ARTICLE CASE

The Cycle of Trust Building, Co-Learning, Capability Development,

and Confidence Building: Application of a Co-Innovation Approach
in a Multi-Stakeholder Project

M . S . S R I N I VA S A N 1 A N D G R A H A M E L L E Y 2
1
Catchment Hydrology, National Institute of Water and Atmospheric Research Limited, 10 Kyle Street, Christchurch, New
Zealand, 2Environmental Monitoring, National Institute of Water and Atmospheric Research Limited, 10 Kyle Street,
Christchurch, New Zealand
Email: [email protected]

ABSTRACT Through learnings and reflections from a water-use efficiency (WUE) pilot study, this paper examines the
use of co-innovation. Led by hydrologists, this paper tracks the cycle of trust building among stakeholders, co-learning
of WUE problem, co-developing of possible solutions and practices, identifying the need for capability development to
overcome constraints, and finally enabling confidence among stakeholders in adapting new practices. The hydrologists
built the trust among stakeholders by matching and validating stakeholders’ experiential knowledge through on-farm
biophysical observations of water use (irrigation) practices. This trust allowed the stakeholder group to identify con-
straints to improving WUE and helped the hydrologists to devise biophysical solutions and practices that support farm-
ers in better managing their irrigations. Observations also indicated that the process of capability development needed
to take into account of farmers’ experiential knowledge and integrate a learning–practice–confirmation cycle that would
boost their (farmers’) confidence in using newly acquired capability.

K E Y M E S S AG E tion, competing demands for water, multiple water users,

• An understanding of co-innovation process in and farmers’ aspirations to achieve greater water-use effi-
practice. ciency (WUE) [2]. Irrigation management in NZ was
• An understanding of how trust among stakeholders identified as a wicked problem [2, 3], and when addressing
could be used in communicating science outputs for a wicked problem, a linear tech-transfer approach, or a lin-
better outcomes. ear sequence of production and diffusion of knowledge,
• An understanding of how co-innovation could lead has been found ineffective [4]. It was highlighted that on-
to co-learning and capability development and farm irrigation management is complex as it is “influenced
improved uptake of science outputs. by several layers of technical, hydrological, climatic, soci-
etal, environmental, economic, regulatory and cultural
I N T R O D U C T I O N A N D B AC KG R O U N D factors, which individually and collectively impose con-
In New Zealand (NZ), the use of technologies such as trols and constraints on farmers’ ability and desire to
soil moisture sensors to assess demand and schedule irri- adopt efficient irrigation practices” [2], pg 139.
gation has remained stagnant over the last three decades, A pilot study was initiated in a river-based irrigation
even though the area under irrigation has been doubling scheme, the Waimakariri Irrigation Scheme (WIS) in
every 12 years since 1970 [1]. It was reasoned that the 2012, to examine the barriers to the uptake of irrigation
use of a tech-transfer approach to disseminate efficient scheduling practices among the farmers and the usefulness
irrigation scheduling practices has failed to include wider of a co-innovation approach in improving the uptake. Co-
water management issues, such as limits on water alloca- innovation is described as a multi-directional, multi-level,

Case Studies in the Environment, 2018, pps. 1–8. electronic ISSN 2473-9510. © 2018 by the Regents of the University of California. All rights reserved.
Please direct all requests for permission to photocopy or reproduce article content through the University of California Press’s Reprints and Permissions web
page, www.ucpress.edu/journals.php?p=reprints. DOI: https://doi.org/10.1525/cse.2018.001255

