March 2022
Appendix : Quantitative research support
to the Conservation Agriculture
Smallholder Farmer Innovation
Programme, KwaZulu-Natal.
Progress report
2
PROGRESS REPORT
Quantitative research support to the
Conservation Agriculture Smallholder
Farmer Innovation Programme, KwaZulu-
Natal
For period:
October 2021 to February 2022
Submitted to The Maize Trust
Compiled by:
E Kruger, M Dlamini, T Mathebula, M Malinga, N
Madondo, L Buthelezi, N Mzobe and HJ Smith
Asset Research
In collaboration with:
Mahlathini Development Foundation
3
Contents
Contents ..................................................................................................................................................................................3
1.Background and project description................................................................................................................4
2.Approach and Methodology................................................................................................................................. 5
3.Key activities: October 2021-February 2022...............................................................................................5
3.1Budget ..................................................................................................................................................................6
3.2Progress...............................................................................................................................................................7
4.Results achieved to date ........................................................................................................................................8
4.1CA collaboratively managed trials (CMTs)..........................................................................................9
4.2Rainfall and runoff .......................................................................................................................................10
4.2Soil health and Nematode indices.........................................................................................................12
4.2.1Samples and sampling procedures .............................................................................................12
4.2.2Biomonitoring using nematode indices....................................................................................14
4.2.3Nematode diagnostics results. ......................................................................................................17
5.Progress summary ................................................................................................................................................25
5.1Bergville ...........................................................................................................................................................25
5.1.1Ezibomvini.............................................................................................................................................27
5.1.2Eqeleni.....................................................................................................................................................30
5.1.3Stulwane .................................................................................................................................................32
5.1.4Vimbukhalo ...........................................................................................................................................34
5.1.5Bergville Fodder trials .....................................................................................................................36
5.2SKZN...................................................................................................................................................................38
5.2.1Springvalley ..........................................................................................................................................38
5.2.2Ngongonini ............................................................................................................................................39
5.2.3Madzikane .............................................................................................................................................41
5.3Midlands ...........................................................................................................................................................42
5.3.1Ozwathini ...............................................................................................................................................42
5.3.2Mayizekanye .........................................................................................................................................45
5.3.3Gobizembe .............................................................................................................................................46
6.Issues and recommendations ...........................................................................................................................46
PHOTO CREDITS: Cover Page: Left and Right: Maize mono-cropped CA plot and SummerCover
Crop plot for Nombono Dladla’s CMT in Ezibomvini, Bergville.
4
1.Background and project description
Mahlathini Development Foundation (2003-2022) is one of the few NGOs in South Africa
focussing on promotingcollaborative pro-poor agricultural innovation. As such MDFis a
specialist NGO working in the fields of participatory research, training and implementation,
focussing on agroecological approaches, including conservation agriculture (CA).
The Maize Trust-funded CA Smallholder Farmer Innovation Programme (SFIP) in South Africa, as
conceptualized and implemented throughMahlathini Development Foundation (MDF), has
pioneered the use of agricultural innovation systems as a methodological approachfor the
promotion of anappropriate smallholder CA farming system, as well as awareness raising and
adaptive research into specific elements of thissystem(Kruger & Smith, 2019). This approach
takes cognizance of the complexity of introducing CA into afarming system, including working
with smallholder farmers as partners in the knowledge co-creation processthrough on-farm
research and experiential learning, as well as embedding the process into the existing socio-
political environments and economic value chains. The overall goal of the CA SFIP is the research,
development and adaptation ofCA by grain farmers to ensure sustainable use and management
of natural resources while enhancing national and household food security and income.
MDF has worked with smallholder farmers in CA learning groups, who implement CA as farmer
level trials and expand their implementation into their whole fields over time. Around 570
smallholders across 40 villages in KwaZulu-Natal (Bergville, SKZN and Midlands) have been
implementingCA for a period of between one to eight years, because of The MaizeTrust SFIP
support. Since 2020, the emphasis has shifted more towards quantitative research andless on
promotion and awareness raising. A selection of the participants now participates in a process of
Mother and Baby trials, with theformer also being known as collaboratively managed trials
(CMTs). In CMTs farmers and researchers work together on problem definition, design,
management and implementation of statistically sound trials aswell as evaluation. TheseCMTs
are considered the ‘Mother”
1
trials and 56 CMTs have been implemented with participants(40)
across the Bergville, SKZN and Midlands sites. Theyare managed alongside the farmer level
experimentation or “baby” trials, of which there have been 90 in this season.
Experimentation protocols chosenby the CMT participants (researchers and farmers) in
participatory review and planning sessions are as follows:
1.10x10’s: This denotes a10-plot layout of 100m2plots, for multiple cropping options(Such
as maize, legumes and cover crops), which is rotated on an annual basis.
2.Strip cropping(1000m2):Planting is done in 4m wide strips on contour, to provide for
soil and water conservation concerns and ease of implementation in larger fields.
3.Short season maize: Planting of two varies of early maturing maize (white-PAN5A172
and yellow-PAN5A190) alongside the normal varieties to test adaptability to climate
variation.
4.Fodder production: Planting ofextra CA trial areas for annual and perennial livestock
fodder species (Teff, Tall fescue, Lespedeza), for both in situ grazing and baling.
1
Sieglinde S., Snapp S., DeDecker, J. and Davis, A. (2019). Farmer Participatory Research Advances
Sustainable Agriculture: Lessons from Michigan and Malawi. Agronomy Journal. Volume 111, Issue 6. Pgs.
2681-2691.
5
2.Approach and Methodology
The CMTsfor the 40participants include different CA and conventional control plots. For the
latter,farmers use their own fertilizer, seed andspacing combinations. Manyhave now used the
2 row planters for their control plots.
Most of the farmer participants have been implementing CAfor a number of years and they will
be provided limited support and mentoring to continue. The intention is to gather yield data also
from the larger group (Baby trials), but to focus on the smaller group (CMTs) for the quantitative
data. This is to include the following:
1.YIELD DATA:
Maize, yields under CA vs conventional cropping systems (80 participants)
Maize yields in CA rotation and intercropping systems (24 participants)
Biomass/dry matter yields for maize for water productivity assessments (9 participants)
Bean and cowpea yield in CA vs conventional tillage systems (50 participants)
Grain yields of cover crops planted (24 participants) and
Biomass /dry matter yields for livestock fodder (9 participants).
2. WATER RELATED DATA:
Local rainfall measurements in each village (x 8 villages)
Weather station and SAEON seasonal data; ET0, temperature, rainfall per site x 3 sites
Run-off (run-off pans installedin CA and conventional control plots and different CAcropping
options e.g. monocropping, intercropping, cover crops); 3-4 run-off pans per participant (9
participants)
Bulk density assessments (9 participants)
Water productivity assessments; 3-4 treatments per participant (9 participants)
3. SOIL RELATED DATA
Soil health analyses (Haney Soil Health Test (SHT), and nematode diagnostic indices) (24
participants, x treatments each, including veld, control and one to two CA samples)
3.Key activities: October 2021-February 2022
CA farmer level experimentation plots have been implemented for 130participants across 12
villages in KZN (Bergville, ZKN and Midlands). The CMT plots (or Mother trials) have been set up
with 9 participants in SKZN, 13 in Midlands and 18in the Bergville study area.
Participants have the following CA treatments: strip cropping, cover crops, different maize
varieties, use of 2 row tractor drawn planters,production of cover crop seed, and annual and
perennial livestock fodder species. Crop growth monitoring has been conducted for all CMTs in
Bergville and SKZN. For the Midlands, given their lateplanting dates (January to February 2022),
crop growth monitoring will be undertaken in April 2022.
Rain gauges and run-off plots have been setup for 11participants across as many villagesand
participant farmers have undertaken to keep ongoing records. The South African Environmental
Observation Network (SAEON) has assisted with weather station data forcomparison with the
farmer level records.
6
Soil health sampleshave been taken for 21 participants across 14villages and have been
submitted to the Soil Health Support Centre (inthe Western Cape) for Haney SHT analyses, as
well as nematode indices (through Northwest University, Potchefstroom) for analyses of soil
health results.
Stakeholder engagement this season has consisted of the following activities:
Liaison with Landbouweekbladto provide for small videos as features for smallholder
success stories inconservation agriculture and for seeking sponsors to assist inhosting
open days
Close cooperation with both the Okhahlamba Local Municipality and the Bergville office
of KZNDARD in use of their tractors and sprayers for planting of CA trials, as well as
potential provision of funding for fencing 100 fields in the Bergville area
Continued support from the LandCare programmeof DALRFfor provision of CA
implements
Showcasing of thesmallholder innovation programme across a number of networks and
stakeholder platforms through meetings, discussions and webinars, including the
Agroecology Network andthe Adaptation Network (focussing on Nature based solutions
and agroecology), and the SANBI catchment-based Indaba,
Collaboration inthe Farming for Climate Justice short action research programme for
committed early careerresearchers in “Transforming Policy and Practice Towards
Agroecological Transitions forResilient and Just Food and Farming Systems in a Changing
Climate”, in association with UCT and Coventry University in the UK, and
Acceptance ofa paper for the South AfricanMountains Conference in March 2022 entitled
“CbCCA in central Drakensberg improves resilience in smallholder farmers”.
3.1Budget
The budgetset aside for the 1st sixmonths (Oct 2021 to March 2022), according to the overall
work planisR605 838. Expenditure for the last five months has been R522845. Expenditure is
well within the present budget range.
Table 1: Budget summary for expenditure between October 2021-March 2022.
Outcomes/
deliverables
Budgets
Oct
Nov
Dec
Jan
Feb
Travel,
accommodation,
admin,
accounting
R12 120
R 1 821
R5 500
R4 750
R1 400
Farmer
experimentation,
researcher
managed
experimentation
R508 358
R 98 982
R82 995
R55 398
R42 685
R32 151
Inputs
R39 570
Laboratory
costs, soil health
samples
R345 600
R112 800
R44 792
TOTAL BUDGET
R866 078
ACTUAL EXPENDITURE
R 522 845
7
3.2Progress
The table below outlines activities related to objectives and key indicators for the period of
October 2021 to February 2022.
