A systemic appreciation of the interplay between technological transformation, rural welfare, collective action and water conflicts in Mount Kenya.

Andrea Cavicchi ^{1 2}, Michelle Bonatti ², Sandro Luis Schlindwein ³, Götz Uckert ², Bancy Mbura Mati⁴, Stefan Sieber ^{1 2}.

Authors Email:

1Humboldt-Universität zu Berlin, Thaer-Institute of Agricultural and Horticultural Sciences, Germany;

Sandro Luis Schlindwein: sandro.schlindwein@ufsc.br

2Leibniz Centre for Agric. Landscape Res. (ZALF), Sustainable Land Use in Developing Countries (SusLAND), Germany;

3Universidade Federal de Santa Catarina, Centro de Ciências Agrárias, Departamento de Eng. Rural, Brazil

4Jomo-Kenyatta University of Agriculture and Technology (JKUAT), Department of Agricultural Engineering, Kenya.

Corresponding Author: Andrea Cavicchi; Email: andrea.cavicchi@zalf.de

First Author: Andrea Cavicchi; Email: andrea.cavicchi@zalf.de

Michelle Bonatti: michelle.bonatti@zalf.de

Götz Uckert: goetz.uckert@zalf.de

Bancy Mbura Mati: b.mati@jkuat.ac.ke

Stefan Sieber: stefan.sieber@zalf.de

Abstract

Erratic rainfall, over-withdrawal, and catchment degradation are intensifying water scarcity and heightening highland-lowland tensions in Mount Kenya. Although rainwater harvesting, micro-irrigation, and solar-powered irrigation systems are increasingly promoted as climate-smart strategies, adoption among smallholders remains limited by upfront costs, technical constraints, poor water quality, restricted access to credit, and weak extension services. To examine the systemic barriers to technology adoption, this study conceptualizes the uptake of RWH and SPIS within a Social-Technical-Ecological Systems (STES) framework and applies a participatory qualitative system dynamics approach. Using Causal Loop Diagrams developed from interviews (N = 40), participatory modelling (N = 8), and stakeholder validation, the research maps feedbacks among ecological processes, governance mechanisms, farmer decision-making, technical components and innovation uptake. Findings show that flat-rate water fees, limited monitoring capacity, and insufficient funding for Water Resource Users Associations reinforce overuse and diminish incentives for efficient irrigation adoption. Ecological degradation -including riparian decline, siltation, and pressure over forest systems- further constrains technology compatibility, while risk perception over crop failures, vandalism, and wildlife incursions affect farmers’ willingness to invest. Conversely, cooperative models, revolving funds, and demonstration farms emerge as important leverage points for expanding access to finance, strengthening technical skills, and improving market linkages. The study concludes that sustainable water management requires coordinated interventions that align governance reform, ecological restoration, and farmer-led technological innovation. A STES perspective, while bridging the gap between socio-technical and social-ecological research in sustainability transition, provides a holistic lens to navigate the social-ecological and technical drivers shaping water security and rural resilience in Mount Kenya.

Key Words:

Causal Loop diagrams

participatory modelling

rainwater harvesting

social-ecological systems

socio-technical systems

solar powered irrigation systems

systems dynamics modelling

technological transition

water governance

List of Abbreviations

ASAL

Arid and Semi-Arid Land

CAS

Complex Adaptive System

CETRAD

Centre for Training and Integrated Research for ASAL Development

CFA

Community Forest Association

Critical Infrastructure

CLD

Causal Loop Diagram

CPR

Common Pool Resource

CSO

Civil Society Organization

CWP

Community Water Project

EEOR

Endogenisation, Encapsulation and Order-oriented Reduction

FES

Forest Ecosystem Services

FGD

Focus Group Discussion

HTE

Human–Technical–Environmental

KFS

Kenya Forest Service

MKEWP

Mount Kenya Ewaso Water Partnership

MLP

Multi-Level Perspective

NDMA

National Drought Management Authority

NWHSA

National Water Harvesting and Storage Authority

NGO

Non-Governmental Organization

QSDM

Qualitative System Dynamics Model

RWH

Rainwater Harvesting

SACCO

Saving and Credit Cooperative Organization

System Dynamics

SES

Socio-Ecological System

SETS

Socio-Ecological-Technical System

SPIS

Solar Powered Irrigation System

STES

Social-Ecological-Technical System

STS

Socio-Technical System

WRA

Water Resource Authority

WRUA

Water Resource Users Association

Introduction

Water access and management have long been sources of tension in the Mount Kenya region (Mutiga et al., 2010). Prior to the establishment of Water Resource Users Associations (WRUAs) in the late ´90s, highland-lowland competition over water frequently escalated into conflict, undermining livelihoods and exacerbating inequality. In a region where agriculture depends heavily on consistent water supply, such instability had and keep retaining serious implications for food security (Liniger et al., 2005; Mutiga et al., 2010; Gichuki, 2002; Kiteme & Gikonyo; 2002). Since the creation of WRUAs under the Water Act (GoK, 2002), some progress has been made in managing local water resources through community-led governance and water rationing programs (Kiteme & Gikonyo, 2002). However, challenges still persist. Over-abstraction for irrigation purposes (often illegal, particularly during the dry season) remains widespread, currently accounting for up to 95% of water withdrawal in several sub-basins of the region, with recurrent disruption of river flows, water scarcity and rationing programs (Mutiga et al., 2010).