1
and multi-stakeholder approach, where knowledge and included an estimate of timing and amount of rainfall
input from every stakeholder are valued in every phase of expected. A daily email was also sent to all pilot farmers,
the project, from defining the problem to evolving solu- displaying the rainfall, irrigation, soil moisture, and soil
tions. The co-innovation principles used in the pilot study temperature data of the last 7 days and the 2-, 6-, and
were (1) take time to understand the problem from mul- 15-day weather forecasts. The hydrologists did not pro-
tiple (stakeholder) perspectives; (2) be inclusive of all per- vide any irrigation recommendation.
spectives; (3) value all sources of knowledge; (4) strive Throughout the season, several one-on-one conversa-
to learn from each other by actively listening and under- tions via email, phone, and in-person about the informa-
standing; (5) keep sight of the shared vision; (6) be hon- tion ensued between the pilot farmers and hydrologists.
est, open, and constructive while interacting with stake- At the end of each irrigation season (typically in May),
holders; (7) be aware of the wider context of the problem; all stakeholders were invited to come together for a work-
(8) be flexible and adaptable; and (9) stick with the co- shop (referred to as “farmer workshop” hereafter). The co-
innovation process despite its frustrations [5, 6]. innovation approach used in the pilot study was aimed
Through a selection of examples from the 5-year pilot at creating and enhancing co-learning opportunities that
study, we reflect on the cycle of trust building, co- would lead to an agreement on problem description, iden-
learning, capability development, and confidence building tification of barriers to change, and opportunities to adopt
as guided by a co-innovation approach. improved irrigation practices. The farmer workshops
would start with a hydrological appraisal of the just-
C A S E E X A M I N AT I O N concluded irrigation season, discussing how much irriga-
Pilot Study Description tion was applied, when, and why, and the differences in
The WIS is a farmer cooperative with ~240 shareholders, irrigation practices between the pilot farms (see an exam-
irrigating a total area of 18,000 ha. The irrigation scheme ple in Figure 2). The hydrologists also presented modelled
is located in the South Island of NZ. The scheme abstracts information on drainage based on measured rainfall and
water from the Waimakariri River, which undergoes sig- irrigation and published soil hydraulic properties. The
nificant flow fluctuations in summer, resulting in poor stakeholder group collectively reflected on the current irri-
supply reliability. An earlier study, which examined the gation practices and explored opportunities to improve
irrigation practices in WIS, concluded that poor supply WUE through improved irrigation scheduling. The dis-
reliability had been an impediment to efficient irrigation cussions were unmoderated, open, and transparent, and
practices and that farmers tend to manage their irrigations provided a window into the diverse perspectives and
based on supply rather than on demand (soil/crop needs) knowledge existing within the stakeholder group.
[7]. Based on this, the hydrologists in the present pilot Detailed description of the workshop processes and the
study hypothesised that irrigation scheduling could be outputs and outcomes from the co-innovation processes
improved by combining current soil moisture (a proxy for used in the pilot study are available elsewhere [2, 8]. Here,
soil and/or crop water demand), supply (river flow), and we have drawn examples from the pilot study to describe
future supplies (next 2-, 6-, and 15-day rainfall). the genesis of trust among the stakeholders and how the
The role of hydrologists in the pilot study was to estab- trust enabled co-learning and capability development. We
lish a co-innovation process that would be inclusive of all also present an example highlighting the need to extend
stakeholders—pilot study funders, hydrologists, farmers, the cycle of trust building, co-learning, and capability
industry professionals, regulators, and irrigation scheme development into confidence building in using new tech-
managers. The pilot study originally included five farms nologies and practices.
(Figure 1). At each pilot farm, a rain gauge and soil mois-
ture sensor were installed under one spray irrigator, and Building Trust
the data were relayed back to farmers electronically every Building trust between the pilot farmers and hydrologists
hour. Pilot farmers were given 24/7 web access to farm- was identified as a key task for the project team to achieve
specific, near real-time rainfall, irrigation, soil moisture its goal of improved WUE. On-farm biophysical obser-
and soil temperature data, and 2-, 6-, and 15-day weather vations of irrigation practices and observations of local
forecasts that were updated every 6 h. The forecast data weather and soil conditions that could influence irrigation

2 CASE STUDIES IN THE ENVIRONMENT 2018

FIGURE 1. Rainfall and soil gradient across the WIS area as defined by the farmers at the start of the pilot. The five pilot farms
(four dairy and one crop) are also identified in the scheme area.

