

•Seed varieties; conscious decisions around POVs, hybrids and GM seeds

•Cost benefit analysis of chosen options

•Fodder crop and supplementation experimentation and

•Farmer level monitoring of trials for selected individuals.

Rainfall

This season rain gauges were installed in six villages within the Bergville study site. The

monthly average rainfall data forthese gauges are summarised in the table below and are

compared to the local weather station data (Davis weather station in Ezibomvini)

Table 5: Rainfall data for 6 villages in the Bergville site; September 2018-May 2019

Rainfall (mm/month) 2018-2019 summer rainfall season; Bergville villages

Village

Weather station

(Ezibomvini)

Month

Stulwane

Ndunwana

Ezibomvini

Eqeleni

Emhlwazin

Thamela

Average

Rain (mm/

month)

ETc

(mm/

day)

Sep-18

30,3

5,8

154,36

Oct-18

19,5

17,8

24,6

117,47

Nov-18

106

68,1

180

74,8

47,7

95,3

50,4

148,16

Dec-18

76,5

56,6

152,34

Jan-19

321

27,5

258,5

290,4

175,2

70,6

142,01

Feb-19

135

253

218,7

254

171,8

356

231,4

139,8

108

Mar-19

177,5

214

205,5

63,2

133,2

212,4

100

Apr-19

136,5

81,7

149,9

100

May-19

0,0

84,92

TOTALS

594,5

937

699,3

1029

676,7

671,7

768,0

744,5

1107,26

Note: values in dark grey were estimated from online weather data for the period –as the weather station was faulty

during this period

The seasonal average for the rain gauges and weather station compare quite well at 768mm and

744mm respectively. This can be considered a reasonably high rainfall for this area, but given

the extremely late onset of rain and the high evapotranspiration values for this season, crop

growth was severely hampered.

The average rainfall recorded for the 2017-2018 season for December- May was averaged at

563mm. For this season in the same time period the average rainfall was 678mm. The actual

evapotranspiration (Etc) for 2017-2018 was however substantially lower at 702,8 mm than

this year, which was calculated at 1107,3 mm for the season. This indicates the major difference

between the two seasons and why the crops fared so badly this year, even with higher rainfall

than last year.

This observation is supported by a number of other studies, indicating the evaporative potential

in a growing season has a much greater potential effect on maize yield potential than overall

rainfall and temperature, as explained in the quote below:

“Recent studies indicate that the negative effect of high summer temperatures is due less to effects

on reproductive growth (e.g., heat damage between anthesis and silking, reducing pollen and grain

set) and more to increased moisture stress driven by vapor pressure deficit (VPD). Rising VPD

increases evapotranspiration, which has a two-fold impact on crop moisture stress: 1)

photosynthesis declines as crops that are unable to meet transpirative demand reduce their

stomatal conductance and 2) soil water supply to the crop declines due to increased evaporation

from the soil surface”

These authors proposed the need for increased soil organic matter to effect greater water

holding capacity (WHC) in the soil to mitigate these effects. They also state that “Other strategies

will be required to complement WHC increases, such as crop genetic improvement, cropping

system design, and irrigation technologies, among others”.

Runoff

This season 4 farmers managed runoff plots in their CA trials alongside their rain gauges to

ascertain the difference in runoff between the conservation agriculture trial plots and a

conventional control plot. The results are summarised below.

Data is summarised on a monthly basis, with the understanding that the runoff is generally

related to amount and intensity of rainfall as well as dryness of the soil. Given that the soils in

Bergville are high clay soils they also tend to be quite compacted and become extremely hard

when dry. This could lead to increased runoff, but this depends on the intensity of the rainfall

events.

Table 6: Runoff results for 4 participants across Bergville; 2018-2019

Stulwane

Ndunwana

Ezibomvini

Eqeleni

Runoff CA

Trial(ml)

Runoff Control

(ml)

Runoff CA

Trial(ml)

Runoff Control

(ml)

Runoff CA

Trial(ml)

Runoff Control

(ml)

Runoff CA

Trial(ml)

Runoff Control

(ml)

Nov-18

2808,0

3267,0

Dec-18

3 343

2 600

35,2

39,5

5 800

5750

Jan-19

5 900

2 250

305

348

30,8

31,0

10 000

12750

Feb-19

3 266

6 275

471

609

66,0

74,5

12710

13 250

Mar-19

2 423

1 615

117

24,1

27,5

9 800

9 000

Apr -19

4 836

5 875

2,7

2,3

4 000

Average

Nov-Apr

3 954

3 723

179,4

223,4

494,5

573,6

8 500

8950

From the table above it can be seen that for 3 of the 4 villages the runoff in the CA trial plots

were on average lower than the conventional control plots. The difference in runoff between the

CA trial and conventional control plots is not as significant as it has been in previous years. This

is likely due to the larger number of small rainfall events this season. In addition, because of the

Williams A, Hunter M.C, Kammerer M,. Kane D.A, Jordan N.R, Mortensen D.A, Smith R.G, Snapp S,

and. Davis A.S. 2016. Water Holding Capacity Mitigates Downside Risk and Volatility in US Rainfed

Maize: Time to Invest in Soil Organic Matter?Published: August 25,

2016https://doi.org/10.1371/journal.pone.0160974Soil.

difficult growing season none of the plots (M+B, M+CP and M only), were well covered with a lot

of exposed soil, with a lot more erosion evident than in previous seasons.

In the section below the effect of different cropping options within each of the CA trials is

explored in more detail.

Stulwane

Table 7: Runoff results or different cropping options within the CA trial; Stulwane 2018-2019

Stulwane; Nelisiwe Msele

Rainfall

CA Plot 1

(M+CP)

CA Plot 3

(Maize)

CA Plot

(Beans)

CA Plot

(M+CP)

CA average

Conventional

Control

Dec-18

3 750

1 170

4 100

4 350

3 343

2 600

Jan-19

11 000

9 600

2 000

1 000

5 900

2 250

Feb-19

135

4 995

2 955

2 135

2 980

3 266

6 275

Mar-19

177,5

3 950

1 050

2 270

2 423

1 615

Apr-19

136,5

6 333

3 910

6 100

3 000

4 836

5 875

Average seasonal runoff

3 954

3 723

For Nelisiwe Msele the expected trend of higher runoff on the CA plots early in the season,

leading into lower runoff values towards the end of the season is clearly visible, as is the trend

for the conventional (ploughed control) of having less runoff early in the season and higher

runoff as the season progresses. This trend has been recorded in the literature and can be

explained through increased macropores in the soil after ploughing, that gradually collapse

throughout the season to lead to higher compaction in the soil. Soils under CA are also generally

more compacted, but aggregate stability and micropores are present that improve water

infiltration and water holding capacity (Cavalieri et al., 2009, Basset ,T.S 2010)

Overall the CA plots for Nelisiwe had slightly greater average runoff than her conventional

control plot. She has been practicing CA for 5 years, but her soil cover has been recorded at

between 1-5% over the years; meaning that it has remained very low primarily due to grazing

of stray livestock.

If one considers the percentage rainfall that has been converted to runoff, as shown in the small

table below, it can be seen that this percentage is quite low, averaging 4,6% for the CA trial plots

and 4,3% for the conventional control plot. This can be related to the general stability of high %

clay soils as well as the reasonably high percentage of organic matter (OM), namely 4,3% in the

CA trial plot.

Table 8: Percentage rainfall converted to runoff for CA trial and conventional control plots in Stulwane; 2018-2019

Cavalieri K.M.V., da Silva A.P., Tormena C.A., Leão T.P., Dexter A.R. and Håkansson I., 2009.

Long-term effects of no-tillage on soil physical properties in a Rhodic Ferrasol in Paraná,

Brazil. Soil and Tillage Research, 103 (158-164).

Basset, T.S. 2010. A comparison of the effects of tillage on Soil physical properties and microbial

Activity at different levels of nitrogen Fertilizer at Gourton farm, Loskop, Kwazulu-Natal. MSC thesis.

Dept of Soil Science, UKZN.

Percentage rainfall converted to runoff

Rainfall

Conv

Dec-18

5,2%

4,1%

Jan-19

10,4%

3,9%

Feb-19

135

2,4%

4,6%

Mar-19

177,5

1,4%

0,9%

Apr-19

136,5

3,5%

4,3%

Average % runoff

4,6%

3,6%

Phumelele Hlongwane: Ezibomvini

Table 9: Runoff results or different cropping options within the CA trial; Ezibomvini 2018-2019

Phumelele Hlongwane: Ezibomvini

Rainfall

runoff (ml)

CA Plot 2

(M+CP)

CA Plot 6

(M+B)

CA Plot

(Maize)

CA trial

ave

control

Conven

contrl

Sep-18

Oct-18

Nov-18

68,1

2393,0

2016,0

4015,0

2808,0

3267,0

Dec-18

35,0

37,0

33,5

35,2

39,5

Jan-19

27,5

35,1

29,4

28,0

30,8

31,0

1007,5

Feb-19

218,7

60,0

72,5

65,5

66,0

74,5

16,5

Mar-19

214

31,7

21,2

19,5

24,1

27,5

3,0

Apr-19

4,0

2,0

2,7

2,3

1,8

Ave Seasonal

runoff

426,5

363,0

693,9

494,5

573,6

257,2

Phumelele has converted most of her farming to CA. She is in her 5th year of implementation.

This year we attempted to find a conventional control plot; the plot selected was planted to

sweet potatoes, which means it was cultivated. For Phumelele her % soil cover linked to crop

residues or stover is around 10%, given that she has fenced her field and control her livestock’s

grazing in this field. Although this percentage is still quite low, it is in fact substantially higher

than for those participants who do not practice controlled grazing, which is the vast majority of

participants. It is also lower this season, given that she has had cover of between 20-35% in

previous seasons. It is likely an indication of the lack of winter grazing in the area, due to

difficult climatic conditions and Phumelele’s temptation to allow the cattle to graze more of the

residue as a result.

This season the average seasonal runoff in her Maize only CA plot was substantially higher than

for her intercropped plots (M+B and M+CP). As Phumelele rotates the crops in her plot every

season, it would appear that the differences in runoff between the plots is related a lot more to

the specific soil properties in each plot, than the specific seasonal cropping option. This result

may also be linked to canopy cover –this season, growth of the crops was impeded by the

weather conditions and canopy cover was never reached, while in the previous season full

canopy cover had been reached by the end of January.

If one considers the percentage rainfall that has been converted to runoff, as shown in the small

table below, it can be seen that this percentage is very low, averaging 0,95% for the CA trial

plots, 1,11% for the CA control plots and 0,36% for the conventional control plot. In Phumelele’s

case her %OM is 3,6% for her CA Trial plot and 2,9% for her conventional control. It is unclear

why the runoff for the conventional control plot is lower than that of the CA trial. It is possible

that the slope of the runoff pans were not well calibrated and that the cultivation practices for

sweet potatoes provide for different runoff conditions in this plot. In retrospect, using a field

allocated to a different crop may not have been such a good idea. The trend for lower runoff

from the CA trial plot, when compared to the CA control plot, which has been observed in the 2

previous seasons has continued into this season.

The percentage rainfall converted to runoff for Phumelele is substantially lower than that of

Nelisiwe (presented above) and attests to her continued good soil management practices.