Table 2:SUMMARY OFPROGRESS(OCTOBER2021-FEBRUARY 2022)RELATEDTO OBJECTIVES AND KEYACTIVITIES
Objectives
Key activities
Summary of progress
%
completion
Objective 1: To
assess the impact
of a range of CA
practices on water,
soil and
productivity
indicators, within
a smallholder
farmer level
experimentation
process
Key activity 1. Participatory planning,
design and layout of experiments
Around 160 participants across 20
villages plant CA trials, with intensive
research managed treatments and
plots for a minimum of 36 of these
participants (across three sites)
Key activity 2.Collection and analysis
of results
Researcher and farmer managed
quantitative and qualitative outcomes
are measured for a minimum of 3
participants across 6 villages
(including run-off plots, soil health
analysis, bulk density and water
productivity calculations, and yield
measurements (maize, cover crops
and fodder).
130 participants across10
villages have planted CA
trials. 40 Collaboratively
managed trials (CMTs) have
been set up
-Soil health samples and
analysis for 22 participants
across 10 villages
-Rain gauges and run-off
plots for 11 participants
across 9 villages
-Crop growth monitoring for
40 participants across 10
villages
100%
50%
Objective 2: To use
results from
qualitative and
quantitative
results to outline
best bet options
for the
smallholder CA
farming system
and to provide
recommendations
for improvement
of the system
Key activity 1: Write up of results
-Interim and annual reports
-Case studies (1 to 2 per site x 3 sites)
-Popular magazine and peer reviewed
articles (minimum of 1 each)
-Posting on appropriate websites and
online newsletters and forums
Key activity 2: Presentation of results
-Results presented to smallholder
farmers in learning and review
sessions, for analysis, interpretation,
and planning purposes (minimum 3,
maximum 6 sessions)
-Results presented to broader farmer
stakeholder forums for networking
and awareness raising (1 to 3 events)
-Interim report submitted in
March 2022
-Short cases presented in this
interim report.
-Conference papers SA
Mountain Conference, Asset
research symposium
-Publication of a chapter
entitled “Conservation
Agriculture Innovation
Systems Build Climate
Resilience for Smallholder
Farmers in South Africa” in
an upcoming CABI
-www.mahlathini.org
-Not yet done
-Bergville Smallholder
Conservation Agriculture
open day 2nd March 2022.
50%
50%
100%
50%
25%
75%
A performance dashboard is indicated below.This provides a snapshot of performance according
to suggested numbers and outputs in the proposal.
8
Table 3:PERFORMANCEDASHBOARD;FEBRUARY2022
Outputs
Proposed (March 2021)
Actual (Feb 2022)
Number of areas of operation
3
3
No farmer level experiments
160
130
No of CMTs
36
40
No of local facilitators
6
VSLAs
18
Participatory monitoring and evaluation
process (farmer level)
60
40
Note: Fewer CA trials have been set up this season, necessitated by the need for participants to
procure their own inputs, as subsidies could not be provided. Funding for this item could not be
secured. In addition, availability of inputs after the social unrest in August 2021 has been very
patchy and new inputs supply coming on stream inJanuary 2022 showed a marked increase in
price. A number of thepoorer participants inthe CA learning groups could not affordto undertake
cropping this season.
4.Results achieved to date
The table below outlines the CMTs planted and monitored, numbersof participants and
experimentation processes for theprogramme.
Table 4:CMTSAND BABY TRIALS,PER VILLAGEFOROCTOBER2021-FEBRUARY2022
CMTS per area
Spraying
dates
Planting
dates
Baby
trials
Area
Village
Name
Surname
10x 10's
Strips
Fodder
Bergville
Ezibomvini
Phumelele
Hlongwane
1
1
21/10/26
21/11/15
52
18
Mantombi
Mabizela
1
1
21/10/26
21/11/26
5 males
Zodwa
Zikode
1
1
21/10/26
21/11/23
13 females
Nombono
Dladla
1
1
21/10/26
21/11/17
Eqeleni
Thulani
Dlamini
1
1
21/11/04
21/11/22
Sthabiso
Manyathi
1
21/11/04
21/11/16
Nomavila
Ndaba
1
21/11/04
21/11/17
Ntombakhe
Zikode
1
1
21/11/04
21/11/19
Thulile
Zikode
1
1
21/11/04
21/11/25
Stulwane
Nothile
Zondi
1
1
21/10/29
21/11/24
Khulekani
Dladla
1
1
21/10/29
21/11/25
Thulani
Dlamini
1
1
1
21/10/29
21/11/25
Dombi
Buthelezi
1
1
21/10/29
21/11/23
Nelisiwe
Msele
1
21/10/29
21/11/24
Vimbukhalo
Sibongile
Mpulo
1
1
21/11/03
21/11/24
Zibonile
Sithole
1
21/11/03
21/11/30
Zweni
Ndaba
1
21/11/03
21/12/01
Bukhisiwe
Ndaba
1
1
21/11/03
21/11/26
SKZN
Spring Valley
Mboniseni
Dlamini
1
21/10/26
21/12/01
17
9
Letta
Ngubo
1
21/10/26
21/11/17
7 males
Bonginhlanhla
Dlamini
1
21/10/29
21/11/26
9
3 females
Gertrude
Khwela
1
21/10/29
21/11/25
Ngongonini
Mandla
Mkhize
1
21/11/04
21/11/15
Leonard
Gamede
1
21/11/04
21/11/23
Moses
Zulu
1
21/11/04
21/12/01
Madzikane
Cosmos
Xaba
1
21/11/03
21/12/02
Nombuyise
Shozi
1
21/11/03
21/12/02
Midlands
Mayizekanye
Babekile
Nene
1
22/01/12
22/01/19
21
14
Ntombi
Shandu
1
1
22/01/12
22/01/17
3 males
Dumazile
Nxusa
1
22/01/12
22/01/17
11 females
Fikelephi
Mapumulo
1
22/01/12
22/01/18
Mavis
Shezi
1
22/01/17
22/01/25
Gobizembe
Rita
ngobese
1
21/11/10
21/11/24
Ozwathini
Martina
Xulu
1
22/01/14
22/01/24
Nora
Sibiya
1
21/12/10
21/12/17
Aaron
Nkomo
1
22/01/10
22/01/17
Ndabenkhulu
Myeza
1
22/01/06
22/01/13
Lindiwe
Khanyile
1
22/01/11
22/01/21
Nomcebo
Zondi
1
22/01/11
22/01/18
Philani
Ngcobo
1
22/01/04
22/01/06
Nokuthula
Dube
1
21/11/24
21/11/30
TOTAL
40
34
11
10
90
Note: SSM = Short Season Maize was planted either in strips, in the fodder trials or separately.
4.1CA collaboratively managed trials (CMTs)
This season, we ensured that all CMTs followed the same plot layout, across all three regions. This
will facilitate a much easier and statistically reliable comparison between the results from
different farmers’ trials. Three different CMTS namely 10x10’s, strips and fodder have been
implemented. The layout for these trials is shown below.
M-Maize (PAN53 or PAN6479 or SC701)
SSM- Short season maize (PAN5A190 or PAN5A172)
B-Beans (PAN9292 or Gadra)
SCC- Summer cover crops (Sun hemp and fodder sorghum and sunflower)
CP- Cowpea (Black eye Betshuana white)
Pk- Pumpkin (Queensland blue or Flat white boer)
WCC (Fodder rye and fodder radish and Saia oats)
BLOCKS (10x10m) x 10
plots
1 M
2 M+B
3 SCC
4 M
5 M+B
10 M+CP/Pk
SCC
8 M+B
7 M
6 SCC
Strips(2mx50m) x10
strips
1 M
2 M+B
3 SCC
4 M
5 M+B
6 SCC
7 M
8 M+B
10
9 SCC
10 M+CP/Pk
Fodder Strips (2mx50m)
x 8 strips
Short season maize
PAN5A190=yellow
PAN5A172=white
SSM
B/WCC relay
SSM
Lespedeza
SSM
Tall Fescue
SSM
B/WCC relay
4.2Rainfall and runoff
This season runoff microplot pans have been installedfor 10 participants in the CMTs. The pans
need to be installed after ploughing for the control plots, but prior to planting for both the control
and CA plots. This provided some logistical difficulties and some of the pans could thenonly be
installed in December 2021 and January 2022.
The table below outlines the rain gauges and run-off pans installed.
Table 5: Rain gauges and runoff pans installed for 10 participants across Bergville, SKZN and Midlands for 2021-2022
Village
Name and Surname
Rain
gauge
Runo
ff
Plot
Ezibomvini
Phumelele
Hlongwane
Yes
Yes
Control, Plot 2(M+B), Plot 4 (M), Plot 7 (M),Plot 9 (SCC)
Stulwane
Nelisiwe Msele
Yes
Yes
Control, Plot 2(M+B), Plot 3 (SCC), Plot 5 (M+B), Plot 7 (M)
Vimbukhalo
Sibongile Mpulo
Yes
Yes
Control, Plot 2(M+B), Plot 3 (SCC), Plot 5 (M+B), Plot 7 (M)
Eqeleni
Thulile Zikode
Yes
Yes
Control, Plot 2(M+B), Plot 3 (SCC), Plot 5 (M+B), Plot 7 (M)
Ngongonini
Mandla Mkhize
Yes
Yes
Control, Plot 2(M+B), Plot 5 (M+B), Plot 7 (M)
Springvalley
Letha Ngubo
Yes
Yes
Control, Plot 2(M+B), Plot 5 (M+B), Plot 7 (M)
Madzikane
Cosmos Xaba
Yes
Yes
Control, Plot 2(M+B), Plot 5 (M+B), Plot 7 (M)
Madzikane
Vakashile Gambu
Yes
Yes
Control, Plot 2(M+B), Plot 5 (M+B), Plot 7 (M)
Ozwathini
Martina Xulu
Yes
Yes
Control, Plot 2(M+B), Plot 5 (M+B), Plot 7 (M)
Gobizembe
Rita Ngobese
Yes
Yes
Control, Plot 2(M+B), Plot 5 (M+B), Plot 7 (M)
Mayizekanye
Dumazile Nxusa
Yes
Yes
Control, Plot 2(M+B), Plot 5 (M+B), Plot 7 (M)
Participant farmers have been provided with monitoring sheets to recordrainfall events and run-
off for their CA trails and control plots. From previous experience it was thought to be an
acceptable strategy to obtain data. In addition, theparticipantswho showed a lack of interestin
previous seasons were replaced with other participants and a few new farmers were brought on
board.