In response, various entities—including county governments, (WRUAs and Civil Society Organizations (CSOs) —are promoting rainwater harvesting (RWH), especially roof top catchments, and on-farm storage solutions, such as ponds, water pans, and storage tanks, to enhance climate change resilience and reduce pressure and dependence on surface water sources (Mati et al., 2022). Despite the efforts, the uptake of these practices is hindered by high initial costs, technical barriers, lack of both know-how and accessible financing mechanisms. Moreover, many storage facilities remain underutilized due to the high cost of pumps and irrigation technologies, inflation-related energy price spikes, and inefficient irrigation practices that lead to the quick depletion of the water collected during the rainy season (ibid). Concurrently, solar-powered irrigation systems (SPIS) have emerged as a promising alternative, becoming widely employed in several commercial farms around Mount Kenya. Their low operating costs, the energy independence, and the reduced emissions are driving the increasingly growing demand for the technology, both locally and worldwide (Mohammed Wazed et al., 2018; Aliyu et al., 2018). Yet the adoption by smallholder farmers active in the region is limited to isolated cases (Lybbert & Sumner, 2010).

Case studies (Alemu et al., 2017; Andersson et al., 2024; Budiman et al., 2024) have displayed how the relationship between technology adoption and social-ecological processes is often non-linear and poorly understood. Traditional models either emphasize social-ecological systems (SES)—treating technology as an external factor—or focus on technology uptake without adequately considering environmental feedback on adoption decision. As Smith & Stirling (2010) argue, overlooking the role of technology as a mediator between society and the environment limits our understanding of technological transformation and resilience. Meanwhile, sustainability transition literature often centres on socio-technical systems (STS), but tends to marginalize ecological dynamics (Ahlborg et al., 2019; Andersson et al., 2024).

This study addresses that gap by advancing a Socio-Technical-Ecological Systems (STES) perspective. Using a participatory system dynamics approach through Causal Loop Diagrams (CLDs), we explore the interactions between water governance, ecological processes, farming practices, and the adoption of climate-smart technologies, specifically RWH, drip and sprinkler irrigation and SPIS. The research focuses on the Timau sub-catchment, a critical yet under-studied sub-catchment of the Mount Kenya water tower. By integrating Systems Theory (White, 2015), Common Pool Resource (Ostrom, 2002) and Collective Action Theory (Ostrom, 2009), this paper aims to (1) map the causal dynamics affecting irrigation practices and highland-lowland water conflicts, and (2) assess how the upscaling of RWH and SPIS might steer the system toward more resilient and equitable outcomes.

The research is guided by the following objectives: i) to systemically examine the dynamics that leads to conflicts over water access, and ii) identify the vicious cycles of barriers that perpetuates and hinder adoption of RWH, micro irrigation technologies and SPIS, while reinforcing water disparity dynamics within the sub-catchment. The study is formulated under the following research questions:

What are the main Social-ecological-technical interactions within the small scale farming sector affecting water resources´ and environmental flows´ dynamics within the Timau river system?

What are the main barriers to the adoption of RWH and SPIS, and how could their upscaling transform the social-ecological dynamics of the system under study?

While participatory CLDs have been widely used in water security and vulnerability studies (Sohns et al., 2021; Polaine et al., 2022; Alamanos et al., 2025; Neely & Walters, 2016; Ram & Irfan, 2018), few have integrated both environmental and technological feedbacks in a single framework. On one side, some system thinking practitioners (Bouchet et al., 2022; Hossain et al., 2020; Noor et al., 2022) have focused on the social-ecological system´s (SES) dynamics that shape specific water-related environmental problems, often leaving technology as an exogenous variable. System´s dynamics studies that focus instead on drivers of technology adoption decision (Adebiyi & Olabisi, 2022; Agnew et al., 2018; Zahng et al., 2021) are limited and focus mostly on the consumers´ behaviour and socio-economic variables affecting adoption choices, treating the environment as a passive context.

This study offers a contribution to the systems thinking and SES literature by advancing a Social-Technical-Ecological framework to address the interplay between water governance, ecological processes, socio-economic conditions, and the adoption of climate-smart technologies. Also, by targeting RWH and solar-powered irrigation systems (SPIS) within a participatory system dynamics approach, the research addresses the barriers and enabling factors for the operationalization of the water-energy-food nexus at the farm level. To date, the modalities through which these specific technologies interact with social, institutional and ecological dynamics to support climate change adaptation in smallholder farming systems, particularly in the Mount Kenya, have not been explored by system´s thinking. Moreover, the application of participatory CLDs enables a grounded understanding of the feedback mechanisms driving both water over-abstraction, social conflicts, and technological adoption, while revealing critical governance gaps—such as weak enforcement capacity and limited institutional support—that constrain sustainable transitions.

Theoretical Background

Complex Adaptive System Theory and SES Research

There is a well-consolidated recognition that environmental and natural resource problems occur at the intersection of complex natural and social systems, where holistic, adaptive and inclusive approaches are essential to disentangle and interpret human-environmental processes (Virapongse et al, 2016). In this context, Social-Ecological Systems (SES) science conceptualize the environment as an open system consisting of ecological and social processes and components, which interact within a dynamic network of structures that facilitate interdependencies and feedback (Virapongse et al., 2016; Berkes et al., 2008). In the definition provided by Anderies et al. (2008), SES is considered as an “ecological system intricately linked with and affected by one or more social systems”. The SES perspective is rooted to the concept of resilience and it is founded on the theories of complex adaptive systems (CAS) and their “conceptual underpinnings of nonlinear dynamics, and uncertainty, as well as notions of multiple stability domains and adaptive management” (Reyers et al., 2018).