Table 10: Percentage rainfall converted to runoff for CA trial and conventional control plots in Ezibomvini; 2018-2019

Percentage rainfall converted to runoff

mm (Weather

station)

CA trial

CA control

Conv control

Nov-18

50,4

5,57%

6,48%

Dec-18

0,04%

0,05%

Jan-19

70,6

0,04%

1,43%

Feb-19

139,8

0,05%

0,01%

Mar-19

212,4

0,01%

0,00%

Apr-19

149,9

0,00%

Average %

runoff

0,95%

1,11%

0,36%

Ntombakhe Zikode: Eqeleni

Table 11: Runoff results or different cropping options within the CA trial; Eqeleni 2018-2019

Ntombakhe Zikode; Eqeleni

Rainfall

Runoff (l)

CA plot

CA plot 2

CA plot

CA Ave

Control

Convenl

Control

Ave

Dec-18

5,5

6,5

5,8

6,5

5,75

Jan-19

258,5

10,5

9,5

10,0

12,5

12,75

Feb-19

254

10,5

13,5

12,7

12,5

13,25

Mar-19

205,5

11,5

9,8

8,5

9,5

Apr-19

4,0

3,5

4,5

Ave Seasonal

runoff

8,5

7,9

8,5

8,8

9,1

8,95

Ntombakhe Zikode is in her 6th year of CA implementation. She also employs a combination of

multi-cropping and crop rotation in her CA trial and has improved her soil management

practices substantially over the last five years. Because of pressure from livestock in the area,

her soil cover from stover is still low; averaging around 3-5%. In addition, the %OM in her CA

trial plot averages around 1,9%, which shows some improvement, but is still quite low for the

area.

It can be seen from the table above that her runoff from both her CA trial plots (Ave 8,8l) are

quite high and much higher than those for Ezibomvini (Ave 0,5l) and Stulwane (Ave 4,4l).This

points towards the damage of her soil caused by long term monocropping and ploughing and

the length of time required to re-build her soil. Ntombakhe has ploughed her fields regularly for

many years, unlike Nelisiwe, who has only done this occasionally and Phumelele who has

always tilled by hand.

Table 12: Percentage rainfall converted to runoff for CA trial and conventional control plots in Eqeleni; 2018-209

Percentage rainfall converted to runoff

CA trial

CA control

Conv

control

Dec-18

9,38%

7,81%

10,16%

Jan-19

258,5

3,87%

5,03%

4,84%

Feb-19

254

5,00%

5,51%

4,92%

Mar-19

205,5

4,77%

4,14%

4,62%

Apr-19

5,97%

5,22%

6,72%

Average % runoff

5,80%

5,54%

6,25%

Predictably, the percentage rainfall converted to runoff in Ntobmakhe’s plots is much higher as

well. Runoff in her CA plots 9both the trial and the control) is lower than her conventionally

tilled plot.

Ndunwana; Boniwe Hlatswhayo

Table 13: Runoff results or different cropping options within the CA trial; Ndunwana 2018-2019

She is in her 4th year of CA implementation and still following the 400m2trial layout of 2 plots of

M+B and M+CP intercrops. She has received good yields averaging around 9,6t/ha for her maize

in the 2017-2018 season. For the CA trail plot the organic matter has been recorded at 2,9% and

Nduwane; Boniwe Hlatshwayo

Rainfall

CA runoff

(M+B)

Conventional

runoff

Dec-18

Jan-19

321

305

348

Feb-19

253

471

609

Mar-19

117

Apr-19

Average seasonal runoff

179,4

223,4

for her conventional control plot at 2,75%. Boniwe recorded very low runoff values, for both

her CA and conventional control plots–with a lower average seasonal runoff value for the CA

plots.

Table 14: Percentage rainfall converted to runoff for CA trial and conventional control plots in Ndunwana; 2018-2019

Percentage rainfall converted to runoff

CA trial

Conv control

Dec-18

0,05%

0,06%

Jan-19

321

0,10%

0,11%

Feb-19

253

0,19%

0,24%

Mar-19

0,09%

0,16%

Apr-19

0,07%

0,05%

Average % runoff

0,10%

0,12%

Boniwe’s percentage of rainfall converted to runoff results are very low and are similar to those

for Phumelele in Ezibomvini. This provides some weight to the argument that in the longer

term, hand tillage, followed by CA has led to stable, well- structured soils.

Conclusions

•Runoff for the 2018-2019 season was much lower than the runoff measured in the two

previous seasons, despite the fact that the overall rainfall was not that different. This can

be attributed mainly to the rainfall intensity and periodicity but also to slowly

improving organic matter content in the soil

•Historical land management practices have a large effect on the localised soil structure

and soil health. It may take many seasons to rebuild a living soil with good aggregate

stability and the related characteristics of reduced runoff and improved infiltration.

There is evidence that those smallholder farmers who have always practiced hand

tillage have soils that are in a much better state than those who ploughed continuously

prior to starting their CA implementation.

•Even within the CA trial plots (which are divided into 10m2 blocks), there can be

considerable variation in soil quality, which again is related to historical management

practices. It is considered that the differences in runoff between these blocks is related

much more to the differences in historical land management practices than the different

cropping options presently implemented.

•On average, the mixed cropped CA trial plots show less runoff than the CA control plots

which have been mono cropped to maize.

•For this season, the conventional control plots (ploughed) have on average shown less

runoff than the CA trial plots. Although there has been a steady, but slow increase in

percentage soil organic carbon (and %SOM) in the CA trial plots, the comparison of

these CA plots with newly ploughed conventional plots has been problematic. As will be

discussed in the following section, we have battled to find appropriate conventionally

tilled plots, as the farmers in question have converted all their production to CA.

Water holding capacity

In the Bergville area, the WHC (water holding capacity) of the soil is naturally high, given the

high clay content and reasonably high SOM content (2-4%). A study conducted with 5

participating smallholder farmers in Stulwane, by a Soil Science Masters student from the

University of Pretoria (Palesa Motaung), confirms these generalisations.

As in many of our present analyses, students, interns and fieldworkers battle to conceptualise

the importance of control samples and also battle to find appropriate controls –as in many

cases the farmers that we are now working with for these measurements have moved across to

CA for their entire cropping areas and do not have conventional tillage control plots. In Palesa

Motaung’s study, given that she is focussing on soil health aspects, she used veld samples as her

controls.

She has used both the Visual Soil Assessment methodology refined by our team as well as the

Cornell comprehensive soil health assessment framework –which uses chemical, biological and

physical soil measurements to provide indices

and scores for soil health.

Among the soil health tests that she conducted, she calculated available water holding capacity

(AWC) for the following plots for five 5thyear CA farming participants in Stulwane:

•CA maize only

•Ca maize and beans

•Veld

The results are shown in the small table below

Water holding capacity (g water per g

soil)

Treatment average of 5 farmers

(Stulwane)

0,58

CA maize only

0,58

CA maize and beans

0,62

Veld

The AWC is the amount of water available to plants –between the field capacity and wilting

points for the particular soil. For the samples tested, the AWC is scored at 100% for all three

treatments (CA maize only, CA maize and beans and Veld). This means that the water holding

capacity of the soils in our study area are high. In addition, the water holding capacity of the CA

trials are very close to the veld benchmark, indicating the benefit of the implemented CA

system. The system consists of rotated plots of different combinations of mono-cropped maize,

legumes and cover crops.

Table 15: Soil quality scores provided by the Cornell soil assessment framework for 5 participants in Stulwane; 2018-

2019

Treatment

Overall

Quality

Score

Overall

Biological

Quality Score

Overall Chemical

Quality Score

Overall Physical

Quality Score

Description

Soil organic

matter, active

carbon, microbial

respiration

Extractable P, K

and pH

Available water

capacity, wet

aggregate stability

CA Maize Only

60,7

48,2

62,6

76,7

B.N. Moebius-Clune, D.J. Moebius-Clune, B.K. Gugino, O.J. Idowu, R.R. Schindelbeck, A.J. Ristow,

H.M. van Es, J.E. Thies, H.A. Shayler, M.B. McBride, K.S.M. Kurtz, D.W. Wolfe, and G.S. Abawi

.2017. Comprehensive Assessment of Soil Health. The Cornell Framework. Third Edition. Cornell

University, Ithaca New York.

CA Maize & Beans

54,7

43,2

51,2

77,3

Veld

63,0

56,4

61,1

75,9

The differences in the scores between the CA maize only and CA maize and bean plots were to

some extent artificial and related to sampling, rather than the treatments. Extractable P for

example was extremely high for a few of the CA plots–but were likely due to recent fertilization

–rather than an overall over supply of P in the soil, but led to much lower scores, as indicated in

the pink shaded block of the table above.

For 3 of the 5 participants, the scores for biological properties were lower for their CA maize

and bean plots than for their CA maize only plots –as indicated in the blue shaded block in the

table above. A trend that has been noticed already in this research process is that soil quality

within participants’ fields can vary considerably and that microbial respiration and active

carbon also varies considerably between the different treatments in a 10- block layout

(10mx10m blocks). Treatments consist of monocropping and intercropping mixes, with cover

crops, which are rotated. This variation is not directly related to the present crop combination

in the block, or rather there have been no discernible trends in the data recorded to date. A

trend that has been noticed, is that the participants who have used both intercropping and crop

rotation in their experimental blocks, have higher average values for these biological properties.

It is postulated here that the basic soil quality within these farmers’fields differ markedly due to

a combination of historical management practices, and natural variability and that the CA

management practices will even these differences out over time.

Conclusions

•The practice of CA has improved the physical properties of the soil over time, to the

extent that both water holding capacity and aggregate stability for the CA fields are

higher than for natural veld in the area (this is a high benchmark for comparison)

•The CA practices have also improved the pH and nutrient availability in the soil

(extractable P and K) to levels equivalent to and higher than the natural veld benchmark

Gravimetric water

The intention of doing the gravimetric water calculations is twofold;

1. To gain a visual representation of water availability in the soil for different cropping

options within the CA system and

2. To ascertain trends in water holding capacity in the soil, given the assumption that CA

and specifically multi- cropping options within the CA system improves the water

holding capacity of the soil.

Results from a gravimetric water content analysis in and of itself, cannot fully answer these

questions, as there are numerous factors at play and a much more in-depth analysis would be

required. This process has thus been exploratory in nature.

This process has been conducted for the last two seasons.

For the 2017-2018 season samples were taken for three participants (Phumelele Hlongwane,

Ntombakhe Zikode and Zodwa Zikode), for different crop combinations within the CA trials (M,

M+B, M+CP, SCC). The results were quite confusing and were only written up for one of the

participants- Phumelele Hlongwane.

This season only one set of soil samples (Phumelele Hlongwane) were taken for gravimetric soil

water assessments, given the time- consuming nature of this activity. These samples would give

an indication of soil water content at different depths (30cm, 60cm, 90cm and 120cm), at

different stages of crop growth, during the season. Samples were combined for her CA trial and

were also taken for a CA control and a conventional control plot.

Right and Far

Right:: Taking

the

gravimetric

soil samples in

Phumelele’s CA

trial plot, at

planting

(2018/11/07)

Below is Phumelele Hlongwane’s 1000m2 CA trial plot layout (2018/2019). Green shading

indicates plots where gravimetric sampling was done.

Plot 5

Plot 4

M+B

Plot 3

M+CP

Plot 2

M+CP

Plot 1

SCC

Plot 6

M+B

Plot 7

M+B

Plot 8

M+B

Plot 9

Plot 10

LAB LAB

Phumelele took a risk and planted a lot earlier in the season than most of the other farmers in

the area, who planted towards the end of November and early December only. Her crops

suffered considerably from the continued lack of rain and high temperatures prevailing during

November and December 2018.

Table 16: Gravimetric soil water sampling dates, compared to average monthly rainfall data

Gravimetric water samples taken

Date of sampling

Average rainfall

for the sampling

period

Planting (0 days)

2018/11/07

Establishment (4-6 leaf stage) (20-30 days)

2019/01/01

Vegetative growth (40-50 days)

2019/02/12

101

Productive stage (tasselling) (60-70 days) and

2019/03/14

212

Harvesting (physiological maturity) (80-110

days).