Weather station data (Bergville)compared for the 2019/20 to2021/22growing season are
compared in the small table below.
11
Month
Rainfall
2019/20
(mm);
Didima
Rainfall
2020/21 (mm);
-3 stations
Rainfall
2020/21 (mm);
rain gauges
Rainfall
2021/22 (mm);
-3 stations
Rainfall
2021/22 (mm);
rain gauges
Oct
131,0
122,7
119,8
88,1
Nov
172,6
217,0
107,3
96,2
Dec
143,5
229,2
265,0
229,4
185,9
Jan
99,1
415,1
323,7
318,9
181,5
Feb
86,1
210,6
340,8
March
49,2
131,0
178,9
April
17,7
55,4
61,0
Totals
699,2
1 381,0
1 276,7
732,6
367,4
Given thehighrainfall variability in the region, due to convective storm events, weather station
data has now been averaged for three SAEON weather stations in the areas of Cathedral peak
office, Mike’s Pass and Vulture’s Rest. These stationsare not representative of the area, but a
combination provides a reasonable average.
The local flooding, increased run off and water logging in the fields is not borne out in the rainfall
periodicity, intensity or amount for the December 2021-January 2022period, as overall rainfall
and daily rainfallevents for thesame period in 2020-2021 was much higher, as shown in the
Figure below.Winter rainfall for 2021(May-September), averaged 162,3mm, compared to 34,6
mm in the winter of 2020. It is assumed that the increased surface flow of water thisseason is
due to saturation of the soil profile as a combined result of a wetter winter and high rainfall in the
2020/21season.
Figure 1: Daily rainfall for 2 weather stations (Cathedral Peak and Mike’s Pass) Oct 2020-Jan 2022
Averages have beencalculated for monthlyrainfall and runoff for each areaas recorded by the
participant farmers. The table below provides asample of records for 3 farmers from Bergville.
12
Bergville (Sibongile Mpulo - Vimbukhalo)
Month
Ave monthly rainfall
(mm)
Ave monthly runoff CA
Control plot (L)
Average monthly runoff
CA Trial plots (L)
Dec-20
143,8
20,85
8,8
Jan-21
183
1,79
6,8
Feb-21
14
1
2,2
Sum
340,8
23,64
17,8
% Rainfall conversion
6,9%
5,2%
Bergville (Thulile Zikode-Eqeleni)
Month
Ave monthly rainfall
(mm)
Ave monthly runoff
Control plot (L)
Average monthly runoff
Trial plots (L)
Dec-20
162
3,15
2,58
Jan-21
179
4,2
3,34
Sum
341
7,35
5,92
% Rainfall conversion
2,2%
1,8%
Bergville (Nelisiwe Msele-Stulwane)
Month
Ave monthly rainfall
(mm)
Ave monthly runoff
Control plot (L)
Average monthly runoff
Trial plots (L)
Nov-20
Dec-20
252
13,01
13,67
Jan-21
182,5
18,85
18,74
Feb-21
Sum
434,5
31,86
32,41
% Rainfall conversion
7,3%
7,4%
Runoffand rainfall results have notbeen well documented this season. Neither the rain gauges,
nor the run-off pan buckets (10 litre) were big enough to accurately measure rainfall, as they
filled up and over-flowed completely for anumber of rainfall events in December and January.
The farmers keeping records and the young field workers supporting them, did not know how to
manage the situation. As aresult, records were notkeptby some farmers and for others thevalues
are much lower than they would have been in reality.
From asample of the results kept and analysed, shown in the table above, thedifference in
percentage runoff betweenthe CA control and trial plots is quite small this season at less than
1%, compared to an average of 5% in the previous seasons.
In future, theserecords will need to be kept by a field staff member/junior researcher. This will
unfortunately increase costs considerably given the increased time and logistics requirements of
such an arrangement.
4.2Soil health and Nematode indices
4.2.1Samples and sampling procedures
Ninety-five (95) soil samples were taken for 22 CMT participants across 10 villages in thethree
areas (Bergville, SKZN and Midlands). Samples were taken in September 2021 prior to the
planting season, to remain coherent with sampling procedures to date. These samples were used
for both the Haney Soil Health tests and for Nematode analysis.
13
Treatments have been adjusted this season to reflectplot number in the CAtrials, rather than
crops planted, as the results for different crops planted within the trialshave notbeen statistically
significant to date. There are localised variations in the soil patterns within each farmers CA trial
and thus sampling has been done across 4 of 10 of the plots for each farmer, to average this result,
as well as for the control plots.
Table 6: Soil samples number, farmers and plot numbers takenfor SH and Nematode analysis: September 2021
Number of samplesSample CodeArea
1P. Hlongwane Conventional Control
Bergville
2P. Hlongwane Plot 2
Bergville
3P. Hlongwane Strip 2
Bergville
4P. Hlongwane Plot 4
Bergville
5P. Hlongwane Strip 4
Bergville
6P. Hlongwane Plot 6
Bergville
7P. Hlongwane Plot 9
Bergville
8P. Hlongwane Veld
Bergville
9Z. Zikode CA Control
Bergville
10 Z. Zikode Plot 2
Bergville
11 Z. Zikone Plot 4
Bergville
12 Z. Zikode Plot 7
Bergville
13 Z. Zikode Veld
Bergville
14 N. Dladla CA Control
Bergville
15 N. Dladla Plot 2
Bergville
16 N. Dladla Plot 4
Bergville
17 N. Dladla Plot 7
Bergville
18 N. Dladla Veld
Bergville
19 D. Hlongwane CA Control
Bergville
20 D. Hlongwane Plot 2
Bergville
21 D. Hlongwane Plot 4
Bergville
22 D. Hlongwane Plot 7
Bergville
23 D. Hlongwane + K. Dladla - Veld
Bergville
24 K. Dladla Plot 2
Bergville
25 K. Dladla Plot 4
Bergville
26 K. Dladla Plot 7
Bergville
27 K. Dladla CA Control
Bergville
28 N. Msele Conventional Control
Bergville
29 N. Msele Plot 2
Bergville
30 N. Msele Plot 4
Bergville
31 N. Msele Plot 7
Bergville
32 N. Msele Veld
Bergville
33 M. Gumbi CA Control
Bergville
34 M. Gumbi Plot 2
Bergville
35 M. Gumbi Plot 4
Bergville
36 M. Gumbi Plot 7
Bergville
37 M. Gumbi Veld
Bergville
38 N. Zikode CA Control
Bergville
39 N. Zikode Plot 2
Bergville
40 N. Zikode Plot 4
Bergville
41 N. Zikode Plot 7
Bergville
42 N. Zikode Veld
Bergville
43 T. Zikode CA Control
Bergville
44 T. Zikode Plot 2
Bergville
45 T. Zikode Plot 4
Bergville
46 T. Zikode Plot 7
Bergville
47 T. Zikode Veld
Bergville
48 S. Mpulo CA Control
Bergville
49 S. Mpulo Plot 2
Bergville
50 S. Mpulo Plot 4
Bergville
51 S. Mpulo Plot 7
Bergville
52 S. Mpulo Veld
Bergville
53 Z. Ndaba CA Control
Bergville
54 Z. Ndaba Plot 2
Bergville
55 Z. Ndaba Plot 7
Bergville
56 Z. Ndaba Veld
Bergville
57 L.Ngubo CA Control
Skzn
58 L. Ngubo CA Trial M+B
Skzn
59 L. Ngubo Veld
Skzn
60 V. Hadebe CA Control
Skzn
61 V. Hadebe CA Trial M+B
Skzn
62 V. Hadebe Veld
Skzn
63 V. Gambu CA Control
Skzn
64 V. Gambu CA Trial M+B
Skzn
65 V. Gambu Veld
Skzn
66 C.Xaba CA Control
Skzn
67 C. Xaba Conventional Control
Skzn
68 C. Xaba Plot 2
Skzn
69 C. Xaba Plot 7
Skzn
70 C. Xaba Veld
Skzn
71 M. Mkhize CA Control
Midlands
72 M. Mkhize CA Trial M+B
Midlands
73 M. Mkhize Veld
Midlands
74 T. Mkhize CA Control
Midlands
75 T. Mkhize CA Trial M+B
Midlands
76 T. Mkhize Veld
Midlands
77 B. Nene CA Control
Midlands
78 B. Nene CA Trial M+B
Midlands
79 B. Nene Veld
Midlands
80 R. Ngobese CA Control
Midlands
81 R. Ngobese Plot 2
Midlands
82 R. Ngobese Plot 7
Midlands
83 R. Ngobese Veld
Midlands
84 D. Chamane CA Control
Midlands
85 D. Chamane Plot 2
Midlands
86 D. Chamane Plot 4
Midlands
87 D. Chamane Plot 7
Midlands
88 D. Chamane Veld
Midlands
89 M. Xulu CA Control
Midlands
90 M. Xulu Plot 2
Midlands
91 M. Xulu Plot 7
Midlands
92 M. Xulu Veld
Midlands
93 A. Nkomo CA Control
Midlands
94 A. Nkomo CA Trial M+B
Midlands
95 A. Nkomo Veld
Midlands
14
Treatments considered this season were:
Conv Control (Conventional control): These plots are ploughed and planted to a maize
monocrop using typicalfertilizerapplicationsand spacing (90cm between andin-row
spacing).
CA control:CA plots planted to a maize monocropusing typical fertilizer applications and
spacing (90cm between and in-row spacing).
CA treatment: CA trial plots: Either plots (from the 10 block 10x10m trial layout) or
strips (from the 10strip 20x5m strip trial layout) were collected from 2-4plots/strips in
the respective trials using plots 2,7 or 2,4,7 or 2,4,6,9 depending on the specific
participant’s trial. This ensures a wide distribution of plots with the full range of cropping
options namely maize only, maize and bean intercrop, summer cover crops, and pumpkin
(M, M+B, SCC, Pk)
Veld: A composite veld sample for each participant with combined subsamples of soil
from veld at the four compass points surrounding the participant’s homestead and field.