Following the SES approach and the concept of CAS, Ostrom (2007) proposes a diagnostic, nested framework for analysing the diverse variables that shape resource governance outcomes, known as SES framework. This framework provides a theoretical lens to address and analyse specific social-ecological configurations connected to Common Pool Resources (CPRs), characterized by rivalry in access and low excludability— i.e. when resources are limited and widely shared, making individual overuse likely if left unregulated (Ostrom, 2002; Ostrom et al., 1994). Contrarily to Hardin´s theory of the Tragedy of the Commons (1968) the SES framework supports a shift away from top-down, one-size-fits-all policy approaches by focusing on context-specific variables that shape collective action and resource outcomes. This approach acknowledges that the dynamic network of interactions responds and behave unpredictably with respect to the behaviour of the components, and that it is adaptive, as the system self-organize in response to change-initiating micro events, or collection of events (Miller & Page, 2009). By understanding interdependent linkages between social and environmental components, the framework allows to identify the main factors influencing the achievement of sustainability goals on different systems, levels and scales (Partelow, 2018; Ostrom, 2009).

With its structured approach to identify and classifying system elements and interactions to understand the system holistically, the SES framework is applicable to a wide range of fields, as the variables highlighted by Ostrom can be defined and modified according to the contexts (Ostrom 2007, 2009; Partelow, 2018). Nevertheless, most studies relating to the SES Framework, which have been described by Grupper et al. (2025) as SES thinking, commonly emphasizes five interrelated system characteristics: causal relationships, feedback loops, non-linearity, heterogeneity, and cross-scale dynamics. These five characteristics enable SES frameworks to examine elements across multiple scales, offering practitioners a more comprehensive view of complex adaptive systems (Virapongse et al., 2016).

On the other hand, while the literature on social-ecological systems recognizes technology as an important influence on resilience, it rarely considers the dynamics of change in depth. Technology plays four main roles in relation to ecosystems: it provides sensors and information on the ecological state, stimulates economic growth and social restructuring with impacts on SES, improves resource efficiency ("cleaner technologies"), and aims to repair existing environmental impacts. Despite its importance, technology is often conceptualized as exogenous in SES (Smith & Stirling, 2008).Yet, while SES thinking avoids a technological focus, it still attributes to technological innovation a critical role to drive sustainable change (Ahlborg et al., 2019). Technological transformations (referred to in SES research as “transformation to sustainability”) are essential when current system dynamics — like inequality, environmental degradation, or entrenched power asymmetries — perpetuate unsustainable development (Reyers et al., 2018; Scoones, 2010; Scheffer et al., 2009). It is important to differentiate here between transformation and incremental adaptation (Reyers et al., 2018). While transformative capacity and adaptive capacity are both crucial for resilience, they require different skills, conditions, and mindsets — and in some cases, efforts to adapt may hinder transformation (Olsson et al., 2017; Marshall et al., 2012).

Studies show that successful transformations often emerge from “under-the-radar” initiatives, but, in order to succeed, they must align with broader enabling conditions and meet key criteria for transformative potential, such as inclusivity, adaptability, and capacity for systemic impact (Reyers et al., 2018). Indeed, in socio-technical systems´ (STS) science, systemic frameworks consider technological innovation as a complex adaptive system per se, with non-linear diffusion and feedback loops, where institutional intervention, social and political factors shape technology and its utilization, and, at the same time, technology shapes societal configurations and interactions (Wei et al., 2018).

But sustainability transitions research and its frameworks (such as the multi-level perspective and innovation systems approaches), often tends to "box" ecological dynamics, making their deep connections with ecosystems invisible within analytical frameworks (Andersson et al., 2024). Instead, the field of socio-technical studies focuses on an understanding of technology as socially contingent and a terrain of struggle with socio-political consequences. In this view, society and technology are mutually constitutive: social processes shape the development and use of technology, but technologies, in turn, embody power relations and come to constitute our lifeworld, reshaping our social interactions and practices (Ahlborg et al., 2019).

Social-Technical-Ecological System Framework

Given the increasing influence of human activity on environmental change and the crucial role of technology in mediating human-environment relationships, there is a strong argument for adopting a co-evolutionary view of society, technology and nature (Andersson et al., 2024). This becomes crucial especially when considering that the resilience of socio-technical regimes (i.e. their capacity to persist and withstand shocks and stresses) can actually undermine the resilience of social-ecological systems (Smith & Stirling, 2008) by perpetuating practices and processes that are detrimental to the their stability (e.g. the intense, and subsidized, plowing and monoculture practices that led to the Dust Bowl in the 1930s in USA and Canada). Therefore, the technology´s interrelatedness with social and ecological elements brings the necessity to treat the technology as a “first class” system component (Mertzani & Pitt, 2023). In this light, technology is an integral part of the ecosystem due to its ability to transform human actions and interaction with the environment, often enabling new kinds of interaction (ibid).