2019/04/25

150

The table above indicates the trend noticed by the farmers; that the rainfall during the

establishment and early vegetative growth stages of the crop was not enough to sustain growth

and rainfall towards the end of the season was unusually high, hampering maturation of the

crops.

Germination and early growth were hampered, but maize growth in the later vegetative stages

improved. Growth of the leguminous crops, specifically beans, was severely hampered, with

almost zero harvests recorded. Lab-lab (Dolichos) and cowpeas survived well, even under these

stressfull conditions. Of the summer cover crops the Sunhemp and millet (babala) survived well,

but sunflowers did not. The photos taken below for Phumllele Hlongwane are indicative.

Right to far-Right:

Growth of different

crops, towards the end

of the productive phase

(2019/04/11);

Dolichos, Sunhemp and

millet (Babala)

Right: Cowpeas grew

well, but because of

heavy rains in the

productive phase did

not seed well

Far Right: Maize

germination was patchy

and growth was

compromised. Late rains

caused a lot of damage to

cobs.

Comparison of gravimetric water content results for two seasons (Phumelele

Hlongwane –Ezibomvini)

For the 2017-2018 season, calculations for gravimetric water content between the different

cropping options were in fact very similar; meaning that the water content at the different

depths were similar within each of the cropping options. There were some interesting

differences between the cropping options.

The figure below indicates the results at 30cm depth.

Figure 1: Gravimetric water content at 30cm depth for different cropping options (Phumelele Hlongwane, 2017-2018)

From the figure above the following trends can be seen:

•At establishment, vegetative stage, productive and harvesting; for depths 30,60,90,120

the values are similar within each plot of the CA trial for 2017-2018, meaning the water

content of the whole profile was similar in each plot (results for 60cm-120cm are not

shown here)

•The water content for the plot planted to Lab-Lab beans (Dolichos) remained higher

than the other plots for most of the season. The assumption here is that the mulching

capability of the Dolichos reduced the evaporation and improved soil water content.

•The soil water content for the summer cover crops, Plot 9, was lower than for the other

cropping options in the trial plots for the entire season. This provides a reasonably clear

indication that the SCC used more water than the other crop combinations tested (Lab-

Lab beans, maize and cowpea intercrop and beans). For the vegetative and productive

growth period the measurements of 0,11 and 0,1 (g/g) of water to soil is considered

suboptimal for unimpeded growth.

•Generally, the CA control and the CA trial plots had similar gravimetric water content

readings for the season, indicating the water holding capacity of the soil is not changed

greatly by the particular cropping options within the CA farming system.

Establishment VegetativeProductiveHarvesting

30 Plot 5 (Lab lab)0.25 0.20 0.18 0.18

30 Plot 6 (M+CP)0.19 0.01 0.14 0.17

30 Plot 8 (B)0.19 0.13 0.14 0.16

30 Plot 9 (SCC)0.18 0.11 0.10 0.13

30 Control (M+B)0.17 0.36 0.13 0.16

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

Ave grav water (g/g)

Phumelele Hlongwane 30cm; 2017-2018

•The gravimetric water content for the maize and cowpea intercrop (Plat 6), indicates a

severe dip in water content in the soil during the vegetative growth phase. It is not clear

why this would be the case, but it could be an indication of temporary competition for

water between the maize and cowpeas in the vegetative growth stage –although the

severity of the result (0,01 g/g) would rather indicate an error in sampling and analysis.

In general, these results indicate that the water holding capacity of these soils under the CA

system of mixed cropping and crop rotation supported good growth of all crop combinations in

this season.

To compare the results of 2017-2018 with the present season (2018-2019), the results for all

trail plots were combined and averaged and were then compared to the CA control and a

conventional control (2018-2019 only). These results are shown in the two figures below.

Figure 2: Comparison of gravimetric water content results between 2017-2018 and 2018-2019 season, for CA trial and

control plots for Phumelele Hlongwane (Ezibomvini)

From the above figures the following observations can be made:

Establishment VegetativeProductiveHarvesting

30 CA trial0.20 0.11 0.14 0.16

30 CA control0.17 0.36 0.13 0.16

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

Ave grav atwr g/g

Phumelele Hlongwane 30cm; 2017-2018

PlantingEstablishment VegetativeProductiveHarvesting

CA trial0.13 0.69 0.09 0.19 0.11

CA control0.68 0.710.15 0.13

Conv control0.14 0.14 0.15 0.11 0.14

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

Ave grav water g/g

Phumelele Hlongwane 30cm; 2018-2019

•Overall the water content was lower at the beginning of the season and higher at the end

of the season for 2018-2019, when compared to 2017-2018. This trend follows the

rainfall patterns and ETo for these two periods.

•For the 2018-2109 season the water content for the CA control for the planting and

establishment phases is relatively high. It then dips sharply during the vegetative phase

(result missing)

•For the CA trial plot water content during the establishment stage is high and dips

sharply to a value below optimal growth during the vegetative growth stage.

•The gravimetric water content for the CA trial and CA control is higher during the

productive phase than the conventional control for the 2018-2019 –indicating potential

for better production from the CA plots.

•During the harvesting phase the water content for the CA trial plot for 2018-2019 is

lower than the two control plots. This is likely an indication of continued active growth

of the cover crops and lab-Lab beans planted in the trial.

The only conclusion that could confidently be drawn from these results is that the soil water

content of the vegetative growth stage in 2018-2019, for the CA trial and CA control pots was

well below the levels required for unimpeded crop growth. The high water content values are

not congruent with the rainfall and ETc data gathered for this season and are hard to explain –

unless per chance samples were taken very soon after rainfall events.

What can be seen from this figure is the following:

•There is a great reduction in water content in the soil, throughout the profile (30-120cm

depth) moving from the vegetative to productive stages and in fact there is too little

water in the soil during that period to sustain the crop growth as a gravimetric water

content in clay-loam soils of 0,11 -0,14 (g/g) is required as a minimum prior to wilting

point being reached

•The CA trial plots recovered well during the productive phase and indicate a higher soil

water content than both the control plots throughout the soil profile. This points

towards better water holding capacity in these soils linked to the multi cropping options

and shows also that the potential competition during the vegetative growth phase did

not continue into the productive phase

30 60 9012030 60 9012030 60 9012030 60 9012030 60 90120

PlantingEstablishment VegetativeProductiveHarvesting

CA control0.68 0.51 0.44 0.17 0.71 0.39 0.31 0.160.15 0.16 0.19 0.09 0.13 0.11 0.12 0.12

CA trial0.13 0.14 0.27 0.11 0.69 0.49 0.21 0.14 0.09 0.14 0.16 0.18 0.19 0.23 0.26 0.22 0.11 0.12 0.08 0.09

Conv control0.14 0.12 0.11 0.14 0.14 0.12 0.140.15 0.08 0.150.11 0.63 0.07 0.14 0.14 0.15 0.17 0.17

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

Ave grav water g/g

Phumelele Hlongwane 30-120cm; 2018-2019

•Towards the end of the season (harvesting stage) the deeper soil levels have dried out

considerably for the CA trial, more so than the control plots; indicating an increased

drying in the lower levels of the soil profile for the multi-species CA trial. This is likely

due to the continued growth of the Lab-Lab beans and cover crops, which were not

present in the control plots.

Overall, for both seasons, the gravimetric soil water content of the CA trials are somewhat lower

than the CA control plots. This indicates that the multi-cropping options used in the CA trial use

more water than a monocropping option (such as used in the CA controls). This result is not

unexpected. There is also an indication that the multi-cropping led to decreased water

availability during the vegetative growth phase for the 2018-2019 season, which could in turn

affect the maize yields for this season. The beans intercropped with maize died back during this

period and no yields have been recorded. Cowpeas however, survived well. This provides a

good indication of the drought tolerance of cowpeas. For the summer cover crop combination,

sunflowers also died during this vegetative growth phase due to water shortages, but the millet

and Sunhemp survived well and seeded. Interestingly the water content is much improved for

the CA trial when compared to the CA and conventional controls- indicating a good recovery for

the CA trial plots in this phase

Bulk Density

Below is a summary of the results of the bulk density calculations for different cropping

practices within the CA system of the three participants. They were chosen for having differing

period of cropping under CA and for inclusion of a number of practices within their CA system;

namely intercropping and planting of summer cover crops (SCC).

Table 17: Bulk density (pb) results for three CA participants

Village

Period

under CA

(yrs)

Name and

Surname

Control CT

Control CA

M+B

M+CP

SCC

Average

Ezibomvini

Phumelele Hlongwane

1,30

1,36

1,38

1,33

1,38

1,28

1,34

Eqeleni

Ntombakhe Zikode

1,35

1,49

1,37

1,32

1,38

Thamela

Mkhuliseni Zwane

1,14

1,08

1,09

1,07

1,10

Average bulk density

1,27

These results indicate an increase in ρb over the period of involvement in CA. This trend is

expected. There is little to no difference between the CA practices, although in all three cases the

planting of SCC has reduced the ρb fractionally.

Soil Health

This season soil health analysis was undertaken for 10 participants across five villages in Bergville;

•Eqeleni (2) Stulwane (2); 6th year of implementation; when was sample taken??

•Ezibomvini (2); 5th year of CA implementation; when

•Mhlwazini (2); 3rd year of CA implementation; when

•Ndunwana (2); 3rd year of CA implementation; when

The intention is to compare the soil health characteristics for a number of cropping options within the CA trials, with conventionally tilled mono-

cropped control plots, over time.

The Haney soil health tests (as analysed by Soil Health Solutions in the Western Cape and Ward Laboratories in the USA) provides insight into

microbial respiration and populations in the soil, organic and inorganic fractions of the main nutrients N, P and K, and assessment of organic carbon

percentage organic matter (%OM). An overall soil health score (SH) is also provided for each sample.

Haney Soil health tests parameters

These analyses are benchmarked against natural veld for each participant, due to high local variation in soil health properties, measured at different

times. The veld scores provide for high benchmarks to compare the cropping practices against.

Soil Respiration 1-day CO2-C: This result is one of the most important numbers in this soil test procedure. This number in ppm is the amount of

CO2-C released in 24 hours from soil microbes after soil has been dried and rewetted (as occurs naturally in the field). This is a measure of the

microbial biomass in the soil and is related to soil fertility and the potential for microbial activity. In most cases, the higher the number, the more

fertile the soil.

Microbes exist in soil in great abundance. They are highly adaptable to their environment and their composition, adaptability, and structure are a

result of the environment they inhabit. They have adapted to the temperature, moisture levels, soil structure, crop and management inputs, as well

Haney/Soil Health Test Information Rev. 1.0 (2019). Lance Gunderson, Ward Laboratories Inc.

as soil nutrient content. Since soil microbes are highly adaptive and are driven by their need to reproduce and by their need for acquiring C, N, and P

in a ratio of 100: 10: 1 (C:N:P), it is safe to assume that soil microbes are a dependable indicator of soil health. Carbon is the driver of the soil

nutrient-microbial recycling system.

Water extractable organic C (WEOC):Consists of sugars from root exudates, plus organic matter degradation. This number (in ppm) is the

amount of organic C extracted from the soil with water. This C pool is roughly 80 times smaller than the total soil organic C pool (% Organic Matter)

and reflects the energy source feeding soil microbes. A soil with 3% soil organic matter when measured with the same method (combustion) at a 0-3

inch sampling depth produces a 20,000 ppm C concentration. When the water extract from the same soil is analysed, the number typically ranges

from 100-300 ppm C. The water extractable organic C reflects the quality of the C in the soil and is highly related to the microbial activity. On the

other hand, % SOM is about the quantity of organic C. In other words, soil organic matter is the house that microbes live in, but what is being

measured is the food they eat (WEOC and WEON).