Soil sampling methods.
Sampling is done at thesame time every year, during September, after harvest and prior to the
start of seasonal rain, according to international conventions
2
.
CA treatment samples:10cm depth cores were taken (with a soil auger), taking 20
samples along a zigzag pattern across each plot or strip chosen. These were combined for
each plot/strip, thoroughly mixed and then 500 g wasplaced in aplastic bag and sealed
as acomposite sample. These bags were kept ina cool, dark place(flash frozen with ice
and kept in a cooler box) until delivery to the different laboratories within four to six
days of taking the sample.
CA and Conv Control plot samples:20 samples were taken in a zigzag pattern across
the dimension of the control plot for a composite sample; these vary from one participant
to the next and are otherwise treated in the same manner as the CA plot samples above.
Veld samples:A patch of undisturbed veld, asclose as possible to the participant’s
cropping field was chosen, to also have the same basicvisual characteristics as thefield
in question. Four subsamples weretaken at 10 cm depth at thefour compass positions
adjacent to the cropping field. Veld in smallholder farming areas under communal tenure
cannot be regarded as “pristine”, given heavy grazing patterns and frequent burning. It is,
however, assumed that the soil is undisturbed in terms of tillage and gives an indication
of the general conditions of the soil in the vicinity of the cropping fields.
4.2.2Biomonitoring using nematode indices.
Nematodes are the most diverse group of soil-inhabiting microorganisms. They play avital role
in organic material decomposition and nutrient recycling.
2
Stolbovoy, V. etal., 2007. Soil sampling protocol to certify the changes of organic carbon stock in minteral
soil of the EuropeanUnion. Version 2., Luxembourg. ISBN: 978-92-79-05379-5: EUR 21576 EN/2. 56 pp.
Office for Official Publications of the European Communities.
15
Temperature, moisture, soil type and structure and soil pH have a significant effect on nematode
populations, thus nematode population quantity and diversity differs in different agroecological
zones. Nematode abundance decreases from alkaline to acidic pH of the soil. Soil nutrients such
as, nitrogen (N) and phosphorus (P) have shown an effect in nematode richness; nematode
diversity andrichness ofgenera are higherat abundant soil N and P levels
3
. Agricultural tillage
is known to reduce nematode populations, as does low levelsof organic matter. Conversely,
minimum tillage and improved organic matter in the soil have shown to increase nematode
populations and diversity.
Free-living nematodescan be divided into fourbroad groups based ontheir diet,existing at
different trophic levels within the soil food web:
Bacterial feedersconsume bacteria. Bacterial-feeders release plant-available nitrogen
when they consume bacteria.
Fungal feedersfeed by puncturing the cell wall of fungi and sucking out the internal
contents. This interaction releases plant-available nitrogen from fungal biomass.
Predatorynematodeseat all types of nematodes and protozoa. They eat smaller
organisms whole, orattach themselves to thecuticle of larger nematodes, scraping away
until the prey’s internal body parts can be extracted.
Omnivores eat a variety of organisms or may have a different diet at each life stage.
Root-feeders are plant parasites, and thus are not free-living in the soil.
Nematodes are concentrated near their prey groups.Bacterial-feedersabound near roots where
bacteria congregate; fungal-feeders are near fungal biomass; root-feeders are concentrated
around roots of stressed or susceptible plants. Predatory nematodes are more likely to be
abundant in soils with high numbers ofnematodes. The proportion of bacterial-feeding and
fungal-feedingnematodes is relatedto the numberof bacteria and fungi in the soil. Commonly,
less disturbed soils contain more predatory nematodes, suggesting that predatory nematodes are
highly sensitiveto a wide range of disturbances.
(https://www.nrcs.usda.gov/wps/portal/nrcs/detailfull/soils/health/biology/?cid=nrcs142p2_053866).
Light soil disturbances especially coupled with compost and manure additions increase the
availability of organic residues to be consumed by bacteria,which in turn stimulatebacterial
feeding nematodes, leading to a net increase of available nitrogen for plant uptake. And although
fungal feeding nematodes are more abundant in no-till and perennial agricultural systems,
bacterial feeding nematodes are better at releasing plant available nitrogen than their fungal
feeding counterparts.
Agricultural systems designed to support a healthy soil ecosystem can more successfully defend
against plant parasiticnematodes and other crop diseases and pests. They canalso facilitate
enhanced nutrient cycling, which supports plant nutrient uptake, leading to overall healthy crop
growth. (https://nwdistrict.ifas.ufl.edu/phag/2018/11/09/benefits-of-nematodes-in-healthy-soil-
ecosystems/).
3
R Nis et al. 2021). Influence of ecological and edaphic factors on biodiversity of soil nematodes. Saudi
Journal of Biological Sciences, Volume 28, Issue 5
16
Nematodes are useful indicators of soil condition because they are ubiquitous, represent different
trophic levels of a soil food web and are convenient to work with. Several quantitative analyses
of nematode assemblages have been developed and used in monitoring programs and by
individual researchers.
Nematode population analysis - biomonitoring
The North WestUniversity Unit for Environmental Sciences and Management (Integrated Pest
Management Programme) analysed 95 soil samples by decanting and sievingfollowed by the
centrifugal-flotation method as described in Marais et al. (2017) and results were compiled in the
online programme NINJA (Nematode Indicator Joint Analysis)
(https://sieriebriennikov.shinyapps.io/ninja/).
Biomonitoring using nematode taxa is builtupon a classification based on life-cycle
characteristics. Specifically, the families, genera and species of free-living (i.e., all but herbivores
and animal parasites) and herbivore nematodes are distributed among five classes in a so-called
coloniser-persister, or c-p, series. Nematodes in c-p1 are regarded as “enrichment opportunists”,
being able tocomplete their life cyclein some cases inas short as two days and their population
sizes respond rapidly topulses of nutrient input. As a result, c-p-1 nematodes are often dominant
in disturbed environments. In contrast, c-p-5 nematodes may be considered “extreme persisters”
that are generally intolerant of disturbance and inhabit stable, mature ecosystems; they have life
cycles that may last several months. The weighted mean of the proportion of afree-living
nematode assemblage in each c-p class is designated the Maturity Index (MI) and is considered
representative of the degree of maturity of an ecosystem.
Food web diagnostics, which characterise the functional aspectsof the nematode assemblage,
consider the functional roles of different taxa in a food web based on their feeding habits. Thus,
all free-living nematode taxaare allocated to three components: a)basal component comprising
bacterivores of c-p-2 and fungivores of c-p-2; b) enrichment component including all c-p-1
nematodes and fungivores of c-p-2; c) structure component represented by allc-p-3-5 nematodes
as well as by predators of c-p-2. Each combination of c-pclass and feeding habit isconsidered a
functional guild.
Weighted proportions of functional guilds are used to infer various attributes of the food weband
give rise to an Enrichment Index (EI) and a Structure Index (SI).EI and SI correlate respectively
with the intensity of nutrient enrichment and the degree of ecosystem maturity. TheChannel
Index (CI) is calculated from the weighted proportions of fungivores of c-p-2 and bacterivores of
c-p-1 and is consideredto indicate whether the “fast” bacterial channel or “slow” fungalchannel
of energy transformation prevails in an ecosystem.
In summary the indices produced can be explained as follows:
Maturity Index 2-5 (Ranges from 2-5)
The Maturity Index serves as a measure of soil ecosystem health and ranges on a scale from 2 (disturbed)
to 5 (stable). Higher Maturity Index scores are therefore indicative of healthier soils.
Enrichment Index and Structure Index (Both indices range from 0-100)
17
The Enrichment Index represents the level of resource availability in the soil. Therefore, higher values
(>50)areindicative of nutrient enriched soils and lower (<50) values of nutrient depleted soils.
The Structure Index serves as a measure of thecomplexity of the soil ecosystem. Higherstructure values
(>50)are thereforeindicative of healthier soils with structured food webs, while lower values (<50) are
indicative of degraded soils.
Channel Index (Ranges from 0-100)
The Channel Index indicates the primary decomposition pathway withlower values (<50) being
representative of bacterial dominated pathways, while higher Channel Index values (>50) indicate fungal
dominated pathways.
The results from the NINJA platform weresubjected to ANOVA analysisto ascertain statistical
significance at (P<0,05) and a further multivariateredundancy analysis (RDA) has been done to
indicate trends across thedifferent sets of variables. Redundancy analysis (RDA) is amethodto
extract and summarise the variation in a set of response variables that can be explained bya set
of explanatory variables. More accurately, RDA is adirect gradient analysis technique which
summarises linear relationships between componentsof responsevariablesthat are "redundant"
with (i.e., "explained" by) aset of explanatory variables. To do this, RDA extends multiplelinear
regression (MLR) by allowing regression of multiple response variables on multiple explanatory
variables. A matrix of the fitted values of all response variables generatedthrough MLR isthen
subject to principal components analysis (PCA).
4.2.3Nematode diagnostics results.
The results presented are an initial and generalized analysis ofthe results. A more in-depth
analysis will follow in the final report.
Metabolic footprints
Metabolic footprints quantify the amplitude of Carbon utilisation by different food web
components, in this case nematodes. The point in the middle of a rhombus represents the
intersection of EI and SI and length of vertical and horizontal axes of the rhombus corresponds to
the footprints of enrichment and structure components respectively
4
.
Nematode metabolic footprints were used to measure/determine the effect of CA practices, using
the pooled datafor treatments,on the functioning of soil ecosystems by indicating the
magnitude/level of functions that are fulfilled by nematodes. These metabolic footprints indicate
the energy and Cflow into and throughout the soil food web via the respective decomposition
channels of the dominating nematode trophic groups.
The graphs (Figures 2,3) belowrepresent the mean metabolic footprints ofthebeneficial
nematodes identified from soil from each of the respective treatments (data was pooled for each
treatment across the area’ssites sampled) for each area.
Below is a small table of the number of samples provided for each treatment from each area.
4
Ferris, H., 2010. Form and function: Metabolic footprints of nematodes in the soil food web. European
Journal of Soil Biology, 46(2), pp.97104.’