This requires a bridging approach that can endogenize both ecological and technological elements, a socio-technical-ecological system (STES) or social-ecological-technical system (SETS) framework (Wei et al., 2018; Mertzani & Pitt, 2023). While there is still no academic consensus or a clear definition of STESs or SETSs, this system´s conceptualization is based on the human, environmental, and technical components, material and non-material, whose presence and interactions determine the state, structure and functions of a system (Selin & Selin, 2023). This system framing can provide a more comprehensive view on the dynamics and outcomes of sustainability transitions, taking into account more effectively ecological elements (Andersson et al., 2024; Ahlborg et al., 2019). Yet, despite repeated calls to bridge the gap between social-ecological and socio-technical systems studies, collaboration between these fields has been limited (ibid).

Some authors have attempted to operationalise this approach and bridge the gap in empirical research. For example, in their study on the impact of an extreme precipitation event, Helmrich et al. (2023) examine critical infrastructures (CIs) in Phoenix, Arizona as complex adaptive systems with multi-level interactions and emergent behaviours. By conceptualizing CIs as STES, they demonstrate how integrated decision-making across technological, ecological, and social dimensions can strengthen infrastructure resilience to climate change. Sorge et al. (2022) introduced the SETFIS framework (Socio-Ecological-Technical-Forest-Innovation Systems) to analyse governance innovations in European forest ecosystem services (FES). Building on the SES framework, SETFIS incorporates socio-technical systems (STS) thinking and the multi-level perspective (MLP) to account for innovation dynamics and long-term transitions.

The “human–technical–environmental systems framework" (HTE) by Selin and Selin (2023), on the other hand, identifies five key components: human, technical, environmental (material), and institutions and knowledge (non-material). By applying this framework through a matrix-based approach, the authors display how upstream interventions, like changing energy sources, yield long-term health benefits, while downstream actions such as food risk awareness campaigns offer short-term but limited impacts. Lastly, Wei et al. (2018) focus on redirecting technological innovation toward sustainability in river basins. While not explicitly proposing a STES model, their work highlights the need to better understand how technologies—especially those related to water supply, use, and regulation—affect both ecological integrity and social outcomes. They identify significant gaps in analysing the evolution and function of technology within social-ecological contexts and propose a mechanistic conceptual framework to evaluate technology's systemic impacts and support more sustainable development trajectories.

Conceptual Framework

System thinking in SES research

SESs and STSs are inherently complex and dynamic, marked by nonlinearity, feedback processes, and time delays, therefore understanding them requires moving beyond reductionist approaches toward a systems-based perspective that emphasizes interactions rather than isolated components. System dynamics (SD) has become a prominent modelling approach for analysing the behaviour of SESs (Filatova et al., 2016). Given its capacity to integrate a wide range of biophysical and socioeconomic variables into a single coherent framework, it offers a comprehensive and dynamic perspective on complex systems (Elsawah et al., 2017). Additionally, SD models support intuitive visualization of complex relationships, enhancing transparency and accessibility for both researchers and stakeholders (Király & Miskolczi, 2019). These strengths make SD particularly valuable for developing participatory models and decision-support tools in SES management, as evidenced by its broad application across a variety of contexts (Elsawah et al., 2017; Martínez-Fernández et al., 2021). Moreover, through CLDs, systems thinking can foster discussion among experts, to test hypotheses, to develop research questions, to identify gaps in policy and be helpful for theory building in general (Helmrich et al., 2023).

In this light, we develop a conceptual framework for the qualitative modelling of the dynamic relationships between water governance, social-economic processes, technological components that affect system´s behaviour, catchment system and water resource dynamics within the sub-catchment of Timau (Upper Ewaso Ng´iro Basin, Kenya). The sub-catchment is affected by sustainability issues which exhibit characteristics of collective action problems typical of common pool resources (CPRs), such as resource competition, over-abstraction and inequity in water access.

This conceptual framework (shown in Fig. 1) serves as a diagnostic tool to understand the catchment system and capture its main dynamics within four specific domains, namely i) ecological sub-system, ii) socio-economic subsystem, iii) governance sub-system and iv) technology sub-system. The SES framework (Ostrom, 2009) is applied at the first-tier level to categorize key system components (e.g., Resource System, Governance System, Resource Units, Actors, Interactions). The framework is an organizing scaffold for coding challenges raised in interviews, without attempting to populate all second-tier variables. This selective use allows flexibility in capturing locally meaningful dynamics while maintaining conceptual coherence. Contrarily to other SES frameworks, we consider technology as an adaptive complex system to understand how technological development influences and is influenced by the co-evolution of socio-economic, governance and eco-hydrological systems. The technology targeted in this study are those related to i) water supply, ii) water use, iii) resource monitoring, iv) water storage, vi) and farm irrigation.

It is important to highlight that the aim of this modelling process is not to achieve a deterministic or predictive representation, but to construct a stakeholder-informed causal model that reveals perceived feedback dynamics and leverage points affecting RWH, irrigation practices and technologies and SPIS adoption. Frameworks are heuristic models and guides for reflection on reality and environmental problems rather than descriptions of reality (Scoones, 2015). We believe that a STES framing can enhance our understanding of system´s dynamics and help to identify processes and feedbacks, specifically the technological contingencies that mediate and shape human-nature interactions and consequences that result in non-linear outcomes. This process can result in a successful mapping of STES dynamics and guide improvement-oriented research.

Fig. 1

Conceptual framework.

Materials and Methods

Case Study Area

The Timau sub-basin (Fig. 2) forms part of the upper Ewaso Ng'iro North basin, covering about 2,175 km², with Timau itself spanning 221 km². Its topography is diverse, ranging from 3,800 to 1,770 m above sea level. Administratively, it lies within Meru and Laikipia Counties, crossing Buuri and Laikipia North constituencies and including parts of four wards such as Timau and Ethi townships. The region experiences a bimodal rainfall pattern, with long rains from March to May and short rains from October to December. However, declining annual precipitation has become frequent in recent years (Timau WRUA, 2023).