FIGURE 2. Rainfall, irrigation, and drainage gradient across the pilot farms resulting in gradients in rainfall, soil type, and
irrigation management. Both rainfall and irrigation were measured and drainage estimated based on soil water holding capacity.
The gradients matched that of farmers shown in Figure 1.
scheduling (e.g., differences in rainfall and soil water hold- demand was a new concept, so the pilot farmers com-
ing capacities across WIS area) were used as tools, or mented on the need for a confirmation of their practice.
boundary objects, to initiate a conversation between the This need was intensified as hydrologists deliberately did
pilot farmers and hydrologists. When considering the not provide any irrigation recommendation. The hydrolo-
selection of pilot farms, the hydrologists met with the gists indicated that farmers need to upskill their decision-
prospective farmers and the irrigation scheme manager to making capability, taking their own constraints and capac-
gain an understanding of local soil and weather conditions ities into consideration, instead of relying on external
and how they were incorporated currently (then) incorpo- agents (e.g., the hydrologists in the pilot study) who may
rated into irrigation management decisions. Even though lack on-farm knowledge. The ability to see others’ irriga-
the farmers lacked scientific skills or tools for monitoring tion data was helpful with this capability building. When
of weather, they possessed an enormous amount of expe- irrigation scheduling between pilot farms coincided, it
riential knowledge on local weather gradients. Farmers provided those pilot farmers a positive confirmation that
understood that the prevailing weather brought more they are on the right track. However, when their schedul-
rainfall to one end of the scheme than the other, although ing did not match, it gave them a moment to reflect on
they (farmers) did not know the strength of this gradient, their practice and that of the other pilot farmer. With-
and they did not use it in their irrigation management out such a reference, these moments of pause-and-reflect
decisions. Similarly, the farmers possessed a good knowl- would not have been possible in the pilot study. Here,
edge of variability in soil water holding capacities, though the role of hydrologists was limited to enabling such co-
they rarely applied that knowledge in practice. Based on learning and capability development opportunities.
these preliminary conversations, the hydrologists deliber-
ately chose pilot farms distributed across the length and Trust Leading to Co-Learning
breadth of the scheme to quantify the gradient, specifically Once a trust in the co-innovation process was established
the rainfall differences (see Figure 1). between the pilot farmers and hydrologists, the latter
Measurement of rainfall at the pilot farms helped to delved into understanding farmers’ perception to the use
quantify the rainfall gradient across the scheme. At the of weather forecasts for irrigation scheduling. A discussion
2013 farmer workshop, when the hydrologists presented at the 2014 farmer workshop indicated that farmers gen-
the gradient information (Figure 2), it matched farmers’ erally use weather forecast at the start and end of irrigation
experiential knowledge (Figure 1). This match between seasons (termed as “shoulder” season, September, Octo-
scientific observations and experiential knowledge ber, March, and April) and rarely during peak seasons
enabled a trust in the co-innovation process. The hydrol- (November through February). Farmers perceived that
ogists took this opportunity to further the use of this any interruption to irrigation during the peak season
knowledge by highlighting the differences in irrigation would aggravate the dry conditions, thereby stressing the
applied between farms, as influenced by weather and soil crop and affecting the yield. While farmers acknowledged
water holding capacity gradients. This also allowed the that rainfall during the peak season would add to soil
irrigation scheme manager to understand the demand moisture, they generally discounted these additions, as
(water ordering) gradient across the scheme. they do not use rainfall forecasts for irrigation scheduling.
The co-innovation-based pilot study was built on the This provided an insight into farmer decision-making and
premise that stakeholders could co-learn and co-identify an opportunity for the hydrologists to analyse irrigation
barriers, opportunities, and constraints to achieve a better applications and drainage during the peak season.
WUE. To enhance this experience, each pilot farmer, in At the following farmer workshop in 2015, based on
addition to their farm, was provided with access to irri- drainage data collected at one of the pilot farms, the hydrol-
gation data from other pilot farms. This open data access ogists presented evidence for significant drainage from irriga-
helped in building trust among the farmers. When the tion during the peak season of 2014 (Figure 3). The hydrol-
pilot farmers arrived at the farmer workshop, since every- ogists highlighted that the drainage during peak season
one had prior knowledge of others’ irrigation practices, a resulted from poor irrigation scheduling, either irrigating
free and transparent discussion ensued. The concept of more than the soil could hold (exceeding soil water holding
matching current and forecast supplies against current capacities), irrigating when soil was wet from previous