18
Sample
Bergville
SKZN
Midlands
CA Control
9
4
7
CA treatment
35
5
11
Conv Control
2
1
-
Veld
10
4
7
As theConventional Control (Conv control) sample is very small, results forthis treatment are
likely to be somewhat skewed.
Figure 2: Left:Faunal analysis and Right: A pooled analysis for CA(treatment and control) samples compared to Conventional
tillage anda veld benchmark, for Bergville (December 2021).
Pooling of data of the respective treatments for the Bergville area revealed that the CA treatment,
CA control andConv treatment all plotted in quadrat B, while the Veldplotted in quadrat C. This
indicates that the ecosystem conditions forthe CA treatment, CA control and Conv treatment were
mature and fertile, while that for the Veld treatment was mature and enriched; both ecosystem
conditions in thetwo quadrants are conducive for beneficial nematodes to fulfil their functions
and roles in terms of soil health.
More specifically:
CA treatments, when compared to CA control and veld samples have the highest Structure
and Enrichment Indexvalues. This indicates that the soil ecosystems for CA treatments
are N-enriched, maturing, with low C:Nvalues and with bacterial dominated nutrient
cycling pathways within a regulated soil ecosystem.
The CA controls indicate a more disturbed soil ecosystem, butisstill conducive towards
a more regulated system and
C
A
B
D
19
The Veld indicates a mature soil ecosystem with balanced bacterial/fungal nutrient
cycling pathways within a suppressive soil ecosystem, meaning that there is a prevalence
of predators and disease suppressing organisms in the system.
In Bergville, the multi-cropping, crop rotation, close spacing, judicious fertilizer use, increased
organic matter, regulated pHand higher aggregate stability found in the CA treatments is conducive
to better soil ecosystem health than mono cropping with Maize only, without addition of lime, wider
plant spacing and inconsistent fertilizer applications in consecutive seasons.
For SKZN the mean value for theCA control treatment, was theonlyone that plotted in quadrat
B, while the CA treatment plotted in quadrat C, the Conv control in quadrat D and the Veld in
quadrat A. Hence the CAcontrol and CA treatments showed the best levels of soil health and the
Conv control and Veld the lowest.
More specifically:
CA treatments indicate a mature soil ecosystem with fertile soil, moderate C:N and
balanced bacterial/fungal nutrient cycling pathways within a suppressive soil ecosystem,
meaning that there is a prevalence of predators and disease suppressing organisms in the
system
CA controls have the highest Structure and Enrichment Index values. Thisindicates that
the soil ecosystems for CA control are N-enriched, maturing, with low C:N values, bacterial
dominated nutrient cycling pathways within a regulated soil ecosystem.
The Veld samples indicate a more disturbed soil ecosystem, but still conducive towards a
more regulated system.
In SKZN the CA treatmentshave moved the system more towards a balanced bacterial/fungal
nutrient cycling pathway, and increased the stability of the nematode populations, when compared
to the CA controls. Enrichment levels however are higher for the CA controls, indicating better
nutrient cycling for the CA control plots in the short term.
In the Midlands all three treatments plotted in quadrat D. The veld treatments had the highest SI,
the CA control had the lowest EI and the CA treatment the lowest SI.This means that theCA
treatments showed the leastabundance and diversity of beneficial nematodes representing the
higher c-p values (3-5) and that these fields overall were least conducive forbeneficial nematodes
to conduct their roles and functions in such ecosystems.
More specifically:
The CA Controls indicate a degraded soil ecosystem, depleted fertility levels, a high C:N
ratio, a fungal nutrient cycling pathway and a conducive soil ecosystem environment for
pest and disease control
The CA treatments indicate a moredisturbed soil ecosystem, which is N enriched, with a
lower C:N ratio and more conducive towards a more regulated system
The Veld samples are similar to the CA control plots.
In the Midlands CA treatments are improving the soil ecosystem, providing for better nutrient
cycling and regulation than the CA controls, but still within a disturbed soil ecosystem with low soil
fertility.
20
Figure 3: A foodweb analysis for CA (treatment andcontrol) samples compared to Conventional tillage anda veld benchmark,
for SKZN (left) and Midlands (right) - December 2021.
It can be concluded that the CA treatments have improved the soil ecosystem in all three
agroecological zones and that a longer periodof implementation (4-8 years), along with the
specific environmental characteristics of Bergville and SKZN, have shown improved structure and
enrichment index levels for both areas.
Nematode feeding/trophic groups
A slightly morein-depth look at thetrophic groupsfor each area, with pooled results for each
treatment, provides for a visual assessment of the proportion of each trophic group and the level
of structure and maturity of the nematode populations for each treatment.
Bergville
Figure 4: Feeding groups of nematodes, including plant-parasitic (upper graph) and beneficial (upper and lower graphs), in
soils from sites in the Bergville area.
21
Visible intheupper graph which reflects thecomposition of the whole nematode assemblage
(plant-parasitic/herbivore and beneficial nematodes) is that the Conv control had thelowest
abundance of plant-parasitic nematodes, compared to the Veld sites which had the highest
(mostly representing spiral nematodes). The Convcontrol had thehighest abundance of
fungivores (nematodes feeding on fungi), with theother three treatments containing
substantially less of this feeding group. As can be seen inthe lower graph theConv control had
the lowest abundance of bacterivores (nematodes feeding onbacteria), which is opposed to the
trend for theother treatments where this feeding group dominated. Omnivore and bacterivores
were very low in terms of abundance in all the treatments, with the Conv control having the least.
It is expected that the bacterivore, fungivore and predator populations in the CA treatment and
CA control samplesshould be higherthan the veld especially given that CA has been
implemented by these participants for >5 years. From the results, these classes aresomewhat
improved in the CA treatments, but not to the extent expected. The high level of herbivore- plant
parasitic nematodes is cause for concern. As will be indicated in the multivariate analysis below,
more attention to building of organic matter, pH regulation and management for aggregate
stability can improve this situation.
SKZN
Figure 5: Feeding groups of nematodes, includingplant-parasitic (upper graph) and beneficial (upper and lower graphs), in
soils from sites in the SKZN area.
The veld, as with Bergville, had the highest plant-parasitic nematode abundance (upper graph),
while the conventional control had the highest fungivore abundance (upper and lower graphs).
In terms of the beneficial nematode assemblages, bacterivores dominated with the highest
abundance in the CA control, followed by the Conv control, the CA treatment and Veld (upper and
lower graphs).
In the lower graph the Conv control and CA control both had similar bacterivore abundance,being
the highest followed by the CA treatment andthe veld. Predator abundance was higherthan in
Bergville sites and occurred in the highestabundance in the CA control, veldand CA treatment
with no predators being present inthe Conv control. Omnivores werethe second lowest in
abundancewith thetrend being that the CA treatment had thehighest, followed by the Veld, CA
control and Conv control with the least.
22
This indicates that in this region the CA treatments are lagging behind the CA control in terms of
soil enrichment organic matter, pH and aggregatestability, but also that CA as a practisehas
provided for improvement of the soil health and structure.
Midlands
Figure 6: Feeding groups of nematodes, includingplant-parasitic (upper graph) and beneficial (upper and lower graphs), in
soils from sites in the Midlands area.
The upper graph shows a trend that the plant-parasitic nematode abundance was similar for the
three treatments in the Midlands area, beingslightly lower inthe Veld treatment. Bacterivores
dominated, as was the case in Bergville and SKZN, being thehighest in the Veld, followed by the
CA control and CA treatment (upper and lower graphs).
In the lower graph fungivore abundance was highestin the CA treatment, followed by the CA
control and Veld. Low abundances of omnivores and predators were present in all three
treatments, with the highest abundance of predators being in the CA control and the highest
omnivore abundance being in the veld.
Comparison of multivariate nematode Indices
In the table below information for the 4 treatments is summarized for each of the Nematode
indices: Maturity, Structure, enrichment and channel indices. This provides for an analysis of the
trends in the different functional and trophic groups forthe treatments across all three areas.,
thus how the treatments are chaining the nematode population dynamics.
The results indicatea trend towards higher enrichment and structure indicesfor the CA
treatments, when compared to the CAcontrol and veld samples. Total biomass for nematode
populations is also higher for theCA treatments than the CA controls. This indicates improved
nematode population complexity and nutrient availability inthe CA treatments and healthier soils
with more stable food webs.
23
Table 7: Mean values for the nematode indices according to the 3 treatments, for all three areas, February 2022 (Note;
Conv controls not included there as there were only 3 samples for this treatment)
The information in this above table shows that the soils in all threeregions are disturbed, with
unstructured food webs, low enrichment levels and with bacterial dominated nutrient cycling
pathways. InBergville however, the soils have reasonably high maturity and structure index
values-indicating that the soils in this area aregenerally more fertile with more stable, structured
food webs thanin theother areas. This is anadvantage that can bebuilt on in the participants’
farming practices. In all three areas, but more so in Bergville (~8 years under CA), followed by
SKZN (~4yearsunder CA), CA practices (CA treatments and CAcontrols)have improved the soil
health and structure and enrichment levels of the soils.
However, when the data is combined into one analysis
the RDA showed significant differences among the
areas. This means that the location of sampling (i.e.
the area) had a significant effect (p<0.05) on the
response variables (i.e., the nematode-based indices)
as indicated by the blue lines and arrows. Higher soil
ecosystem health (MI2-5), as well as food web
structure (Structure Index) and enrichment
(Enrichment Index) were recorded in the Bergville
area.
Figure 7: RDA for the nematode indices in the three areas (Bgvl, SZKN
and Midlands)
Index nameComment
CA
Control
CA
Treatment
Veld
CA
Control
CA
Treatment
Veld
CA
control
CA
treatment
Veld
Maturity Index 2-
5[MEAN]
2.92 2.902.96 2.82 2.702.38 2.27 2.132.43
MI higher for CA (control and treaments) than conventional control plots. MI
for CA control plots higher than CA treatment plots. AllMI indices (including
veld), low and indicative of disturbed soils
Structure Index
[MEAN]
50.90 62.9067.27 71.47 56.9939.68 29.79 16.0042.65
SI higher for CA treatments than CA control plots in Bergville and SKZN, but
not Midlands. SI in Midlands higher for CA (treatment and control) than
Conventionally tilled plots. For Berville the SI is highest for veld, as expected
and is generally indicative of reaosnalby structured food web. For the
Midlandsthe CA (treaments nad controls) have imporved the structure of
the food web considerably and for SKZN food webs are unstructured.