If this value is low, it will reflect in the C02 evolution, which will also be low. So less organic carbon means less respiration from microorganisms, but

again this relationship is unlikely to be linear. The Microbially Active Carbon (MAC = WEOC / ppm CO2) content is an expression of this relationship.

If the percentage MAC is low, it means that nutrient cycling will also be low. One needs a %MAC of at least 20% for efficient nutrient cycling.

Water extractable organic N (WEON):Consists of Atmospheric N2 sequestration from free living N fixers, plus organic matter degradation. This

number is the amount of the total water extractable N minus the inorganic N (NH4-N + NO3-N). This N pool is highly related to the water extractable

organic C pool and will be easily broken down by soil microbes and released to the soil in inorganic N forms that are readily plant available.

Organic C: Organic N: This number is the ratio of organic C from the water extract to the amount of organic N in the water extract. This C:N ratio is a

critical component of the nutrient cycle. Soil organic C and soil organic N are highly related to each other as well as the water extractable organic C

and organic N pools. Therefore, we use the organic C:N ratio of the water extract since this is the ratio the soil microbes have readily available to

them and is a more sensitive indicator than the soil C:N ratio. A soil C:N ratio above 20:1 generally indicates that no net N and P mineralization will

occur. As the ratio decreases, more N and P are released to the soil solution which can be taken up by growing plants. This same mechanism is

applied to the water extract. The lower this ratio is, the more organisms are active and the more available the food is to the plants. Good C:N ratios

for plant growth are <15:1. The most ideal values for this ratio are between 8:1 and 15:1.

Soil Health Calculation: This number is calculated as 1-day CO2-C/10 plus WEOC/50 plus WEON/10 to include a weighted contribution of water

extractable organic C and organic N. It represents the overall health of the soil system. It combines 5 independent measurements of the soil’s

biological properties. The calculation looks at the balance of soil C and N and their relationship to microbial activity. This soil health calculation

number can vary from 0 to more than 50. This number should be above 7 and increase over time.

Some of the inter relationships between these variables are explored below

Figure 3: Comparison of the SH scores for Bergville participants (N=10) with microbial respiration and organic carbon.

The general assumption here is that if the level of organic C in a plot is high, then the microbial respiration will also be high, as will the soil health

scores and vice versa. This is not always the case, as the relationship is not necessarily a linear one.

The CO2-C respiration also gives an indication of the potential mineralisation of N for the soil as well as organic matter content. The small table below

indicates these relationships.

M+B VeldB Conv

C SPSCC Labl

ab MM+B M+C

PSCCVeld Conv

C SPM+B Veld M+C

PSCCVeld Conv

C MMM+B M+C

PSCCVeld Conv

C MB

EqeleniEzibomviniMhlwazini NdunwanneStulwane

Average of CO2 - C, ppm C129. 277. 91.3 48.4 83.1 68.5 24.7 82.3 126. 129. 137. 49.7 110. 174. 195. 143. 63.6 396. 136. 80.4 169. 265. 307. 158. 75.9

Average of Organic C ppm C179. 250. 241. 163. 276. 237. 187. 185. 176. 126. 261. 198. 224. 196. 236. 213. 167. 492.251. 142. 192. 264. 307. 225. 122.

Average of Soil health calculation (new)13.7 26.1 15.69.315.7 13.17.612.6 16.2 14.2 17.7 10.5 16.0 19.7 21.2 17.5 10.9 35.8 18.1 11.7 17.9 26.0 28.3 19.2 10.7

0.0

100.0

200.0

300.0

400.0

500.0

600.0

Axis Title

SHo scores,CO2 respiration and organic carbon; Bergville 2018-2019

Test results ppm CO2-C

N mineralisation potential

Biomass

>100

High-N potential soil. Likely sufficient N for most

crops

Soil very well supplied with organic matter.

Biomass>2500ppm

61-100

Moderately-high. This soil has limited need for N

supplementation

Ideal state of biological activity and adequate organic matter

31-60

Moderate. Supplemental N required

Requires new applications of stable organic matter. Biomass

<1200ppm

6-30

Moderate-low. Will not provide sufficient N for most

crops

Low in organic structure and microbial activity Biomass

<500ppm

0-5

Little biological activity; requires significant

fertilisation

Very inactive soil. Biomass<100ppm. Consider long term care

For the above figure the following trends can be seen:

➢All the CA samples for all five villages fall within the >100ppm and 61-100ppm C02—C respiration categories; indicating adequate to high

levels of organic matter, an ideal state of biological activity and a moderate to high N- mineralisaton potential.

➢The two Conventional tillage samples (sweet potato) fall within

the moderate category where addition of organic matter is

required as well as supplemental N. the Conventional maize

control for Ndunwana however has extremely high respiration

and organic carbon values –This value is somewhat of a

mystery- as the benchmark veld samples for Ndunwana are quite

low. The fact that it is a newly tilled plot, leading to very high

microbial activity, especially bacteria, and the release of

nutrients from the organic matter in the soil, might be the best

explanation of the result.

In conclusion the soil health status of the CA trial plots are moderately

high to high, with good organic matter content and ideal states of

biological activity, as indicated in the small figure alongside. The highest

values for %OM are for the M+CP and SCC plots –which confirms the

observations that these crop combinations are the best at improving soil

health in the short term.

BConv C

SP LablabMM+B M+CPSCCVeld

Total 4.8 2.8 3.2 3.8 3.9 4.9 4.8 6.2

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

% OM

% OM for different crop combinations, Bergville

2018-2019

Figure 4: % OM for different CA crop combinations in Bergville; 2018-2019

Below is a comparison of the soil health status for Ezibomvini across two seasons.

Figure 5: Comparison of Soil health indicators for Ezibomvini across two cropping seasons; 2017/18 and 2018/19

Average

of % OM

Average

of CO2 -

C, ppm C

Average

Organic

C ppm C

Average

Organic

N ppm N

Average

of C:N

ratio

Average

of Soil

health

calculati

on (new)

Cont M3.873.1 233.5 19.112.613.9

M+B 4.769.9 243.5 22.211.213.2

SCC 4.073.7 263.3 20.313.114.0

Veld 3.984.8 285.3 17.816.315.2

0.0

50.0

100.0

150.0

200.0

250.0

300.0

Soil health Ezibomvini 4th yr (N=3)

Average

of % OM

Average

of CO2 -

C, ppm C

Average

Organic

C ppm C

Average

Organic

N ppm N

Average

of C:N

ratio

Average

of Soil

health

calculati

on (new)

Conv C SP2.949.7198.015.8 12.5 10.5

Lablab 3.268.5237.014.7 16.1 13.1

M3.624.7 187.0 13.813.67.6

M+B 3.682.3185.714.4 12.8 12.6

M+CP 4.5 126.2176.515.1 11.7 16.2

SCC 5.3129.6 126.08.814.314.2

Veld 5.8 137.0261.715.4 17.0 17.7

0.0

50.0

100.0

150.0

200.0

250.0

300.0

Axis Title

Soil health Ezibomvini 5th yr (N=3)

Legend: CONV C SP - conventional control sweet potatoes, LabLab - Dolichos lab lab beans, M - Maize, M+B -maize and bean intercrop, M+CP - Maize and cowpea intercrop, SCC -summer

cover crop mix (millet, Sunhemp and sunflower)

When comparing the two graphs (4th and 5th year) above, it can be seen that the soil health scores (SH) are comparable for the CA cropping options

across the two seasons compared:

➢SCC (5th); SH=14,2 and SCC (4th); SH =14,0

➢M+B (5th); SH=12,6 and M+B (4th); SH=13,2

The SH score for the veld samples however differ quite a lot, mainly due to a difference in measured Organic C and Organic N. IN addition, these

samples were not taken in exactly the same place every year –although they were taken in the veld directly adjacent to the homesteads. What this

means, is that the soil health parameters vary in the veld as well, depending on where the sample was taken, also over a very small area. It is thought

that these differences can be smoothed out by taking the veld samples in different places around the homestead and combining them into one

sample. This will be attempted in future seasons.

In general, the Organic C and Organic N values for the 4th year are markedly higher than those measured for the 2018-2019 season. But the microbial

respiration values for comparable CA samples (M+B and SCC) are markedly higher for the 5th year. While the flux and flow of organic nutrient

availability and microbial growth are quite complex, with many interrelated parameters, the trends in decrease in organic C and N are considered to

be related primarily to a slow, but definite drying of the soil profile over the last two years. The trend towards increased microbial activity in the

multi-cropped (M+B, M+BP and SCC) and legume (Lab-Lab) plots in the 5th year clearly indicate the value of these practices for sustained soil health

under conditions of climate variability (late onset of rain, variable rainfall and increased temperatures)

For 2018-2019 (5th year) the soil health results indicate the following trends:

➢The average % OM is higher for all the CA cropping options when compared to the conventional control. Plot 1 (SCC CA treatment, 2018/19)

has a value close to that of the natural veld sample, indicating the greatest build- up of organic carbon for this cropping system within a

season. This trend was also noticed for the 2017-2018 cropping season (4th year), where the SCCs were planted in Plot 9

➢The microbial respiration is highest for the SCC CA plot, followed by the maize and legume (cowpea, bean) intercropped plots and Lab-Lab

beans and is lowest for the mono-cropped maize. A similar trend was noticed for the 2017-2018 cropping season (4th year).

➢The average organic N is the highest for the three CA plots containing legumes (Lab-Lab, M+B and M+CP)(plots 10, 4 and 2 respectively).

And lowest for the SCC plot. A similar trend was noticed for the 2017-2018 cropping season (4th year).

➢A low C:N ratio is considered beneficial for nutrient availability for crop growth. The lowest values are found for the CA intercropped plots

(M+B and M+CP), followed by the CA maize plot. Again the trend is similar to the 2017-2018 results

The Conventional control plot showed the highest average organic N value (15,8ppm).

As mentioned above in the discussions around soil water content and water holding capacity, finding appropriate controls to compare the CA results

against, has been a challenge. This season a conventional control plot was chosen where increased tillage and mono-cropping is practiced. The plot

was planted to sweet potatoes. We however, did not take into account the historical land use of this plot, so while the lower % OM and microbial

respiration was expected, the higher levels of organic N were not. We have not compared the CA and conventional plots directly for this reason.

In addition, the CA maize plot for 2018-2019 (5th year), shows a very low microbial respiration rate, despite having reasonably high organic C and

Organic N values. The understanding here is that there are localised differences in soil qualitybetween the 10x10m CA plots in Phumelele

Hlongwane’s field that have reduced these values considerably. These differences are not directly related to the multi-cropping and crop-rotation

practices for the CA trial, but are more likely due to a lower microbial count, or localised soil pathogens. This was reported on in the 2016-17 report,

where a supplementary soil pathogen study conducted by the ARC showed high levels of root and crown rot fungal species in her CA plots; notably

Fusarium and Phoma species.

. The data indicated that the severity of root rots is higher in the CA plots than the conventionally tilled plots.

The local variations in soil health quality for different plots within the same field are due to soil management practices bythe farmer prior to starting

their CA experimentation combined with the combination of intercropping and crop rotation practices used during the CA experimentation process.

Repeat soil health samples are taken from the same plots within each participant’s field. A sampling set of15-20 cores is sued for each plot ; meaning

that these 20 sub-sub samples per plot are combined to provide for one reading for the plot. As these participants use both intercropping and crop

rotation, the crops in these specific plots are not the same every year. The assumptions used here are the following:

1. That a specific crop combination in a plot has an effect on the soil health; e.g. maize vs an intercrop or cover crops

2. That the history of the plot, in other words the rotations used also have an effect on the soil health and thus

3. That overtime the soil health scores across the plots will similar as they slowly increase.

To date, there is however still a high level of variability between the plots. This is explored in a little more detail below.