4 CASE STUDIES IN THE ENVIRONMENT 2018

FIGURE 3. Rainfall, irrigation, and estimated drainage at one of the pilot farms during an irrigation season. During the shoulder
(irrigation) seasons, farmers use weather forecast but rarely during the peak season. The evidence presented here shows that use of
weather forecast during the peak season could have reduced the number of instances of irrigation-induced drainage.

irrigation and/or rainfall events, or irrigating before a signif- leaching management issues was conveyed directly to the
icant rainfall event. Unlike the earlier example, where the group, allowing them sufficient time to look for solutions.
farmers’ experiential knowledge and the hydrologists’ science At the 2015 workshop, the hydrologists, based on a
data agreed, here they contradicted. However, the trust built water balance model developed from observed drainage
in the co-innovation process helped in effectively communi- data from one of the pilot farms, presented estimations
cating the results. This also led to further discussion among of drainage at all five pilot farms (Figure 2). These data
the stakeholders, which revealed that farmers lacked capabil- provided an indication of the extent of drainage during
ity to manage irrigation and drainage together, an important the irrigation seasons, but these data were not available to
step to better WUE. farmers in real time. This meant that the farmers could not
effect a change to their irrigation practice in real time. The
Co-Learning Leading to Capability Building pilot study farmers urged the hydrologists for a solution
At the start of the pilot study (2012), soil moisture sensors that would allow them (farmers) to assess the impact of
were installed at 20 cm (depth to root zone, 40 cm), and their irrigation in real time. The hydrologists sensed that
soil moisture was used as a proxy to trigger irrigations. In real-time knowledge of drainage would provide an oppor-
2014, the local regulatory authority, Environment Can- tunity to enhance the uptake of irrigation scheduling prac-
terbury (ECan), published agricultural good management tices and weather forecast.
guidelines [9] that emphasised the need to reduce Following the 2015 farmer workshop, at each pilot farm,
drainage to control nutrient leaching, as leaching below the hydrologists installed a profile soil moisture sensor that
root zone in irrigated farms was identified as a key envi- measured soil moisture within (0–40 cm) and below
ronmental issue in NZ [10]. Since ECan was a stakeholder (40–80 cm) the root zone. The soil moisture measurements
at the farmer workshop, the emergence of drainage and within the root zone were used as a proxy for irrigation