Hydrologically, Timau has a dense river network where springs provide the main domestic water source. Wetlands exist but are increasingly threatened by activities such as grazing and logging. Land use is diverse, encompassing small-scale mixed farming, large-scale agriculture, livestock rearing, wildlife conservation, and pastoralism—the latter dominating in the lower zones. Smallholders practice both rain-fed and irrigated agriculture (Timau WRUA, 2023). The sub-basin’s 2019 population stood at about 17,000 people spread across nine sub-counties. Livelihoods have shifted significantly in recent decades, with agriculture and livestock remaining the dominant economic activities (Timau WRUA, 2023; Dell’Angelo et al., 2016).

Freshwater management poses a major challenge, driven by rising demand and declining availability (Mutiga et al., 2010; Lesrima et al., 2021). Irrigation accounts for roughly 64% of all freshwater use in the basin (Timau WRUA, 2023; Rural Focus Ltd, 2022). A study by Rural Focus Ltd under the Timau River Project revealed that small-scale farmers represent 86% of surface water abstractors, responsible for 58% of the abstracted volume (25,693 m³/day), while large-scale farms account for only 6% of abstractors but 13% of the volume (5,841 m³/day). Most abstractions are unauthorized, with 81% lacking a permit, and only 12% of the total volume is officially licensed (Rural Focus Ltd, 2022). The sub-basin is therefore classified as “Alert,” indicating increasing scarcity, groundwater over-abstraction, deteriorating water quality, and a rising potential for water-related conflicts (Timau WRUA, 2023).

Fig. 2

Administrative boundaries of Timau WRUA (own elaboration).

TIMAU WRUA

Water governance in the sub-basin relies largely on the Timau WRUA. The WRUAs were established in the Ewaso Ng'iro basin following awareness-raising campaigns for participatory water management and conflict resolution (Liniger et al., 2005). WRUAs are catchment-level decision-making units (McCord et al., 2017) following water sector reforms, particularly the Water Act of 2002 (GoK, 2002). The WRUA represent all members of its basin of jurisdiction who hold a water use permit from the Water Resources Authority (WRA) (McCord et al., 2017).

Timau WRUA, established in 2005, is managed by a committee representing community user groups (or Community Water Projects), large-scale farms, and households. It collaborates closely with the local WRA offices. Its main role is to prevent and resolve water-use conflicts through allocation agreements, often via rotational schedules. It also monitors resources, disseminates information, and builds user capacity (McCord et al., 2017). The WRUA is legally registered and works with key partners such as WRA, Community Water Projects (CWPs), Community Forest Associations (CFAs), and county authorities to integrate community-based participation in water and environmental governance (Timau WRUA, 2023).

Community Water Projects

CWPs form the grassroots units of WRUAs. Some date back to the colonial period, while others expanded after independence through government support (McCord et al., 2016). Timau WRUA comprises 49 CWPs (Timau WRUA, 2023), each typically governed by a management committee (chairperson, secretary, treasurer, etc.) responsible for water abstraction, distribution, and infrastructure maintenance at the local level.

CWPs draw water mainly from rivers and springs and enjoy autonomy in defining institutional rules, monitoring systems, and rotational water schedules. Their size often determines the complexity of these arrangements, especially during dry seasons (Richards & Syallow, 2018). CWP representatives sit on the WRUA committee to participate in catchment-level decision-making on water allocation and conservation strategies. However, they must also comply with policy directions from higher institutions such as the WRA and WRUA (McCord et al., 2016; Richards & Syallow, 2018).

Community Forest Associations (CFA)

CFAs are community-based associations responsible for participatory forest management in Kenya, established under the Forest Act, 2005. CFAs work in partnership with the Kenya Forest Service (KFS) to improve forest cover and livelihoods of surrounding communities. These activities include forest patrolling, fire-fighting, tree nurseries and plantations. CFAs also engage in income-generating activities, such as the Plantation Establishment and Livelihood Improvement System (PELIS) (Musyoki et al., 2016). These include access to forest products (firewood, pasture/fodder, medicinal products, honey, etc.), monetary benefits or income-generating activities (Musyoki et al., 2016; Ongugo et al., 2008).

CFAs also help monitor forest conditions and manage disputes internally and externally. Their collaboration with WRUAs is crucial where forest and riparian ecosystems intersect. Since forest degradation directly affects river flow and water quality, CFAs’ conservation work aligns with WRUAs’ objectives to protect watersheds and rehabilitate degraded catchment areas (Timau WRUA, 2023). Both groups engage in joint initiatives such as tree planting and watershed restoration, sharing overlapping mandates in sustainable livelihood promotion and conflict resolution (Ongugo et al., 2008; Timau WRUA, 2023).

Water Resource Authority (WRA)

The WRA (formerly Water Resource Management Authrity, or WRMA) serves as Kenya’s central regulatory institution for water resource management. It oversees water allocation, monitoring, and protection to balance agricultural productivity with ecological sustainability. Through the permit system, it regulates abstraction levels to maintain environmental flows while ensuring equitable access for users (GoK, 2016).