Degraded soils are present in both the Midlands and SKZN.
Enrichment Index
[MEAN]
54.41 59.0439.28 53.83 21.6553.33 29.73 49.4237.59
The EI for CA treatments are higher than the CA controls for Bergville and
SKZN, but not Midlands. Nutrient enrichment levels are generally quite low
for all the treatments including veld.
Channel Index
[MEAN]
40.09 26.2914.10 37.50 79.4940.00 40.00 40.3026.47
CI for all treaments are higher than the veld samples. CA treament CI index
higher for CA treatments than CA controls in SKZN and Midlands, but not
Bergville.For all three reas the veld indicate bacterial dominated nutrient
cycling pahtways that are shifted towards more fungal domianted nutrient
cycling pathways through the CA cropping practises.
Total biomass,
mg [MEAN]
4.55 5.342.38 4.70 1.951.10 3.30 10.913.45
Total biomass of nematode populations have increased under CA practices ,
when compared to the veld samples in all three areas
SKZN
Midlands
Bergville
24
This result echoes the analysis done in the 2020/21 cyclewith the first ‘visible”signs of change
being ashift to a more fungal dominated ecosystem followed in subsequent years by agreater
structure and maturity of the nematode populations.
For the Bergville site a further multi variate
redundancy analysis, comparing the
treatments and various abiotic factors was
undertaken to get an indication of the
importance of these factors (e.g., pH, aggregate
stability, soil organic carbon, % clay) on the
nematode indices.
Figure 8: RDA analysis includingabiotic /edaphic factors
From this figure the edaphic conditions
(indicated by red arrows) contribute to the
observed variation in the response variables
(nematode indices), whereas the treatments do
not (indicated by red triangles). Soil pH and
aggregate stability presents a positive
association with theMI2-5 (i.e., ecosystem
health) and Structure (i.e., foodweb complexity) indices.Nematode biomass is associated with
higher clay and organic carbon content.
This outcome lends credence to the research team’s hunch/feeling/observation that positive soil
health outcomes due to CA practises rely heavily on the status of the farmer’s soil and their long-
term treatment of their soils, alongside the edaphic soil parameters of each area.
To further explore this possibility an RDA
was conducted forthe participant farmers
in Bergville, thus checking for
associations betweentreatments,
nematode indices, abiotic factors and
specific farmers.
Figure 9: Nematode-based indices and associations
with treatments, selected abiotic variables and
farmers
Farmer Numbers:
Farmer_1:P Hlongwane (Ezibomvini)
Farmer_2: Z Zikode (Ezbomvini)
Farmer_3: N Dladla (Ezibomvini)
Farmer_4: D Hlongwane (Stulwane)
Farmer_5: K Dladla (Stulwane)
Farmer_6: N Msele (Stulwane)
Farmer_7: M Gumbi (Stulwane)
Farmer_8: N Zikode (Eqeleni)
Farmer_9: T Zikode (Eqeleni)
25
The farmer number and edaphic conditions that cluster together are associated and the length of
the arrows provide an indication of the strength of this association.
From this graph farmers 5-7 show a very strong positive association of the nematode indices with
soil aggregate stability, soil organic carbon and % clay. The three farmers arefrom one village,
Stulwane, where the % clay is higher than theothertwo villages (Ezibomvini and Eqeleni),
showing the influence of the basic soil type on soil health parameters, but the aggregate stability
and soil organic carbon areparameters that are higher for these participants due to their long-
term soil management practices. For two of the participants in Ezibomvini (farmers 1 and 3) the
pH of their soils has a positive association with the nematode indices here pointing to their
improved management practices as well. Soils in the area are naturally strongly acidic. For the
other farmers in this group (farmers 2,4, and 8) thereare no significant positive associations. This
is more likely tobe related to their longer term (mis)management of their soils, or shorter CA
history, as they fall across three different villages,an observation which is corroborated from
what the research team knows about these farmers. Mrs Zikode (Ezibomvini) and Mr Hlongwane
(Stulwane) have been farming formany years, without provision of fertility enhancement
measures in their fields and Mrs Zikode (Eqeleni) ploughed her fields for more than a decade
prior to starting her CA experimentation.
5.Progress summary
This season, the support for CA experimentation and inputs has not been provided by LandCare
(KZNDARD) as has been the case for the past 4 years. Farmers were required tobuy their own
inputs, while MDF assisted in the bulk orders and transport. Due mainly to the increasingly
difficult economic situation for small scale farmers and the steep increase in prices for inputs, the
number of farmers undertaking the baby trials was much smaller this year being 85 participants
in total, compared to 185 participants in 2021/21. The total number of CMTs also reduced from
73 to 40 this season.
Care was taken to ensure that the research team was present at the lay out and planting of each
CMT this season, to ensure strict adherence to the layout and crop combinations proposed.
5.1Bergville
For this adaptive researchprocess, we have focused mainly on four villages: Stulwane,
Ezibomvini, Vimbukhalo and Eqeleni with a total of 18 CMTs and 47 baby trials.
The Okhahlamba Local Municipalityand thelocal KZNDARD extension office assisted with
tractors for both spraying and planting the CA control plots, primarily in Vimbukhalo and
Stulwane, but also in the other two villages. 20 Control plots in Stulwane, 28 inVimbukhalo and
5 in Eqeleni. They made use of the 2-row no till planters, owned by the farmer learning groups in
these villages.
CMTs were sprayed using knapsack sprayers, alsousing a tank mix of Round-up and Kemprin
(600 and 250 ml respectively,in 2x16 litres ofwater).The learning group members in each village
assisted in all CMT plantings.
26
Figure 10: The Okhahlamba LM, municipal tractor and boom sprayer provided to participating farmers in B
Figure 11: Spraying of herbicide and
pesticide 1-2 weeks prior to planting. Far
Right: Learninggroup members in
Ezibomvini assist with planting of the
1000m2 CMTs.
The main issues in Bergville this
season have been difficulty with
weeding, as participants planted
large control plots with assistance
from both KZNDARD and GrainSA
and then could not keep up with weeding. In addition, there has been substantial run-off damage
in some plots, especially those planted in January, fodder and short season maize. Insect damage
has also been substantial, specifically stalk borer as well as CMR and flea beetles. Farmers were
responsible for buying their own insecticides and many did not.
LAN fertiliserprovided through the bulk buying process was notenough, which meant a large
proportion of the CMTs were not properly top dressed. This was partly due to a lack of supply of
LAN in the local towns during December-January and partly due to a very sharp increase in prices
thereafter.
Figure 12: Above Left: Runoff damage to a short-season maize CA trial plot planted by Phumelele Hlongwane (Ezibomvini) in
January 2022. Above centre: Weedy and yellowing maize in a CA control plot which was not top-dressed using LAN for
Khulekani Dladla (Stulwane). Above right: CMR beetle infestation on sunflowers in SCC CA trail plot for Bukisiwe Ndaba
(Vimbukhalo).
27
5.1.1Ezibomvini
NOMBONO DLADLA (7th year)
10x10 Block CA trial 2021/22
Nombono planted a buffer strip of
maize and runner beans around her
1000m2trial.
Stalk borer was treated using Kemprin. Germination was goodat between 85-95% and crop growth was very good. Some signs of limited run off in
the CA trial plot.
1 M
2 M+B
3 SCC
4 M
5 M+B
10 M+CP/Pk
9 SCC
8 M+B
7 M
6 SCC
28
She also planted an extra trial plot of M+B+Pk (550m2). Nombono’s husband ploughed a plot (1650m2) below her extended trial. The maize and beans
he planted washed away and there was excessive runoff in this plot. He was not convinced about the CA practice, but now has been given a very visual
lesson on the need for this practice to save soil and crops. Nombono has recently replanted this plot. Nombono’s CA control plot is directly below her
homestead and is fedwith manure enriched run off from her kraal just above the plot. She planted a mixture of maize, beans and turnips in this plot.
Effectively there is little difference in the treatments between her CA control and trial plots; the only difference being in the spacing of maize, as close
spacing was not used in the control plot. The control plot showed signs of run-off and yellowing, due to being directly below the homestead. With the
high level of rainfall in November and December this field was inundated with run-off from the homestead above.
Right: Nombono’s extended CA trial with amaize,
bean and pumpkin intercrop
Middle right: The plot below this strip which was
ploughed by her husband, at the same time , which
now shows extensive run -off damage and all crops
were washed away.
Far-right: Nombono’s CAcontrol plot directly below
her homestead, showing some yellowing and run-
off damage.
29
ZODWA ZIKODE (7th year) -
20mx50m Strip cropping CA
trial 2021/22
Zodwa also planted a further
200m2maize only CA plot
as well as her CA control plot of Maize only (250m2).
Again, there is virtually no difference between the CA trial and CA control plots, as the same Hybrid (PAN53) was used
for both, awa the same fertilization regime. The only difference is in spacing, with a wider spacing used in the control.
As is the case in a few of the other CMT households, daughters and orgranddaughters, mostly young teenagers, were
also given a small plot each to plant.
1 M
2 M+B
3 SCC
4 M
5 M+B
10 M+CP/Pk
9 SCC
8 M+B
7 M
6 SCC
30
5.1.2Eqeleni
STHABISO MANYATHI (2nd year)-10x10 block CA trial
Mr Manyathi and a few other members of his agricultural cooperative joined the CA learning group last year. These men plant in larger fields close to
streams, wheresome irrigation is practiced using pumpsdirectly from the streams.The CA trials however have been planted asdryland plots, closer
to their homesteads. Mr Manyathi was very pleased with the increase performance in his trial plot, which is evident after1 year only. Withinthe CA
trial (planted end Nov 2021) it was obvious that weeding of the food crops was prioritized,which hampered the growth ofthe cover crop plots
somewhat. Otherwise, both germination and growth has been very good. He sprayed Kemprin for the CMR beetles and thus managed a much better
growth of cowpeas than most other participants in the area, who did not control this pest
His control plot (1600m2) was planted using the 2-row planter (mid Dec 2021), with tractor supplied through the arrangement made between MDF
KZNDARD and the Okhahlamba LM. Gemination in the control plot was not very good (~60%) and spacing was also not carefully done. It is difficult to
work with the Municipal tractor drivers, who generally do things in a haphazard way and plant at high speed. Mr Manyathi replanted seed into empty
spaces by hand.