The table below indicates Phumelele’s rotations in the last four years.

Agricultural Research Council. Plant Protection Research Institute. P/Bag X134, Queenswood, Pretoria 0121. Preliminary Consultation Report-Analyses Of

Soil borne Diseases Of Maize, Soybean And Sunflower –Soil Health Project. Prepared by: Dr Sandra Lamprecht and Thabo Phasoana. Tel: (021) 887 4690

Fax: (021) 887 5096. Email: lamprechts@arc.agric.za

If one now compares the soil health results for the specific plots within Phumelele’s trial, across a number of seasons, it is expected that the soil

health scores between the plots should even out and become more similar over time. These scores are also expected to increase over time. This is

shown in the figure below

Plot

2015/16

2016/17

2017/18

2018/19

Run off plots

M+B

M +WCC

SCC

Shaded cells indicate runoff plots

SCC

M+B

M+CP

Rotations have been done attempting to ensure a

different crop/crop mix on each plot in each consecutive

year.

A further refinement of the schedule to be a 3- year

rotation of; single crop –intercrop- cover crop, will be

adhered to into the future

M+SCC+WCC

M+B

MCP

M+B

M+LL

SCC

M+CP

M+B

M+CP

M+B

M+CP

M+B

M+CP

M+B

SCC

M+B

11?

CA Control:

CA Control M

12?

Control: M

(CA)

CA Control:

M+B (CA)

Conventional

control: SP

Figure 6: Soil health scores for a number of plots in Phumelele Hlongwane’s CA trial between 2016-2018

From the figure above the following can be seen:

➢The SH scores have not necessarily increased over the three seasons measured; it has increased for one of the plots (pot 10), decreased or

another (Plot 5), stayed the same for another plot (Plot 2) and has been variable for another (Plot 4), with similar differences not in the

microbial respiration.

➢It the SH scores are average for the plots then the veld sample has the highest SH score (17,73), followed by plot 2 (13,8, with all other plots,

including the control plots being in the range of 11,6-11,9. This does then indicate a tendency for the SH scores to even out between the plots

over time.

➢The fact that the CA control and Ca trial plot average soil health scores are the same; appears to indicate that the intercropping and rotations

used have a smaller effect on soil health than the practice of CA (minimum tillage)

This means that the assumption that the soil health scores will even out over time is correct, but that there are local and seasonal variations that can

mask this trend. Thus far the overall increase in soil health scores expected from the intercropping and crop rotation is not evident.

VeldVeldVeldPlot 2Plot 2Plot 4Plot 4Plot 4Plot 5Plot 5Plot 5Plot 10Plot 10

Cont M

(ConvSP

)

Cont M

(CA)

Cont M

(CA)

2018 2017 2016 2018 2016 2018 2017 2016 2018 2017 2016 2018 2017 2018 2017 2016

CO2 - C, ppm C141.8 81.6113 82.4 78.4 48.3 77.4 54.5 24.7 45.5 90.2 68.5 30.8 49.7 59.6 62.7

SH 20.116.52 16.5713.813.810.613.81 10.877.612.73 14.4513.110.8410.513.36 11.54

100

120

140

160

SH scoress for Phumelele Hlongwane accroding to plots within her trial: 2016-2018

Using the soil health test results, it is also possible to explore the composition of the microbial population in the soil, looking at the different types of

microorganisms and their prevalence using data from the PLFA test.

Generally, it is known that conventional tillage systems favour decomposer/saprophytic fungi, with small hyphal networks. These are important in

soil fertility but play a very small role in carbon storage. Conservation Agriculture systems favour Mycorrhizal fungi which have large hyphal

networks and play a major role in carbon storage. Mycorrhizal fungi get their energy in a liquid form, as soluble carbon directly from actively

growing plant roots. They access and transport water - plus nutrients such as phosphorus, nitrogen and zinc - in exchange for liquid carbon from

plant roots. Soluble carbon is also channelled into soil aggregates via the hyphae of mycorrhizal fungi and can undergo humification, a process in

which simple sugars are made up into highly complex carbon polymers. Aggregate stability is thus an important emerging quality of the soil under

CA. It is measured as % volumetric stability, as shown in the small table below.

From the soil health, microbial respiration and organic carbon data for Ezibomvini and Ndunwana, the expectation is that aggregate stability will be

good to excellent. This is indeed the case for Ndunwana (as shown in Figure 5), where the values range from 45-46,5%. For Ezibomvini however,

there is a range of values from low, to average and good. This would mean, among other things, that the soil health status, including Mycorrhizal

fungi populations in the Ezibomvini soils are not building up as expected and shows high variation between plots (within one field), as discussed in

the section above.

Volumetric Aggregate stability %

0 - 15 %

15 - 30 %

30 - 45 %

45 - 60%

> 60%

Very low

Low

Average

Good

Excellent

Figure 7: A comparison of % aggregate stability for soil health samples from Ezibomvini and Ndunwana

The PLFA analysis conducted and presented below, sheds some light on this.

PLFA analysis

PLFA (Phospholipid fatty acid) analysis of the microbial populations in the samples provides a breakdown (snapshot) of the type of organisms and

their ratio’s present, e.g. bacteria, fungi and protozoa, as well as their relative abundance. This is based on the different and distinguishable

biochemical structures and processes for these organisms. Although this analysis can get very complex, two simplified snapshots of the process are

provided in the figures below.

Conv

C SP

Labla

bMM+BM+CP SCCVeldM+CP SCCVeld

Ezibomvini Ndunwanne

Average of Soil aggregates25.0 32.0 45.0 25.3 31.5 27.0 42.0 46.5 45.0 40.0

Average of Soil health calculation

(new) 10.5 13.17.612.6 16.2 14.2 17.7 21.2 17.5 10.9

Average of CO2 - C, ppm C49.768.524.782.3126.2 129.6 137.0 195.7 143.763.6

0.0

50.0

100.0

150.0

200.0

250.0

Axis Title

Soil Health scores compared to soil aggregates and CO2-C

for Ezibomvini and Ndunwana 2018-2019

Figure 8: PLFA results for microbial populations from Ezibomvini and Ndunwana soil health samples; Bergville 2018-2019

From the above figure on PLFA results the following trends can be seen:

➢Mycorrhizal fungi populations for the CA maize (Mtrial) in Ezibomvini are extremely low, when compared to the veld sample and the

samples from Ndunwana; although the Mycorrhizal populations are quite small when compared to the overall microbial populations present

in these sites.

➢For the Ezibomvini samples the total microbial biomass for the Mtrial sample is lower than the Conventional control sample. This low

microbial mass is not reflected in the %OM (3,65) or the organic carbon (187ppm) and organic nitrogen (13,8ppm) content of the plot; these

values being quite high. This means that the microbial biomass in this particular plot is being dampened for another reason, the most likely

being disease, shown in the 2nd graph in Figure 6 above. From this and other analyses done, it would appear that this situation is specific to

this plot (and perhaps 2 others) in Phumelele Hlongwane’s CA trail.

➢Mycorrhizal fungi populations in the CA trial plots (Maize and SCC) are considerably higher than the veld benchmark for Ndunwana,

indicating the expected build-up of these fungi in the CA cropping system

MtrialVeldConv C SPMtrialVeldSCC

Ezibomvini Ndunwana

Average of Arbuscular Mycorrhizal9.4170.8 140.9 217.468.179.4

Average of Saprophytes104.6 982.4 305.3 581.5 253.7 264.1

Average of Actinomycetes biomass208.4 376.7 253.2 352.8 223.5 259.4

Average of Rhizobia0.0 126.0 0.021.30.00.0

Average of Protozoa biomass0.029.7 17.9 44.7 13.08.8

Average of Bacteria biomass1504.0 2257.3 2015.7 2679.2 1433.2 1917.0

Average of Fungi114.01153.2446.2 798.9 321.8 343.4

0.0

500.0

1000.0

1500.0

2000.0

2500.0

3000.0

ng/g

PLFA results ; avreage of different microbial popluations for Ezibomvini and Ndunwana 2018-

2019

If is also possible to analyse the ratios of the different organisms to each other

Fungi:Bacteria; Bacteria tend to dominate in systems with fewer organic inputs or residues possibly leading to a

lower C:N ratio. In addition, bacteria can be more prominent in the early spring or late fall as soil temperatures are

usually cooler and vegetation is less active or absent. Dry conditions, or increased land disturbance through

prolonged and extensive tillage, grazing, or compaction may also favour bacteria. While bacteria are important and

needed in the soil ecosystem, fungi are desired and more often considered indicators of good soil health. Increased

use of cover crops and/or other organic inputs and less soildisturbance should help the soil support more fungi.

Gram+ve:Gram-ve;Gram+ve bacteria typically dominate early in the growing seasons and/or following a fallow

period. They also survive better under certain environmental conditions or stressors such as drought or extreme

temperatures due to their ability to form spores. Therefore, it is common to see higher values when a community is

coming out of dormancy or is stressed. These values will typically become more balanced as the soil conditions

become more favourable throughout the growing season. A Gram-ve dominated soil may be due to anaerobic

conditions or other stressors such as pesticide application or heavy metal contamination

Predator:Prey; This ratio is also expressed as protozoa to bacteria. Protozoa feed on bacteria, which helps release

nutrients, especially nitrogen. A higher ratio indicates an active community where base levels of nutrients are

sufficient to support higher trophic levels of predators. However, this ratio will always be relatively low as prey

generally greatly outnumber prey.

The figure below indicates these ratios for the soil health samples taken in Ezibomvini and Ndunwana

Figure 9: PLFA results; ratios of different types of organisms for Bergville 2018-2019

It can be seen that:

➢The fungi:bacteria ratios for the CA Mtrial plots are above average to good. The low ratio for the Convention control plot (Conv C SP) is

expected, given the lower organic matter content and tillage of this plot.

➢The Gram+ve;Gram-ve ratios for all the samples veld control and trial plots fall between 1-2 and indicated a balanced bacterial community.

➢The predator:prey ratios show some variation: From Phumelele Hlonwgane (Ezibomvini) the ratios for the M trial plot are average and

those for her Conv C are very poor. This is also indicated in Figure 8 above showing very low populations of both rhizobia and protozoa in

Phumelele’s soilThe occurrence of natural strains of Rhizobia in soils are affected by pH, tillage, high temperatures and chemical residues.

As Phumelele has given attention to liming, reduced use of chemicals and has used CA now for 6 seasons running, it is unclear why these

populations are so depleted. It is suggested that introduction of rhizobia strains for root nodulation and nitrogen fixation by legumes will

have a highly beneficial effect.

➢For Lethiwe Mofukeng (Ndunwana) both her Mtrial and SCC plots show a very good ratio.

Conv C SPMtrialVeldMtrialSCCVeld

Ezibomvini Ndunwana

Average of Fungi:Bacteria0.08 0.22 0.510.30 0.18 0.22

Average of Gram(+):Gram(-)1.77 1.00 1.49 1.24 1.79 1.58

Average of Predator:Prey00.0089 0.0132 0.0167 0.0046 0.0111

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

1.80

2.00

ng/g

PLFA results; ratios of different organism types; Bergville 2018-2019

Nitrogen

In the dryland cropping system around Bergville, as in most other dryland cropping areas in South Africa, supplementation with inorganic Nitrogen

is considered an important strategy for optimal crop growth. In our CA study different crop combinations and cropping options are being explored

to assess the potential of providing this nitrogen through improvement of natural nutrient flow cycles. Inorganic N, besides being expensive, also has

been shown to dampen the natural microbial activity in the soil and can also be partially ineffective under extreme conditions of drought and heat.

An analysis of immediate release N has been done, as well as an estimation of the rand value of inorganic nitrogen saved /ha for different cropping

options under CA. The immediate release N- is the water extractable organic Nitrogen, which is immediately available to the next crop.