Multi-Stakeholder Co-Innovation Processes 5

demand and below the root zone as an indicator of drainage. ture increases in the below root-zone layer to be minimal
Whenever irrigation (or rainfall) drains below the root zone, during irrigation events, which would give them the confir-
this was recorded as an increase in soil moisture in that zone, mation that all applied irrigation had been held within the
and thus the farmers were able to see the impact of the last root zone. This was a case of self-referencing and confirming
irrigation (or rainfall). Also, combining the management of their practices based on real time observations. Following a
irrigation and drainage using one sensor greatly increased the few one-on-one training sessions, one of the pilot farmers
capability of farmers and helped them to provide evidence reported significant changes to their irrigation practice. Over
to the regulator of their irrigation practices. However, these the month following the training, none of the irrigations in
changes demanded that the hydrologists remain open and that farm resulted in increases in soil moisture below root
flexible to alter their biophysical observations in response to zone, confirming an irrigation practice that resulted in no
emerging signals. drainage (see Figure 4, section A). The pilot farmer even
skipped a few irrigation events based on root zone soil mois-
Co-Development of Solutions ture measured during this period and weather forecast,
Co-development of solutions is central to a co-innovation thereby saving on time, labour, money, and water. However,
process. At WIS, farmers are required to place their irrevoca- the same pilot farmer, 2 months later scheduled irrigations
ble water orders 48-hour ahead of time. However, the pilot that had resulted in frequent increases in soil moisture in the
farmers found that the 2-day rainfall forecast to be the most below root-zone layer (Figure 4, section B). On enquiry, the
reliable, which meant that there was very little time available farmer indicated that they scheduled those later irrigations as
for them to process the forecast information and decide on they perceived that the grass growth was low, which they con-
water ordering. This quandary was discussed in one of the cluded due to dry soil conditions. Even though they could
farmer workshops. One of the pilot farmers questioned the access soil moisture data from within the root zone, in the
usefulness of 2-day rainfall forecast for their farm where no absence of any explicit reason for the lack of grass growth,
storage was available to store the ordered water in case of a they reverted back to their original irrigation practice. Fur-
significant rainfall forecast. The neighbouring farmer (not a ther investigation proved that the poor grass growth might
pilot study farmer), who was present at the workshop, offered have resulted from poor nutrient availability. The drainage
their pond for storage if such a situation arises. Neither the and leaching resulting from poor irrigation practices further
constraint nor the solution emerged from the hydrologists. depleted nutrient availability, eventually impacting the grass
The workshop offered the stakeholders a forum to discuss growth. While the capability development allowed a change
such constraints and co-evolve solutions. Here is a case where early on, a lack of confidence in that newly acquired capabil-
the farmers, because of their proximity to each other and ity resulted in the pilot farmer falling back to their original
being a part of the same irrigation scheme, were able to devise practice. This emphasised the need to take into account the
a solution within constraints. The role of the hydrologists farmer’s inherent knowledge and integrate an optimal learn-
was limited to providing an enabling environment, such as ing–practice–confirmation cycle when such new technolo-
the farmer workshop, where such constraints to change could gies and practices are introduced.
be discussed and resolved, where possible.
CONCLUSION
Transiting from Capability Building to Confidence Building A co-innovation approach provides opportunity to all stake-
Finally, the cycle of learning, practice, and confirmation was holders to explore and learn about a problem from multiple
found to be very important in sustaining the uptake of new perspectives. The role of researchers in a co-innovation set-
technology or practice, such as the use of profile soil moisture ting varies with context, which itself could vary during the
sensor. Such a cycle needs to be repeated more than once span of a project. Trust among stakeholders and in the co-
to ensure the learning is well integrated into practice. When innovation process is seldom recognised as an outcome due
the profile soil moisture sensors were installed in 2015, the to its non-tangible nature. However, in a multi-stakeholder,
hydrologists worked with the pilot farmers to ensure that building trust among participants is a key enabler to sharing
they understood its operation, i.e., soil moisture within root perspectives, co-learning and co-definition of problem space,
zone was proxy to irrigation demand and below root zone to and co-development of solutions. Biophysical observations,
drainage. The pilot farmers were told to keep the soil mois- specifically observations of stakeholders’ (on-farm) practices,

6 CASE STUDIES IN THE ENVIRONMENT 2018

FIGURE 4. The cycle of learning–confirmation–practice is critical in enabling and establishing new practices. The farmers
trained in managing irrigation and drainage simultaneously were able to eliminate irrigation-induced drainage in December and
January (see section A), but practiced irrigation that led to drainage in February and March (see section B).