He also planted a further plot of the PAN5a190 white short season maize (1200m2), in the 1stweek of Jan 2022. Here he planted first, then sprayed
with Roundup after planting, as he felt the 2-week waiting period, provided too much opportunity for weeds to regrow. He was advised against doing
this again, but perhaps could reduce the waiting period to 1 week, before planting
Above from Left to right: 1. The CA trial maize and cowpea intercrop.2. The SCCplot which was not weeded. 3. Maize only plotin the CA trial. 4. The short season maize
plot, recently planted with wide spacing. 6. Mr Manyathi in his CA control plot, with wide spacing and only around 60% germination.
31
THULANI DLAMINI (2nd year) Eqeleni
20mx50m CA strip cropping trial
Mr Dlamini is amember of the cooperative and has alarge vegetable garden as well as his fields. He pumps water from asmall stream nearby. He
planted the strip trial as well as a 600m2 plot of short season maize and a 1000m2 CA control plot. As with other participants the only difference in the
control plot is wider spacing of the maize. Germination and growth for Mr Dlamini’s trial was excellent.
Above Left: A view of maizeand SCC strips in Mr Dlamini’s CA trial plot. Above middle: A view of a maize and bean intercropped strip plot. Above right: The field containing
both the short season maize, (upper half) and the CA control plot (bottom half)
32
5.1.3Stulwane
NOTHILE ZONDI (5thyear)
10x10 block CA trial
Nothile’s trial is doing reasonably
well, with 85-90% germination
and meticulous weeding. There
are however some signs of
yellowing in her intercrop and
cover crop plots.
Her CAcontrol plot (28m x
115m) was planted usinga two-
row planter tomaize only
(PAN53). It showed reasonable
germination (95%) and growth. A
wide spacing was used. She
also planted a strip cropping CA trial (27m x46m) in January 2022, where she planted short season maize, beans,
lespedeza, turnips and tall fescue strips. This trial however showed very bad germination due to bird damage.
1 M
2 M+B
3 SCC
4 M
5 M+B
10 M+CP/Pk
9 SCC
8 M+B
7 M
6 SCC
33
DOMBI BUTHELEZI (6thyear)
10x10 block CA trial
She planted a monocrop maize CA
control plot (29x140m), with
wider spacing (90cm between
row). Yellowing is due to water
logging.
All crops germinated and are growing well. Dombi has maintained reasonable residue cover in her CA trial plots. She
planted a short season maize trial plot (37x44m) in January. The maize is growing despite obvious run-off issues.
1 M
2 M+B
3 SCC
4 M
5 M+B
10 M+CP/Pk
9 SCC
8 M+B
7 M
6 SCC
34
5.1.4Vimbukhalo
SIBONGILE MPULO (6th
year) 10x10 block CA trial
Sibongile’s germination and
growth have been very
impressive in her trial plot.
She also planted a monocrop
M CA control plot (23x109m). Residue in both trial and control is reasonable.
She planted a short season maize and fodder trial in January, and as with other farmers there have been issueswith
germination and growth due to run off
1 M
2 M+B
3 SCC
4 M
5 M+B
10 M+CP/Pk
9 SCC
8 M+B
7 M
6 SCC
35
BUKISIWE NDABA (4th year)
10x10 block CA trial
Bukisiwe is a member of a
cooperative which is also supported
by KZNDARD. She planted a 0,43ha
CA control plot using the two-row
planter to mono-cropped maize. She
has however struggled to keep up with weeding, as she also planted other fields and has had problems with weeds
and pests in her fields.
1 M
1 M+B
3 SCC
4 M
5 M+B
10 M+B
9 SCC
4 M+B
7 M
6 SCC
36
5.1.5Bergville Fodder trials
The outline of strip cropping (600m2) with short season maize with fodder species (Lespedeza,
Tall Fescue and turnips) was followed by only 1 of the 9 participants. Theother 7 participants
planted small mixed plots of the fodder species, partly as this planting from late December- to
Late January coincided with heavy and almost continuous rainfall which made field preparation
and planting very difficult.
Table 8: Fodder trails planted by 8 participants in the Bergville area 2021-2022.
Village
Name & Surname
Plot Type
Area Planted (m2)
Vimbukhalo
Sibongile Mpulo
Mixed Intercrop
101
Ezibomvini
Mantombi Mabizela
Mixed Intercrop
35
Zodwa Zikode
Mixed Intercrop
96
Eqeleni
Ntombakhe Zikode
Mixed Intercrop
40
Thulile Zikode
Mixed Intercrop
19
Stulwane
Khulekani Dladla
Intercropped 10x10s
500
Thulani Dlamini
Fodder
700
SSM+CP
1300
Nothile Zondi
Strip
600
Sabelo Mbhele
Mixed Intercrop
17
STULWANE
Nothile Zondi planted according to the outlined fodder strip cropping plan. Thulani Dlamini used
an animal drawn Knapik planter and planted 1-row intercrops ofshort season maize and cowpea,
with the fodder species planted inblocks towards the bottom of his field. Khulekani Dladla
planted five 10x10m plots-(2plots sorghum and sunflower, 2 plots lespedeza, tall fescue, and
turnip and 1 plot SSM and late season beanintercrop). One of his fodder plots has tall fescue
intercropped with lespedeza that was planted in the previous season 202/21.
Figure 13: Clockwise from Top Left: Thulani Dlamini and
Dlezkahe Hlongwane preparing the Knapik planter for plnting
foddr strips. Khulekani Dladla standing is his fodder trail and
strip cropping of tall fescue between Lespedeza re-growth from
202/21.
37
EZIBOMVINI
Mantombi Mabizela,who has goats in her homestead, planted tall fescue, turnipand lespedeza.
She has squeezed inrows of turnips in most of her trial plots,having found last season that the
goats really likedgrazing on the turnip tops and then continued grazing on the tubers into the dry
winter months. She alsoplanted 2 intercropped plotsnear her homestead (1 plot OPV maize +
Turnip and 1 plot Tall fescue + Lespedeza).
Figure 14: Above left: Mrs Mabizela planted turnips in between her SCC’s in her main CA trial and Abve Right: Planted small
plots of Tall fescue intercropped with Lespedeza close to her homestead.
EQELENI
Ntobmakhe Zikode planted a separate plot of short season maize and another of her fodder crops,
instead of planting the strips as suggested. She struggled with weed control and gemination. In
addition, the stalk borer load in the late maize plantings has been substantial.
Figure 155: Above Left: Mono-cropped short season maize with weed control issues in Ntombakhe Zikode’s field (Eqeleni).
Above Centre: lack of germination and weed control for the fodder plots means weeds are indistinguishable from the
planted fodder and Above Right: Sever stalk borer infestations in late maize plantings.
The fodder that was planted last year was used in different ways:
Khulekani Dlada harvested the lespedeza seeds for replanting and expansion of his
fodder plots. He then harvested, dried and chopped the vegetative part into pieces to
make a mixture of Lespedeza, crushed short season maize and mineral saltto feed his
cow as an alternative to a protein block that during the winter fodder supplementation.
38
Thulani Dlamini allowed his livestock to eat the fodder, SSM Stover and cowpeain the
field when they come back from mountain during theharvesting season as he didnot
harvest and bale his fodder.
Mantombi Mabizelaallowedher goats to eatthe turnips from herfields. This yearshe
intends to additionally cut and bale grass and fodder for her goats and to prevent
neighbouring livestock from grazing in her fields, to preserve some of the stover.
5.2SKZN
5.2.1Springvalley
GETRUDE KHWELA 1000m2 strip cropping trial (4th year)
Mrs Khwela opted to use her own traditional bean variety in the trial given the poor performance
of the PAN9292 beans the previous season. She planted onthe 25th of November 2021 and top-
dressed LAN in early January. Germination (95%) and growth have been excellent, with the
expected early canopy formation and reduced weeding pressure later in the season. She only
needed to weed twice.
Figure 16: Above Left: Mrs Khwela’s cowpea and maize and bean intercropped strips andAbove Right; Her summer cover
crop strip in her CA trial
Mrs Khwela”s CA control plot (12x5m) isplanted amaize-bean intercrop, using traditional seed
for both, with close spacing (50x50cm). Yields here may bereduced dueto invasion bythe
neighbour’s goats.
LETTA NGUBO:10x10 block CA trial (5th year)
Planting of the CMT was done together with the rest of the learning group members. Germination
and growth have been very good. Mrs Ngubo has found that the Gadra bean variety grows and
pods a lot betterin thisarea than the two Pannar varieties tried out.Sheto keep the seed to ensure
seed availability should there be challenges in sourcing seed next season.
39
She planted a 1000m2 monocroppingmaize
CA control plot., using traditional seed.
Figure 17: Letta’s Maize and bean intercropped plot in her
CA trial
BONGINHLANHLA DLAMINI: 1000m2 strip
cropping trial (5th year)
Mr Dlamini continues to hold his role as the
local facilitator helping the group mostly with
sprayingofchemicals. Heassisted the group
with pre spray before planting and will be
helping them with spraying Decis Forte for
stalk borer that was supplied on the 3rd of February 2022.
Germination and growth inhis CA trial has been very goodand canopycover here was also
reached after two weedings. In his caseSun hemp is dominating the summer cover crop plots,
unlike the other participants in the area, where sunflower is dominating. He is intending to
harvest seed from the SCC plots to milla feed for his traditional chickens and his 50 Lohman
brown layer hens.
Figure 18: Above Left: Mrs Dlamini in the strip cropping CMT and Above Right: their layers.
5.2.2Ngongonini
This group saw a substantial decrease in the number of participants for the 2021/2022season;
due in part to the poor performance of CA trial the past season, linked to the further requirement
of having to source their own inputs, without subsidization. In total only6 farmers remain, with
Mr Moses Zulu being the sole participant in Nkoneni.