The WRA collaborates with WRUAs in implementing catchment management strategies, monitoring water withdrawals, and addressing local disputes. While the Authority establishes national-level rules and enforces compliance, WRUAs translate these regulations into local action, coordinating rotation schedules, identifying illegal abstractions, and encouraging conservation. WRA, in turn, provides oversight, technical assistance, and limited funding to strengthen grassroots governance. This interdependence reinforces a multi-level water governance system linking national policy frameworks with local water user initiatives (ibid).

Methodological Framework

The research approach consist of a multi-stage participatory process for Qualitative System Dynamics Modelling (QSDM) based on CLDs to support the analysis of interactions between the governance, socio-economic, natural resources and technological systems. The methodological step and methodological framework are shown in Figs. 3 and 4 respectively.

CLDs are a qualitative tool used in the framework of system dynamics to visualize the complex interactions and feedback loops within and between components of a system (Haraldsson, 2004), helping to describe the network of interconnections that influence the system´s dynamic evolution (Hanf et al., 2025; Coletta et al., 2024). CLDs illustrate cause-effect relationships between variables using arrows with positive or negative polarity—where positive links indicate that variables change in the same direction, whereas negative links indicate that variables change in the opposite direction. These links often form feedback loops, either reinforcing loops that amplify changes and can lead to exponential growth or decline, or balancing loops that counteract changes, promoting stability and equilibrium within the system. CLDs can also include time lags, represented by a double perpendicular bar on the arrow (Haraldsson, 2004).

Fig. 3

Methodological approach.

Successful application of a systems dynamics modelling approach involves five main steps, namely: i) problem articulation, ii) formulation of dynamic hypothesis and causal loop modelling, iii) formulation of quantitative simulation model, iv) testing and validation, and v) policy design and evaluation (Sterman, 2002). The objective of this study is to develop a holistic and multi-sectorial understanding of water scarcity dynamics and hindering factors for SPIS and RWH adoption, therefore it only includes the causal loop modelling phase.

The methodology is based on a participatory approach to better integrate different fields of knowledge. This approach is built considering the wide recognition of the inclusion of stakeholders and participatory practices for the development of Qualitative System Dynamics Models (QSDMs) (Noor et al., 2022), and it is aimed at providing a richer contribution in the understanding of the interactions between the elements of the system (Coletta et al., 2024). SES research acknowledges that systems are framed depending on the position of the observer (Armitage, 2011), and that the interconnected causes and sub-systems are understood through the promotion of multi-actor’s dialogue and the engagement of different sectors and different perceptions (Hanf et al., 2025).

Fig. 4

Methodological Framework.

Exploratory interviews respondents selection

The first phase of the research consisted in a literature review of governmental sources. Through consultation with local partners working in Academia and Non-Governmental Organizations, the target actors were shortlisted to save time and resources while allowing the comprehensiveness of the research output. Other actors were identified by partners if aimed relevant to the scope of the study. Further actors were included based on the first respondents input, following a snowball sampling. A total of 40 participants from CSOs, grass-root organizations and governmental agencies were included in the first round of data collection (the list of actors interviewed is shown in Table 1).

Table 1

List of Respondents.
Actors Typology	Semi-Structured Interviews	Semi-Structured Group Interviews	Open-ended Discussions
Water Users	Small Scale Farmers (2)
	Large Scale Farmers (2)
	Riparian Land Owners (1)
	Women´s Group Representatives (1)
Grass Root Organizations	CWPs (1)	Timau WRUA (6)
	CFA (1)	Ngusishi WRUA (5)
	Timau WRUA (1)	Isiolo WRUA (3)
NGOs		MKEWP (6)	MKEWP (2)
NGOs		SNV (1)
Consultancy (Private Sector)	Rural Focus Ltd (1)
Bilateral Research Organizations	CETRAD (1)
Corporate Bodies and Parastatal Organisations	Local WRA (1)	National WRA (2)
		NWHSA (1)
		NDMA (4)
County Government	Laikipia Water Department (1)
National Government	Water Department (1)
National Government	Irrigation Department (1)

Semi-Structured Interviews for Stakeholder Identification and Prioritization, Problem Specification, System Boundaries Definition and Identification of Preliminary Variables

Interviews were carried out with the selected actors between 08.10. 2024 and 25.10.2024.

All interviews were conducted in person, lasted between 30 and 120 minutes, and were audio-recorded with participant consent. The interviews were articulated under the following topics to guide the discussion:

Role of the organization and structure;

ii)

Main challenges faced by the actor, community and/or organization;

iii)

State of water resources and main drivers of resource degradation;

iv)

Ongoing initiatives to tackle mentioned challenges and hindering factors;

Collaborations and actors links.

Specific questions were adapted iteratively based on the position, role and sector of the respondents. This phase was carried out to refine the research problem, target the research area and system boundaries, and produce a preliminary set of variables to guide the individual CLD building phase. Interview transcripts were examined through a structured thematic extraction process. Salient terms, variables, and causal expressions mentioned by participants were identified, extrapolated and grouped into five domains (social, economic, governance, ecological, and technical). This process informed the identification of key system variables and relationships, which were subsequently used to inform and support the individual and thematic CLDs.