Figure 10:Comparison of immediate release N and Rand value of

inorganic Nitrogen substituted for organic N for 5 villages in Bergville;

2018-2019

From this figure the expected progression of increase in

available N from a CA maize monocrop –a summer cover

crop mix to a maize and bean intercrop –a maize and

cowpea intercrop is clearly visible. The CA beans only plot

has a somewhat unexpectedly low result. On average the

rand value of inorganic N saved in this process is

R318/ha. If a recommendation of 60Kg/ha of N is used,

this equates to a saving of around 47% on inorganic

fertilizer –more specifically for the plots that integrate

legumes (M+B, M+CP and Lab-Lab beans). The average

rand value for inorganic N saved in the previous season

(2017), was R393. It is assumed that this value is higher

because of the higher soil water content (better soil water

distribution in the soil profile throughout the season).

This indicates the effect of heat and dry soil profiles on the ability of the soils to process and maintain nutrients.

BConv C

SP LablabMM+B M+CPSCCVeld

Average of N Immediate release25.0 30.5 33.0 26.5 30.7 32.2 28.7 34.0

Average of R value of Org N278.50 341.00 369.00 299.50 346.29 364.83 320.33 379.56

0.0

50.0

100.0

150.0

200.0

250.0

300.0

350.0

400.0

Immediate release N and Rand values of N for different CA cropping

options in Bergville; 2018-2019

Comparison of SH test results 2015-2018

One can compare the soil health data for the different participants over time to track improvement in soil health scores. The assumption is that soil

health will improve over time with CA implementation. The figure below summarises the data for five participants between 2015/16 to 2018/19

Figure 11: Soil health data for 5 participants from Bergville;2015-2018

From the above figure the following trends are visible:

➢Soil health scores have increased (although not linearly) between 2015-2018 and the average SH score for 2017/18 is 18,74.

➢Despite the fluctuations in CO2_C (microbial respiration) organic carbon and organic nitrogen in the four years of measurement, the overall

values have increased substantially since 2015.

CO2 - C(ppm)Organic C (ppm)Organic N (ppm)C:N ratioSoil health calculation

2015 132.7 118.5 11.6512.113.6

2016 72.95198.55 14.02514.558.525

2017 86.2 280.5 19.414.015.2

2018 149.68 227.0214.6216.5818.74

100

150

200

250

300

SH scores for bergville 2015-2018

➢Interestingly the C:N ratio has been systematically increasing –rather than the expected reduction. It indicates a higher proportional

increase in organic carbonin the soil, as compared to organic nitrogen through the CA practices employed in the programme. It also shows

that improvement of soil Nitrogen is still an issue, despite the legume based intercrops and cover crops that have been used.

The extreme climatic conditions in the area, including heat and dry soil profiles, reduces the soil health impact of the CA practices and also increases

variability in the results for different seasons

Implementation snapshots across different villages

Information provided in the interim report is not repeated here. Snapshots are provided for a

few of the different villages.

Ngoba

Ngoba is a village now in its fourth year of programme participation with eleven members in the

group. The village practicescrop rotation, intercropping and cover crop experimentation. Mam

Vimbephi Dladla is one of the participants of the group. Her trial was planted 28 November 2018

andshe sharedthat while her trial suffered during the drier period, the CA trial has thrived under

the unexpected late rains in March-April 2019. She feels that one of the main reasons for this is

the CA planting, that has helped crops to withstand these harsh conditions.

Thamela

Constance Hlongwane is the local facilitator for theThamela learning group,which is now in their

third year of programme participation. Her trial has performed poorly this season and she

attributes this to both dry conditions and late planting. Her crops are characterized by stunted

growth, yellowing and poor germination. She wasunable to replant in plots where germination

was poor due to lack of planting material.

Above Left & Centre-Pictures of Mam Constance fields showing signs of yellowing and stunted

growth. Right- This plot has a contrast of yellowing and green leaves.

Ndunwana

The Seasonal conditions also caused a lot of late season weed growth, which some participants

were unable to manage well. The pictures for Lethiwe Zimba below are indicative.

Above left and right; Weed infestation in Letiwe Zimba’s trial plots for summer cover crops and

maize.

Late beans

Farmers in the villages of Ndunwana, Thamela, Stulwane and Ezibomvini planted experiments of late

beans in early March. They planter200 m2 plots,of which100 m2was planted to dry beans PAN 148

withfertilizer(LAN)and 100 m2was planted incorporatingbothfertilizer and lime. Fertilizer was

applied at a rate 40 kg/ha and lime at 1.7 t/ha.

Table 18: Showing participants for late bean experimentation

Village

Participant name

Ndunwana

Lethiwe Zimba

Thamela

Constance Hlongwane

Ezibomvini

Nombono Dladla

Zodwa Zikode

Phumelele Hlongwane

Landiwe Dlamini

Stulwane

Phasazile Sithebe

Hluphizwe Zondo

Dombi Dlamini

Constance Hlongwane–Thamela

Constance Hlongwane planted her late bean trial on the 7th of March

2019 and sprayed with Roundup the following day, due to high weed

pressure. Monitoring showed even germination of the beans and

subsequent better growth for the trial plot that included lime.

Right: Even germination for Constance’s late bean trial

Lethiwe Zimba -Ndunwana

A similar result was found for other participants

including Lethiwe Zimba from Ndunwana,

although her germination was somewhat patchy.

This is considered an effect of Roundup which can

reduce germination of legumes, especially when

sprayed after planting.

Right: Somewhat patchy germination of

Lethiwe’s late bean trial.

Cover crops

For the majority of participants who planted summer cover crops (around 56 participants),

growth was compromised due to the weather conditions and late season infestation of weeds.

Very few participants (around 8

in total) harvested seed from

these and then only for the

Sunhemp, fodder sorghum and

millet (Babala). Sunflowers did

not survive to maturity.

Right: Sunhemp and fodder

sorghum cover crops seedinng in

Ntombakhe Zikode’s field

(Eqeleni)

Far right: Millet seeding in

Phumelele Hlongwane’s field

(Ezibomvini)

Legumes such as lab-lab (dolichos) and cowpea also fared reasonably well under these harsh

conditions.

Winter cover crops were only

attempted by a handful of

participants this year. Those

planted by Ntombakhe in

Eqeleni did surprisingly well.

Right: Phumelele’s Lab-Lab plot

growing well and

Far right: Winter cover crops for

Ntombakhe

Given the extremely erratic performance of the cover crops this season, obtaining of yield data

has not been attempted.

Fodder production and supplementation

As mentioned in the interim report, most of

the fodder species planted (including

Lucerne, white and red clover and turnips)

did not germinate. Only two participants

managed a reasonable harvest of Teff, which

was one of the few crops planted that grew.

Right: MT visitors standing on Ntombakhe’s

Teff plot, for which the grass was cut and put

into bags for provision of fodder to livestock

In addition to attempting to grow different

fodder crops, the learning group members

have been provided with input on fodder

supplementation and making of bales to

improve their winter fodder supply for their

livestock

During the fodder and supplementation

learning workshop (April 2019) a number of

topics were covered including:

-Why are livestock thin in winter and

why does it matter

-What to feed and how much

-Supplementation

-How can we tell if it’s making a

difference (condition scoring sheet)

-Making hay

-Experimentation

Right; Participants in the fodder and supplementation workshop

learn about using the condition scoring sheet to monitor their

livestock and supplementation experiments.

The major issue in winter, or more generally is protein.

Carbohydrates can be found in grass and maize. The type of

grass makes a difference; Themida (Rooigrass) is good, other

grasses such as Ngongoni (Aristida), Mtshiki and Uqunga are not

very palatable.Urea supplementationcan be done using a

powder (Premix 450) or liquid (LS33) and protein licks.

Cutting of grass and making of hay bales was discussed and

small mechanical balers were provided to the group for their

experimentation process.

Right: the small balers to be used for making bales.

For the experimentation process participants undertook to do the following:

-Collect monies for the supplements

-Have these supplements, as well as rolls of twine for the bales available through the

farmer centres

-Set up and participate in condition scoring days for the livestock; the first one in mid-

End June and to

-Ensure that they have a viable experiment; cattle in the experiment and cattle not in the

experiment

In Ezibomvini and Stulwane farmers have been preparing

for the fodder supplementation experiments undertaken in

early June. They have cut grass for baling and will now start

to make bales, as the 2nd baler has been delivered; meaning

there is one baler for each of the respective areas. The idea

was the farmer centresin these two villages would procure

and supply the premix and the LS33. This has worked well

for the protein blocks as well. For the LS33, there was none

available from their closest town for a period and thus they

have only now bought this liquid supplement.

Right: Dried grass ready for baling at Phumelele

Hlogwane’s homestead.

Farmers have approached the experimentation process a

little haphazardly –feeding all their cows every now and

again, rather than having a more controlled

experimentation process. The idea was thus re-introduced.

There is also the issue that the few bales that they will be

able to make (usually not more than 10 per participant),

are not likely to last long, and thus their attempts at

introducing the supplements directly. This will of course be

rather difficult with the liquid version (LS33), but has in

fact been working quite well with the pre-mix 450.

Right: Phumlele measuring out the Premix 450 to feed to her

two cows directly.

The table below outlines the fodder and supplementation

experiments now being undertaken .

Table 19: fodder experimentation participants from Stulwane, Ezibomvini and Eqeleni.

Experiment

Names

Comments

Bales with LS33

Mtholeni Dlamini

Phumlani Dladla

Mkhathini Dladla

Phumelele Hlongwane

Ntombenhle Hlongwane

They have now undertaken to feed 2-3

pregnant or lactating cows with the bales

mixed with LS 33 and to have the rest of their

hers as their control

Ntombakhe Zikode

Bales with Premix 450

Mtholeni Dlamini

Dlezakhe Hlongwane

Mkhathini Dladla

Khulekani Dladla

Phumelele Hlongwane

Ntombakhe Zikode

They have now undertaken to feed 2-3

pregnant or lactating cows with the bales

mixed with Premix 450 and to have the rest

of their hers as their control

Teff Bales

Ntombakhe Zikode

Mtholeni Dlamini

Baling of the Teff they have grown

Veld with protein blocks

Lungile Dladla

Matholozane Gumbi

Zodwa Zikode

Phasazile Sitheb

Here it has been very difficult to allow for a

control as all cattle in the kraal are given

access to the block; it is thus a practice that is

being introduced without conscious

experimentation

Veld with Premix 450

Ntobmi Dlamini

Fikile Hltashwayo

Hlupizile Zondi

Thulile Zikode

Hlanganise Hlongwane

Phumelele Hlongwane

They have now undertaken to feed 2-3

pregnant or lactating cows with the Premix

450 and to have the rest of their hers as their

control

Yields; 2018-2019 planting season

Yields have been measured for around 50% of all participants, also focusing on villages where

yield summaries were not done in the previous season.

Separation of the maize yields per plot of the trials has been an ongoing difficulty, although a

number of participants in Stulwane managed this, due to the diligence of their local facilitator,

Nelisiwe Msele. In addition, some participants combined their trial and control plot yields- in

this case due to very low yields and these results could not be used.

An ongoing issue is the

lack of appropriate

storage options. In this

season a number of 200l

bins were bought by

participants, with

assistance by MDF.

Right: Decobbing and

weighing maize yields

separated per plot in

Stulwane and Far right:

Harvested maize piled up

in a yard for final drying

before storage. (Constance

Hlongwane- Ndunwana)

This season also saw a much higher incidence of cob rots in harvested maize, due to the wet

conditions towards the

end of the season.

Right and Far Right:

Examples of cob rot found

in harvested maize.

These are likely caused by

Fusarium and Diplodia

species and carry the risk

of mycotoxins.