MANDLA MKHIZE: 10x10block CA trial (5th year)
Mr Mkhize’s CMT is not looking good, with patchy germination and an overgrowth of weeds. The
SCC plots showed almost 0% germination. Mostly the trial clearly has not ben attended to as
needed.
40
MrMkhize also planted a 1000m² CA control plot using the two-row tractor drawn planter after
a pre-spray of Roundup. Germination and growth here have been somewhat better than the CA
trial He planted PAN 6479 basefertilized with MAP that they together with fellow members
bought collectively. He also sprayed herbicide after germination, despitehaving been repeatedly
advised against this.
Figure 19:Above Left; Mr Mkhize’s weed fille CA trial plot and Above right: His CA control, also weedy with patchy germination
and yellowing of maize towards the bottom of the plot.
MOSES ZULU: 1000m2CA strip cropping trial (5th year)
Mr Zulu’s trial was planted on the 1stof December with the help of his family; wife, two sons and
a daughter. Maize and bean strips were planted on the day; cover crops were panted muchlater
using a Haraka planter and are also growing well. Germination and growth for his trail hasbeen
good. A 1000m²control plot was planted a week earlier than the trial, the ploughed plotwas
planted with PAN 6479 maize only using the two-row planter. farmers do not appreciate time CA
takes to build up soil and arenot realizing that they are doing more damage than good with the
ploughing.
Figure 20: Above left: Moses Zulu’s Maize-SCC strip. Above right: The maize-bean intercrop strip. Mazie was showing heat
stress on the day of monitoring.
41
5.2.3Madzikane
This learning group is left with avery few members, mostly members of an agricultural
cooperative, growing maize at scale and using GM seed and ploughing. It has been arather slow
process winning farmers over to OPV’s and hybrids as well as minimal use of herbicide which are
characteristic of our CAefforts in environmental and financial sustainability. We have
successfully ticked the boxes with regards to minimum tillage, intercropping, rotations and the
use of cover crops.
COSMAS XABA: 1000m2 CA strip cropping trial (5th year)
MrXaba really admires the two-row tractor drawn planter notonly for its efficiency insaving
time and money but also forsparing the soil needless ploughing. The CA plot wherea combination
of practices; micro dosing, rotation, intercropping,relay cropping, retaining residue; are
implemented simultaneously is provingit’s worth when compared to a plot in the same field just
next toit. Maize growth, colourand health differencesare vividand provide evidence of gradually
improving results. He has also planted a 1000m2 conventional control plot, ploughed and planted
to GM maize, supported by KZNDARD.
Figure 21: Above left: CA strip cropping trial visible in the background, with yellowing stunted maize in the foreground. The
latter was not planted to CA and clearly shows the differences in production due to CA being implemented over a period of
time, vs the conventional tillage typical in the area. Above right: tow of Mx Xaba’s strip plots in his CA trial late season
weeds are present, but not impacting growth of the maize all that much.
NOMBUYISELO SHOZI:1000m2 CA strip cropping trial (5thyear)
Mrs Shozi works with her husband, and they are also members of the local savings group. Their
maize is grown more forselling than eating; maize is dried, threshed and bagged and sold locally.
She planted her strip cropping trial using the two-row no-till tractor drawn planter, on the 6thof
November 2021. She planted a small CA control plot (15x12m) to GM maize. As with Mr Xaba, late
season weeds are prevalent, due to heavy and ongoing rains through most of January.
42
Figure 22: Above left: One of Mrs Shozi’s SCC strips in her CA trial, with very good germination and growth and Above right:
A maize and bean intercropped strip.
5.3Midlands
14 CMT’s wereplanted and 2 fodder experimentation plots. One CA demonstration was
conducted with a new group which is interested in learning more about regenerative agriculture.
Given the need for theteam to be at the planting of each CMT,planting took some time. In addition,
heavy rainfall and hailstorms inDecemberdamaged the trials of thosefew participants who
planted in November. In this region most participants still prefer planting in January- due to their
supply of green mealies to the market.
For all CMTs, germination of maize has been good(85-95%), germination of beans and
subsequent growth has been poor (40-75%), germination of theSCCs has been poor and there
has been an abundance of weeds. The high level of rain fall has made it difficult for participants
to get into theirfields to undertake weeding. The season has also shown a high level of stalk borer
infestation. Kemprin has been ineffective in curbing the infestations.
5.3.1Ozwathini
NORA SIBIYA: 10X10m block CA trial (3rd yr)
Like most farmers in Ozwathini her cover crops did not look good at all. It is suspected that there
may have beenan issue with the settings on the planter where a smaller plate was used, which
released the seed more sparsely than is required for cover crops. This is possibly true, because in
the picture below the cover crops in the trial have large patches whereas the ones she planted on
her own in a separate strip had95 % germination and were growing much better. She also planted
a plot of cowpeas on the side which had over 90% germination and were growing vigorously. In
her trial, only themaize seemed to be growing well. She explained that since she planted on the
17thof December the hailstorm on the 23rdand 27thof December 2021, and a series of heavy
rains after that were detrimental to the growth of her beans.
43
Figure 23: Nora Sibiya’s CMT trial and the two separate plots with sunflowers and cowpeas.
MARTINA XULU: 10X10m block CA trial (4rdyr) from Swidi, planted her trial onthe 26thof
January this year. All of her crops had good germination including beans and summer cover
crops. The maize in plot 10 was damaged by one of her calves.
Figure 24: Martina Xulu's Trial (Swidi)
44
NOMCEBO ZONDI:
10x10mblock CA trial
(3rdyear)from
Gobinsimbi had a good-
looking trial where the
beans had a germination
of 85%and were
growing vigorously.
However, the beans in
plot 2 were growing a
lot better than thosein
plots 5 and 8. The
summer cover crops
were not doing well at
all due tothe
overgrowth of weeds. In
plot 6 and plot 9 they
appeared to be light
green and only sorghum
and SunHemp were
visible.
Trial layout:
Plot 1
M
Plot 2
M+B
Plot 3
SCC
Plot 4
M
Plot 5
M+B
Plot 6
SCC
Plot 7
M
Plot 8
M+B
Plot 9
SCC
Plot 10
M+CP/P
Figure25: Nomcebo Zondi's 10x10 trial (Gobinsimbi)
Plot 1
Plot 2
Plot 3
Plot 4
Plot 5
Plot 6
Plot 7
Plot 8
Plot 9
Plot 10
45
5.3.2Mayizekanye
DUMAZILE NXUSA:1000m2 strip cropping trial (4thyr). She used a two row no till planterand
planted her trial towards he ned of January. Her overall germination was 85% and all of the crops
emerged including beans. One highlight of planting using the tractor drawn planted was how well
the summer cover crops and beans have done this season. Her CMT, upon monitoring was very
weedy
BABHEKILE NENE: 10x10m block CA trial (4thyr).She has been one of the most consistent
farmers in implementing CA. Althoughher crops appeared to be growing well, themaize plots
had many patches as a result of birds eating her seed as it was sprouting. The bean plots had
varying degrees of germination where plot 2, had less than30% germination, plot 5 had 50%
germination and plot 8 had around 40%germination for beans. Her cover crops had over 90%
germination across all plots.
Figure 17: Babhekile Nene’s 10x10 plot
Figure 16: Above Left: Mrs Nxusa’s CA strip cropping trial and Above Right: Her CA
mono-cropped maize control plot
46
5.3.3Gobizembe
RITA NGOBESE: 10x10m block CA trail (5thyr). She has been inthe program for5 years and
has planted her trials consistently despite experiencing an array of challenges, from stunted
growth to yellowing and dwarf cobs. Over timethe CA has led to a significant improvement over
time. This season she planted her trial on the 24thofNovember. The hailstorm in December
damaged both her crops and her homestead. She was convinced there would be nothing left. The
maize and SCCs however survived reasonably well. Empty patches left by the beans that were
completely destroyed, were replanted. Her trial looks passable, despite clear variations in growth
and in the colour of her crops.
Figure 25: Mrs Ngobese’s 10 block CMT showing the crops in each plot
6.Issues and recommendations
1.For the 2ndseason in a row, COVID-19 combined with the aftereffects of the social unrest
in the province, implementation of CA has reduced by participants, wo can ill afford the
much more expensive inputs. A substantial number of participants didnot plant control
plots but paid only for the inputs for the CMTs.
2.Record keeping for the raingauges and runoff plots this season, for both the farmer
participants and the research support staff has been very limited and unreliable. This is
partly due to both the rain gauges and runoff pan buckets spilling over dueto largerainfall
events, or a number of consecutive days of raining and the farmers and staff did notdeal
well with this ‘new’ situation.A dedicated field staff member will need to be brought on
board to undertake a more methodical and monitoring of these instruments.
3.As these communities fall within one of South Africa’s strategic water source areas, there
is a strong interest from various stakeholders to support evidence-based activities that
improved water use efficiency in the system. It is thus proposed to further concentrate on
the water related indicators that have been monitored namely runoffand water
47
productivity. These two sets of indicators have also shown the most consistently positive
results for the smallholder experimentation process in CA.
4.Results for soil health and nematode populationindicators have been very inconsistent
and variable in the 3-4 years that they have been measured. It is recommended to
continue with this process, given thegreater attention now being paid to treatment
replicates and consistency across the farmers’ trials, but at a slightly less intensive level –
so fewer samples for fewer farmers to provide opportunity to focus more on the water
related indicators in the system
5.It is proposed to start with small, fenced CA trials as demonstrations - where the soil
cover (stover) is not removed at all through grazingand to compare these results with
various intensities of grazing and stover removal. Generally, the smallholder farmers are
not payingenough attention to stover retention, and this is alsoreducing the potential
benefits of the CA system
6.It is also proposed to initiate anew set of CA trials, tofocus onremediation of soil and
fields prior to starting theCA treatments. This is proposed as some of the farmers’ soil
conditions are so limiting, that the CA implementation is not showing the desired soil
fertility and soil health improvements. Results are very variable and slow to improve. The
suggestion is to do liming (to the recommended amount for those fields), do contouring
and undertake soil conservation works (swales and ditches) and to do an intensive cover
cropping process for 1-2 seasons, priorto startingagain with the maize-based
intercropping and crop rotation process.
It is envisaged that the improvement in soil health and fertility will then show a much
more consistent and positive trend.