Second Round of Interviews and Individual CLDs elaboration

During the second phase for the construction of the Individual CLD models, participants where shortlisted based on their relevance in the system dynamics and network centrality. Each model was focused on the interviewee's understanding of the system dynamics within their area of expertise. The individual CLDs interviews took place between the 6th and 30th of May, 2025. In some instances, individual interviews were preferred to a single Focus Group Discussion (FGD) to facilitate the organization of the activity and to allow a pluralistic provision of input and reduce peer pressure, as the one-on-one interview format create an open environment to share individual´s views (Peerone et al., 2020). In the event of FGD, a local facilitator was present. The selected actors included:

Timau WRUA representatives;

ii)

Extension officers;

iii)

Members of Mount Kenya Ewaso Water Partnership;

iv)

A local WRA officer (done via phone interview);

Representatives of Laikipia County Government, from the Water, Irrigation and Agricultural Departments respectively;

vi)

A consultant from Rural Focus L.t.d.;

vii)

A representative from SNV;

viii)

Members of Ontulili CFA.

The process of individual CLD construction was structured following a systematic procedure., consisting of i) an introduction, description of the System Thinking approach and the CLD methodology; ii) open ended discussion for the selection of initial variable relevant to the problem and system under consideration; iii) collaborative CLD building; iv) evaluation of the CLD´s representativeness of the system.

During the open ended discussion, notes where took by the researcher and each statement was translated into quantitatively measurable variables. When the interview expressed a clear causal relationship, notes were taken to facilitate the next phase. In some cases, the researcher may enrich the discussion by directing questions to address the preliminary list of variables collected from the first interview round. The researcher and interviewees occasionally agreed on rephrasing some of the variables mentioned by the respondent to ensure consistency and integration with other individual models in the later stage.

Variables were written on sticky notes or stickers and then placed on a large paper sheet. The interviewees then arranged the variables and draw the causal links (arrows) between them. During this process, especially for respondents less familiar with the method, the researcher assisted the interviewee in drawing the causal links, often by stimulating reflection on causal relationships by providing question such as “Do you think that variables A and variable B are interconnected? If yes, why? Is this relationship bimodal, as variable A influences variable B and vice versa?” During the drawing of each link, the researcher asked for clarification to ensure clarity and appropriatedness. Finally, participants where asked if there was any significant delay or time lag between the dynamic behaviour of variables, requesting a further elaboration if the respondent´s answer was positive.

Digitalization of individual CLD and Merging in an Individual QSDM

The stakeholders´ diagrams were digitalized via Vensim 10.3.0, a software extensively used for system dynamics modelling (Noor et al., 2022). The digitalized models where then sent to each stakeholder via e-mail or Whatsapp for validation. Once the individual CLDs have been validated by respondents, they were joined together into an aggregate or "merged" STES model representing the perspectives and mental models of all individual stakeholders into one single holistic model, considering both complementary and controversial elements. The merging of the single CLD was initiated starting from the mental model with the greatest number of variables, and variables from the other stakeholders' models were added progressively. Following the methodology of Noor et al. (2022), complementary, redundant and controversial elements have been addressed in the merged CLD via remote stakeholders´ consultation (see Fig. 4).

After the merging of the CLDs into a single QSDM, the model was then downscaled to reduce the complexity and facilitate its readability, following the EEOR method (endogenisation, encapsulation and order-oriented reduction) (Asif et al., 2023). Sources indicate that holistic models developed by combining CLDs built by individual stakeholders can lead to over-complexity (Inam et al., 2015; Bureš, 2017; Noor et al., 2022; Asif et al., 2023), which can limit the role of stakeholders in the next modelling phases and lead to incorrect inferences, and therefore requires a process of simplification, or model downscaling (Noor et al., 2022, Bureš, 2017). It must be acknowledged that, during this process, some informational value is lost (Bureš, 2017; Asif et al., 2023). Therefore downscaling only addressed the variables that did not affect the model readability. Duplicate (ghost) variables were used only as inputs to other variables.

Thematic CLDs extraction

After finalizing the CLD, the large unified CLD was divided into several thematic sub-models. These sub-models addressed significant enforcing and balancing loops that were particularly relevant to the research questions. This division helps to assess the details of different aspects in isolation and simplifies the model structure (Noor et al., 2022). During this phase, the thematic models went through a refinement process, which involved: (i) the addition of variables emerging from cross-comparison of thematic diagrams and secondary literature; (ii) renaming and rephrasing of specific variables to ensure conceptual consistency across sub-models; and (iii) merging of multiple variables and feedbacks into a single variable.

This process carried out through an iterative model refinement was supported by literature review and by the stakeholders involved in the modelling phase (engaged remotely), ensuring that adjustments remained grounded in local insights.

Results

Problem definition

The central problem identified through the perspectives of the actors interviewed concerns the difficulty in achieving sustainable and equitable water management for agricultural production in the region. This challenge is rooted in a dynamic interplay of environmental, socio-economic, institutional, and technological factors that interact to create systemic barriers to effective water governance. Central to the problem is the unsustainable water abstraction practices, particularly illegal withdrawals that are difficult to monitor or regulate effectively. The capacity to enforce existing water use regulations remains weak, undermining formal governance structures and allowing harmful practices to persist. The anthropogenic pressures are further intensified by droughts and long-term climate variability. Intensive irrigation has been identified as a key factor in the onset of water conflicts in the basin, with small scale farmers being the main contributors to water abstraction in the basins (accounting for around 80% of the total water withdrawal) (Timau WRUA, 2023).