Participants have been

strongly warned against

consuming this maize and

have been advised to

remove the infected cobs

from the rest of their harvest.

The table below provides a comparison between the 2018/19 season and the previous season

2017/18

Table 20: Yield summaries for Maize; 2017/18 and 2018/19 seasons, for a number of Bergville villages

Village

2017/18

Maize

(Trial) t/ha

2018/19

Maize (Trial)

t/ha

2017/18

Maize

(Control)

t/ha

2018/19

Maize

(Control)

t/ha

2017/18

Beans

(t/ha)

2018/19

Beans

(t/ha)

Stulwane

3,0

2,4

4,4

0,84

Ndunwana

3,8

3,7

0,39

0,56

1,46

Ezibomvini

8,1

2,6

7,5

2,2

1,62

Eqeleni

4,4

3,7

1,37

0,6

Thamela

4,5

1,8

0,51

Emabunzini

4,6

5,2

0,94

Emazimbeni

7,9

4,4

1,28

Vimbukhalo

7,9

1,29

Emangweni-

Emaqeleni

5,9

3,8

1,80

Okhombe

3,5

2,3

Magangangozi

2,8

4,9

2,5

3,5

0,5

Nsuka

2,8

Thunzini

3,6

1,02

1,2

Ngoba

4,2

0,5

Mhlwazini

6,6

Emahlathini

5,9

3,6

1,09

Emafafatheni

1,5

0,625

Emadakeni

4,4

0,45

Nsuka

5,6

AVERAGE

5,0 t/ha

3,5t/ha

3,4 t/ha

2,7t/ha

1,22 t/ha

0,56 t/ha

NOTE: Grey blocks indicate villages where yield measurements are still being finalised. Green blocks indicate

newer villages, where yields have been measured for the first time this season

The overall average yield for this season has been compromised due to the weather conditions

and is 41% lower than 2017/18 for maize and 55% lower for beans.

Yields in intercropped and rotated plots

Maize yields for different plots within the experimentation regime were taken for a few of the participants and are summarized in the table below

Table 21: Maize yields measured per plot of the experimental fields for participants from Eqeleni, Stulwane and Ndunwana;

2018/2019

Yield in kg

2018

/ 19

2017

/18

Village

Name

and

Surname

Trial description

PLOT 1

PLOT 2

PLOT3

PLOT 4

PLOT 5

PLOT 6

PLOT 7

PLOT 8

PLOT 9

PLOT

t/ha

Eqeleni

Ntombakhe

Zikode

plt1(m),plt2(sccmix),plt3(

m),plt4(b),plt5(m),plt6(m+

LL),plt7(m+b),plt8(m+b),pl

t9(m+b),plt10(lab lab)

35,6

scc mix

22,8

beans

0,0

106,8

56,2

133,5

101,0

lab lab

7,6

4,9

Stulwane

Dlezakhe

Hlongwane

plt1(m+b),plt2(m+c),plt3(s

ccmix),plt4(m+c),plt5(m),pl

t6(b),plt7(m),plt8(m+b),plt

9(lb lab),plt10(m+c)

34,2

20,5

scc mix

11,9

65,6

beans

68,8

lab lab

12,7

3,7

4,2

Khulekani

Dladla

plt1(sccmix),plt2(m+b),plt3

(m+c),plt4(m+b),plt5(b),plt

6(m),plt7(lablab),plt8(m),p

lt9(b),plt10(m)

scc mix

53,2

54,0

33,1

beans

lab lab

25,2

beans

24,0

3,6

3,5

Fikile

Hlatshwayo

plt1(m+b),plt2(sccmix),plt3

(m+b),plt4(lablab),plt5(m+

c),plt6(b),plt7(m),plt8(b),pl

t9(m0,plt10(c)

39,0

16,1

1,3

Thulani

Dlamini

plt1(sccmix),plt2(m+b),plt3

(m+c),plt4(b),plt5(m+c),plt

6(m),plt7(c),plt8(b),plt9(m

),plt10(lab lab)

scc mix

23,2

34,5

beans

34,3

56,1

cowpeas

beans

31,7

lab lab

4,2

2,3

Dombolo

Dlamini

plt1(m+c),plt2(m+b),plt3(s

ccmix),plt4(m+c),plt5(m),pl

t6(b),plt7(m),plt8(c)plt9m)

,plt10(lab lab)

20,4

60,3

scc mix

0,0

35,8

beans

23,1

cowpeas

49,3

lab lab

3,6

2,7

Nothile

Zondi

plt1(sccmix),plt2(m+b),plt3

(m+c),plt4(b),plt5(m),plt6(l

ablab),plt7(m),plt8(c),plt9(

b),plt10(m)

scc mix

0,0

beans

14,8

lab lab

14,7

cowpeas

beans

29,5

1,7

2,1

Matolozana

Gumbi

plt1(m),plt2(b),plt3(m+b),p

lt4(sccmix),plt5(m+b),plt6(

lablab),plt7(m+b),plt8(c),pl

t9(m),plt10(b)

41,7

beans

24,6

scc mix

0,0

lab lab

0,0

cowpeas

37,0

beans

2,4

4,5

Mthuleni

Dlamini

plt1(m+b),plt2(m+c),plt3(b

),plt4(m),plt5(lablab),plt6(

m),plt7(c),plt8(m),plt9(b),p

lt10(scc mix)

36,8

39,4

beans

81,2

lab lab

77,2

cowpeas

0,0

beans

Scc mix

5,3

0,7

Cuphile

Buthelezi

plt1(m+b),plt2(m),plt3(c),p

lt4(m),plt5(b),plt6(sccmix),

plt7(m),plt8(b),plt9(m+c),p

lt10(lab-lab)

13,8

0,0

cowpeas

68,4

beans

scc mix

67,6

beans

14,3

lab lab

3,6

2,9

Thembi

Mpinga

plt1(m+b),plt2(m+c),plt3(

m+b),plt4(m+c)

31,6

0,0

21,5

15,6

2,3

Ndunw

ana

Boniwe

Hlatshwayo

plt1(m+b),plt2(m+c),plt3(

m+b),plt4(m+c)

42,4

51,4

33,3

16,9

4,8

10,9

Shiyiwe

Mazibuko

plt1(m+c)plt2(m+c),plt3(m

+b),plt4(m+c)

13,9

5,5

20,4

9,4

1,6

2,5

Matozo

Zondo

plt1(m+b),plt2(m+c),plt3(

m+b),plt4(m+c)

63,8

35,8

32,6

29,0

5,4

9,6

Makhu

Mdluli

plt1(m+b),plt2(m+c),plt3(

m+b),plt4(m+c)

99,5

57,6

49,1

64,7

9,0

4,0

Vayiswa

Hlatshwayo

plt1(m+b),plt2(m+c),plt3(

m+b),plt4(m+c)

40,9

21,3

21,1

20,6

4,0

Lethiwe

Zimba

plt1(m+b),plt2(m+c),plt3(

m+c),plt4(m),plt5(sccmix),

plt6(m),plt7(lablab),plt8(m

+c),plt9(m+b,plt10(m+b)

57,3

41,3

36,3

42,3

scc mix

55,5

lab lab

39,5

45,1

0,0

4,9

AVERAGE

4,1

4,2

From the above table it can be seen that overall the average yields in the experimental plots were in fact very similar for 2017/18 and the 2018/19

seasons, being 4,2 and 4,1 t/ha respectively. The other notable point is the continued large variation in yields between the different plots within each

of the participants’ fields. The expected evening out of yields between plots, due to the crop rotation and multi-cropping options, is not really

happening. These yield differences are due in part to historical difference in land management, even at this micro-scale and partly due to different

practices between plots-such as more meticulous weeding in some plots than others and livestock invasion only in parts of fields. It also points

towards the confusion in soil health test results when trying to compare different cropping options in different plots within one field.

Bean Yields for the 2018/19 season

The 2018/2019 season has been a very difficult seasons for bean production; In a number of

villages, no beans were harvested at all. Below an indicative summary is provided for three

villages where bean harvests were recorded.

Village

% of

participants

who obtained a

harvest

Average

harvest (Kg

Yield

(t/ha)

Eqeleni

55%

12,12kg

0,6

Emafefetheni (new)

44%

5kg

0,625

Emadakaneni (new)

63%

3,6kg

0,45

Average

54%

0,56t/ha

Right: A basin of beans being weighed at Emadakaneni during yield

determinations

Farmer Centres

Case study 1: Ezibomvini Farmers Centre

Above left: Phumelele Hlogwnae making a sale of a

bunch of spinach and Above right: A view of thefarmer centre; which includes seed, fertilizer,

pesticides, knapsack prayers, seedlings and 200l storage drums.

The Ezibomvini farmers centre is owned by Phumelele Hlongwane and Zodwa Zikode. It has been

operating for 3 years, in Phumelele’s homestead. The enterprise sells agricultural inputs and non-

agricultural items including fertilizers (MAP and LAN), herbicides (Round up), knapsack

sprayers, preservation pills, maize and bean seed, vegetables, seedlings, protein premix(fodder

supplement) and 200l storage drums. She decided to include the non-farmingitems to sell during

off-peak times. The agricultural inputs are sold to local farmers in the village and farms from

surrounding villages. She obtained the initial investment from Mahlathini Development

Foundation asaloanpayable after three months. Subsequent investments were obtained from

savings group and profits made through sales.

The incremental success of the farmers centre has continued to grow year after year since its

inception. The owners are still as surprised by the success the business has generated over the

years because they were ‘venturing into something completely new and risky’. Local farmers and

neighbours have continued to support the enterprise;fertilizers, seed and herbicides are still the

most sold products by the centre. These sales are made mainly during the planting season. But

over the years sales have also increased during off-peak times through the sale of non-farming

items mentioned above. The inputs sold in small quantities have been the main attraction to the

centre and the ability to purchase on credit on conditions specified by the owners. For example,

customers are only allowed to buy on credit if they do not have existing debt or if there’s enough

stock.

We recommended that she start selling the fertilizer, maize seeds and other seeds, herbicide as a

combo to local farmers. She added that this would be a great idea because some farmers end up

planting late due lack of access to inputs when the rains start. This could ensure that the farmers

have all the inputs they need to start planting immediately and increase her sales throughout the

planting season. She will be working on what the combos will include, price and quantities of each

time.

Case study 2: Thamela Farmers centre

Mam’Constance Hlongwane is the sole owner of the Thamela farmers centre which has been

operating from December 2018. She sells

LAN and MAP fertilizers, eggs and non-

farming items such as candles and soap to

boost sales during throughout the year. She

purchased the inputs late in December

during a dry and hot season, most farmers

did not plant and most of those who planted

had completed.

Right:The initial inputspurchased for the

farmers centre.

She received her initial R1643.06 loan from

Mahlathini Development Foundation. She

bought the following inputs:

Item

Quantity

Price per item

Amount

MAP fertilizer

R447.01

UERA

R350

R700

LAN

R284.05

Round up

R106

R212

Total amount

R1643.06

During our visits we noticed that her confidence had decreased due to the low sales. We

recommended that she may want to think of other items she could include. She later added some

household items for sale. Subsequent investments were made using profits from sales and her

own money to buy additional eggs and non-farming products such as candles and soaps. She had

low confidence in the beginning that her enterprise could continue due to the low agricultural

input sales in December due to the late purchase of the stock. After taking the recommendation

to include other items that could be sold during off-peak times, she was regained her confidence

and continued. Her sales are lower than expected but she has made personal investments to

purchase more products to boost sales.

Case study 3: Ndunwana Farmers centre

The Ndunwana Farmers centre is owned by Lethiwe Zimba and Boniwe Hlatswayo. The

enterprise operates fromBoniwe Hlatshwayo’s

homestead. The owners areneighbours so

customers can go to Mam’ Zimba if Mam’

Hlatshwayo is away. They received an initial loan

of R2758 from Mahlathini Development

Foundation to start the business. They both share

the responsibility of paying back the loan.