provide a good platform in establishing stakeholders’ trust in 2. Apart from biophysical observations of
a co-innovation process. stakeholders’ irrigation practices, how else
Even when on-farm observations contradicted stake- could trust be established in a multi-
holder perceptions, the trust in co-innovation process could stakeholder setting?
enable moments of reflection and (co-)learning. In the irri-
3. While transferring specific on-farm observa-
gation pilot study described here, the co-learning was notice-
tions of irrigation practices from one stake-
able among all stakeholders and not just among pilot farmers
holder group to another, how could the
(end-users). For instance, the hydrologists learned to present
experiences and lessons be transformed across
information that takes into account of stakeholders’ percep-
the groups?
tions and knowledge gaps. One of the key learnings for all
stakeholders was the need to spend more time in practising
the lessons learned. In the absence of any confirmation of AUTHOR CONTRIBUTIONS
their practice, the confidence in using new practices could The lead author MS is the programme manager for the
decline over time, resulting in old habits to creep up. research programme “Justified Irrigation” and the lead
researcher of “Primary Innovation” programme. GE is an irri-
gation scientist who contributed to the designing and imple-
C A S E S T U DY Q U E S T I O N S
mentation of Justified Irrigation and Primary Innovation pro-
1. What are the advantages and limitations of a
grammes. Both authors contributed equally to idea concep-
co-innovation process in a multi-stakeholder
tualisation, funding acquisition, and investigation.
setting?

Multi-Stakeholder Co-Innovation Processes 7

AC K N O W L E D G E M E N T S Applying a cognitive framework to a pilot study. Rur Ext Inno
We thank the pilot study farmers and the Waimakariri Sys J. 2018;14(1): 137–141.
4. Leeuwis C, Aarts N. Rethinking communication in
Irrigation Scheme for their continued support and coop-
innovation processes: creating space for change in complex
eration in undertaking this study. systems. J Agric Educ Ext. 2011;17: 21–36.
5. Coutts J, White T, Blackett P et al. Evaluating a space for
FUNDING co-innovation: the practical application of nine principles for
We acknowledge the New Zealand Ministry of Business, co-innovation in five innovation projects. Outlook Agric. 2017:
doi:10.1177/0030727017708453.
Innovation and Employment for funding the programmes 6. Nederlof S, Wongtschowski M, van der Lee F. Putting
and Primary Innovation (CONT-30071-BITR-AGR) and Heads Together. Agricultural Innovation Platforms in Practice.
Justified Irrigation (CO1X1617) that supported the work Amsterdam, The Netherlands: KIT Publishers; 2011.
presented here. 7. Srinivasan MS, Duncan MJ. Droughts and irrigation:
study in a river-based irrigation scheme in New Zealand.
2012;138(1). doi:10.1061/(ASCE)IR.1943-4774.0000372.
COMPETING INTERESTS 8. Srinivasan MS, Jongmans C, Bewsell D, Elley G. Research
The authors declare that no competing interests exist. idea to science for impact: tracing the significant moments in
an innovation based irrigation study. Agric Water Manage. 2018:
REFERENCES doi:10.1016/j.agwat.2018.08.045.
1. Irrigation NZ. Irrigation in New Zealand. 2016. Available: 9. Environment Canterbury. Canterbury Water. 2017.
https://www.irrigationnz.co.nz/. Available: http://www.canterburywater.farm/.
2. Srinivasan MS, Bewsell D, Jongmans C, Elley G. Just-in- 10. Cameron KC, Di HJ, Moir JL. Nitrogen losses from the
case to justified irrigation: applying co-innovation principles in soil/plant system: a review. Ann Appl Biol. 2014: doi:10.1111/
an irrigation scheme. Outlook Agric. 2017;46: 138–145. aab.12014.
3. Boyce W, Srinivasan MS, Turner JA, Fielke SJ, Percy H.
When to co-innovate to solve complex agricultural challenges?

8 CASE STUDIES IN THE ENVIRONMENT 2018