Subsequent investments have been made by

Mam’Constance to purchase eggs and paraffin to

boost sales during off-peak times.

Right: The owners of the farmers centre recording

delivery of inputs

The following items were purchased using the loan:

Item

Quantity

Price per item

Amount

MAP fertilizer

R447.01

R894.02

UERA

R350

R1050

LAN

R284.05

Round up

R176.64

R529.93

Total amount

R2758

It was recommended that they sell the eggs in smaller quantities, than the packs of 30 they were

selling; given that this market is mostly available at month-end only and eggs would spoil if kept

on the shelf for long periods.

Case Study 5: Stulwane Farmers centre

Nelisiwe Msele is the owner of the Stulwane Farmers Centre for about 2 months since 15 May

2019. Following the

workshop on fodder

production for the winter

season, the facilitator

suggested to the farmers to

include protein

supplements for livestock to

sell to livestock farmers and

preservation pill. She

bought additional candles

and salt sold at the farmers

centre.

Right: Stulwane farmers

centre

The initial investment was used to purchase the following items:

Item

Quantity

Price per item

Amount

Voermol Premix 450

R230

R920

Protein Block

R149.50

R299

Quick phos tab (pack)

R302

Chemical measuring cup

R7.50

R30

TOTAL

R1551

She has made subsequent investments due to the increasing demand for Voermol premix 450.

She initially thought people would not buy from her but we encouraged her to market her items

more by spreading the word to her neighbours at all gatherings.

Case study 6: Emabunzini Farmers centre

Mam’Valindaba Khumalo also has a farmer centre in her homestead. The enterprise began

operating from December 2018. She has knowledge of livestock vaccination due to her

membership in the Executive Committee of the local dipping committee. In winter she plans to

include livestock vaccines to her stock to sell to livestock farmers. She is the only one from the

three other farmers centre opened in 2018 that was ready to return the loan to Mahlathini

Development Foundation.

Shevisited her neighbours going from door to door marketing her products which was successful.

Right: Mrs

Valindaba

Khumalo’s

farmer centre

and Far right:

Phumzile

sitting with her

to consolidate

her record

keeping

process.

The following inputs were purchased with the initial loam:

Item

Quantity

Price per item

Amount

MAP fertilizer

R447.01

UERA

R350

LAN

R284.05

Round up

R106

R318

Total amount

R1399.06

Personal subsequent investments were used to purchase candles, salt and soap which were sold

to increase profit and pay back the loan as soon as possible. Mam’ Valindaba can’t write so we

devised a system with her that she could use to record. We offered her the opportunity to continue

to use the initial investment to purchase livestockvaccines and other livestock items she had

suggested.

Below is a table outlining the incomes for each farmer centre

Village

Initial

loan

Items sold

Total

Profit

Stulwane

R1821

Premix 450, protein blocks, Quick phos

tablets. Salt, britelit soap, Bulala Zonke

R6 237,50

R1 247,50

Eqeleni

R1 553

Protein block, premix 450

R1 523,75

R304,75

Thamela

R2 800

Urea, MAP, LAN, Roundup, paraffin,

soap, snuff, eggs

R3 839,20

R778,64

Ezibomvini

Seed PAN 6671, Pan 413, PAN 53,

Quickphos, Fert 2:3:2 (22), MAP Fert

(33), Bulala Zonke, Blue death,

seedlings, premix 450, protein blocks

R21 490,92

R4 958,40

Village Saving and LoanAssociations

In the course of the month of April, the MDF team comprising of Mr Madondo, Tema and

Nontokozo undertook a task to visit VSLA groups in Bergville. This was part of yearly group

reviews with the aim to assess their progress and address any challenges. The groups visited by

the teamare situated in the villages of Vimbukhalo, Ezibomvini, Eqeleni, Ndunwana, Emabunzini,

Stulwane and Bethany (Gudlintaba). MDF is currently supporting 15 groups in Bergville, 2 in

Creighton and 1 in Nokweja area bringing the total number of groups to 18.

The groups are making good progress and all of them are fully operational. The oldest group is

uMntwana Savings Group based in Stulwane Village which is saving for the seventh year this year.

The majority of the groups have been operational for a period between three and five years.

Majority of the groups have a membership ranging from 20 to 30 members with 98 percent of the

members being female. The groups consist of youth members, middle aged women as well as

elderly women, most of whom are unemployed and rely on socialand pension grants as well as

remittances from family members residing outside Bergville.

Savings meetings are conducted monthly until the conclusion of the 13-month cycle wherethe

group dissolves and starts again. Members who are not present at meetings canrequest someone

to save on their behalf but must be present when borrowing to sign off their loans.

All of the groups have had their share-outs for the 2018/19 cycle. During the reviews, the groups

were asked to talk about how the money from savings has made a difference in their lives and

whether they save for agricultural inputs. The biggest items on which share-out money is utilized

include furniture; fridges, wardrobes and televisions in particular which were mentioned by 95%

of the members. Other uses include payment of school fees, groceries and household renovations

which were also mentioned in more than 80%of the members of the savings groups. A small

percentage (5-10 %) used the money to purchase agricultural inputs, i.e. fertiliser, seed and other

agriculture related products such as meat and eggs.

Reasons for the aforementioned are that in most savings groups, up to half of the group members

are not part of the CA programme,but joined the groups as they wanted to learn about savings

hencetheir priorities lean moretowards household consumption. However, even the participants

who practice CA spend the majority of their savings on household needs.

The groups visited distributed a total of R 879 963.00 amongst 258 members (see Table 22). The

majority of the groups had a share increase value ranging between 25 and 49 percent.

Right: A shar out session

being conducted in Ngoba

(Sakokuhle group)

The savings groups have

been supported as an

important element in the

stability of the groups and

in support of livelihoods

in these rural

communities.

There is no direct evidence that the hoped-for increase in savings for inputs and agricultural

uses is materialising. Work with the VSLAs will thus no longer be consciously included as a part

of this programme.

Table 22: Group Share-out Bergville, 2018/19

Village

GROUP NAME

NO. OF

MEMB

ERS

NEW

VALUE

TOTAL

AMOUNT

SHARED

OUT

MAX

AMOUNT

/MEMBE

MIN

AMOUNT/

MEMBER

AVERAGE

AMOUNT/M

EMBER

USES

Vimbukhalo

Ukhamba

R130.00

R75,000

R8,000

R0.00

R1,300

Inputs, school fees, cutlery, blankets,

renovations, furniture

Eqeleni

Masithuthuke

R130.00

R80,000

R7,800

R1,560

R3,680

Christmas and school clothes, new TV,

tiles, renovations

Eqeleni

Masibambane

R136.00

R79,698

R8,000

R700

R4,000

Fertiliser, LAN, Maize seed, lounge suite,

floor tiles

Stulwane

uMntwana

R130.00

R140,000

R7,500

R1,200

R3,200

Inputs, furniture, other household needs

Stulwane

Mbalenhle

R149.00

R108,000

R9,000

R1,600

R5,000

Electricity installation, furniture, serviced

debts, business stock, groceries

Ngoba

Sakhokuhle

R145.00

R105,000

R8,000

R3,500

smart phones, clothing, fertiliser, seed,

wardrobe, building material, fencing

Ngoba

Isibonelo

R152.00

R100,000

R9,000

R1,200

R4,000

Wedding celebration, furniture, livestock,

groceries, investments

Bethany

Gudlintaba

R161.00

R86,070

R9,600

R1,400

R4,500

New stock for meat business (tripe), eggs

to sell, poultry, medical bills, College fees

Vimbukhalo

Inyonyana

R130.00

R41,210

R3,770

R780

R1,950

Furniture, groceries, school fees

Ezibomvini

Ukuzama

R125.00

R23,375

R3,375

R2,000

Inputs, household needs

Ndunwana

Mphelandaa

R149.00

R41,610

R3,576

R700

R2,200

Christmas and school clothes, renovations

TOTAL

258

R879,963

Stakeholder engagement

Maize Trust Board visit

6 Members of the Maize Trust Board, journeyed from Pretoria to the cathedral Peak area in

Bergville for a smallholder conservation agriculture day hosted by the MDF team. The intention

was to provide informationand practical examples of the innovation development approach used

for adaptive and participatory research into smallholder CA systems. Both this approach to

research and the emphasis on livelihoods and adaptation were new to these important decision

makers in the maize industry. The day was designed also to showcase some of the work

smallholders have undertaken.

Below are a few illustrative photographs of the farmer visits.

Above Clockwise from top left: Visiting Ntombakhe Zikode’s field in Eqeleni where a plot of winter

cover crops is seen in the fore ground; Her maize crop maturing; the farmers’ meeting with the

board members and a view of a portion of the farmer centre for the village.

Issues, comments and suggestions

1. This season saw a dip in the number of CA trial participants, inaddition to which those

who did plant their subsidised trials for the most part did not plant the rest of their fields.

This is a reasonably strong indication that these really poor smallholders use the

subsidised CA production process as a risk management strategy for their cropping. This

was not the intention, but the feeling is that the gains in food availability and food security,

as well as general soil health are high enough to justify this “investment”.

2. These seasons of high climate variability affect the ability of these smallholder farmers to

crop in a commercially viable way, but the CA process has assisted to build resilience in

the overall cropping system and improve the resilience of the livelihoods of these people.

3. Yields have decreased by around 40% in this season, primarily due to late seasonal rains

and coupled with very high evapotranspiration rates in the hot summer months, despite

overall rainfall being higher than last season.

4. The partnership with the KZNDARD Landcare unit is both rewarding and very frustrating.

This season late and haphazard delivery of inputs from the Department led to a lot of time

wastage and planting even later than needed. Thankfully the Department wasvery aware

of this shortcoming and has initiated deliveries of inputsfor the upcoming season already.

5. The local facilitators have played a much stronger role this season in organising

monitoring and managing their learning groups. Their role will become more crucial in

future as this responsibility will be handed to the groups themselves to free the MDF field

staff up to focus on the research aspects of the programme

6. A focus on integration of livestock has been initiated and the initial fodder production and

supplementation experimentation has been very interesting for all involved. This aspect

is to be strengthened in the upcoming season through a partnership with the soil science

division of KZNDARD at Cedara and AGT Food for supply of a range of different cover crop

and fodder seed varieties.

7. A partnership also with a research process on mycotoxins through the ARC has been set

in place for the upcoming season, as the slowly increasing prevalence of cob rots is of

concern.

8. There has been a high turnover of interns at MDF and although this is to be expected, it is

difficult to ensure quality of work and the monitoring these interns have undertaken.

Although MDF is committed to this capacity building process, the resultant compromise

in quality of data for some of the monitoring processes is seen as regrettable. The process

of instituting an initial probation period of 1-3months, is now being held to more firmly.

Summary of annual expenses as on August 2019

Date of transaction

Type of transaction

Amount

2018/10/18

AGT FOODS : Seed

16 175,90

2018/10/18

AGT FOODS : Seed

1 932,00

2018/10/26

Monthly expenses

64 169,40

2018/12/11

Monthly expenses

52 358,17

2019/01/23

Monthly expenses

117 993,96

2019/02/28

Monthly expenses

75 295,69

2019/03/29

Monthly expenses

45 358,22

2019/04/30

Monthly expenses

55 445,11

2019/05/31

Monthly expenses

45 780,95

2019/06/28

Monthly expenses

65 330,30

2019/06/30

Monthly expenses

47 979,38

2019/07/31

Monthly expenses

48 490,22

2019/09/09

Monthly expenses

50 545,50

TOTAL AUG 2019

636309,30