Although the WRUA has contributed to the resolution of disputes among its members, challenges remain in managing conflicts between water users (Timau WRUA, 2023; Dell´Angelo et al., 2016). These conflicts often involve farmers (from both upstream and downstream communities), pastoralists, forest association members and wildlife. While some large-scale commercial farms are present in the area, their pressure on the surface water resources has decreased drastically in the last years, due to the implementation of large scale water reservoirs recharged by rainfall through greenhouse rooftop conveyance systems, drilling of boreholes and adoption of water efficient irrigation practices and technologies. Contrarily, the majority of small-scale farmers practicing irrigation relies on flood irrigation practices, making their storage facilities inadequate to supply enough water during the dry season. Watershed degradation due to practices such as overgrazing and illegal deforestation further reduces water availability. Weak governance, characterized by lack of coordination, overlapping of functions between entities, limited human and financial resources for the WRUA and county government departments, and problems of corruption and political interference, further aggravate the issue.

Thematic CLDs

This section presents the results emerging from the participatory qualitative system dynamics modelling process carried out in the Timau sub-catchment. Insights from interviews and focus group discussions were synthesized into eight individual CLDs, which were subsequently merged and refined into a single, comprehensive Socio-Techno-Ecological System (STES) model (see supplementary material). The unified model was then disaggregated into five thematic CLDs to enhance readability and analytical depth: 1) Overuse and Scarcity, 2) Ecological Restoration, 3) Monitoring and Water Governance, 4) Enabling Conditions for Technological Transition, and 5) Opportunities of Solar-Powered Irrigation Systems (SPIS). Overuse and Scarcity (CLD 1) illustrates how illegal withdrawals and flat-rate water fees drive inequitable access and downstream shortages. Ecological Restoration (CLD 2) shows how catchment degradation, overgrazing, and poor land-use practices exacerbate hydrological instability, while community-based conservation efforts by WRUAs and CFAs can restore flows and reduce conflicts. Monitoring and Water Governance (CLD 3) highlights the crucial yet underfunded role of WRUAs in enforcement and sensitization, emphasizing the need for better financing and metering systems. Enabling Conditions for Technological Transition (CLD 4) demonstrates that cooperative models and access to affordable credit are central to overcoming financial barriers to technology adoption. Finally, Opportunities of SPIS (CLD 5) illustrates how solar-powered irrigation systems can lower energy costs and enhance water-use efficiency, though uptake remains limited by technical know-how and taxation. Together, these results point to the need for integrated interventions that align governance reform, ecological restoration, and technological innovation to foster resilience and equitable water management.

Overuse and Scarcity - CLD

Fig. 5

Thematic CLD 1 (Socio-economic variables marked in yellow, technological in orange, ecological in green and governance in purple).

The first thematic CLD (Fig. 5) illustrates how agricultural water withdrawals drive downstream scarcity, especially during dry seasons. As water availability diminishes, downstream users report to the WRUA, which issues a request to the WRA for water rationing. Once in place, the water rationing program temporarily limits upstream abstraction to ensure minimal downstream flow (balancing loop B1). Yet, its impact is weakened by illegal withdrawals conducted by i) seasonal farmers who lease land and abstract without a permit, and, ii) CWPs that applied for a water permit solely for human consumptions (as this is prioritized during rationing programs), but still withdraw water for irrigation practices.

Water abstraction is regulated by the water abstraction permit system, issued by the WRA following ongoing hydrological assessments, with a fixed amount of water abstraction allowed on a quarterly basis. The permit system is designed to make sure that each users activities are sustained throughout the year while guaranteeing an equitable allocation of water resources across all users along the catchment. If the cumulative water consumption reaches a certain threshold, no more permits are issued. But this governance mechanism did not take into account the illegal water withdrawals. These activities reduce the monitoring capacity of the WRA on actual water abstraction volumes, resulting in the issuing of more permits than the system can sustain.

Moreover, upstream farmers who initially obtained permits for specific water demands have progressively expanded their production, withdrawing more than authorized while other farmers were granted with new permits. As land parcels shrunk, intensification drove higher abstraction, further straining WRA’s monitoring capacity, which kept providing new permits (reinforcing loop R1). This vicious cycle would be balanced by the rationing programs, which limits the water withdrawal and therefore the amount of agricultural intensification (balancing loop B2). But the balancing effect of this loop is limited by the over-abstraction of permitted users, even during rationing programs.

Over-abstraction remains economically viable due to the historically low fines, set at 1 KSH (0.0077 USD) per m³, until their recent increase to 5 KSH in April 2025. In addition, CWPs charge members a flat rate, since only the common intake is monitored and household meters are absent. This system encourages excessive withdrawals and exacerbates disparities, even among upstream users subject to rationing. Because scarcity is not perceived, many farmers see little incentive to adopt efficient irrigation technologies or expand storage, especially given financial barriers and limited know-how. Those who comply with rules typically lack the resources to invest, resulting in continued over-irrigation and increasing upstream withdrawals (reinforcing loops R2 and R3).

Declining river flows undermine downstream livelihoods through recurrent crop failures, creating a “poverty trap” in which low incomes from poor yields restrict access to efficient irrigation technologies. As a result, farmers remain dependent on flood irrigation, further intensifying water scarcity (reinforcing loop R4), despite their willingness to invest in RWH and micro-irrigation technologies due to the recurrent droughts and dry spells. Scarcity heightens tensions among downstream users, with theft and vandalism becoming frequent responses to resource shortages. These conflicts reduce household income and elevate investment risks for both farmers and external actors, discouraging the adoption of improved practices and reinforcing cycles of scarcity and conflict (reinforcing loop R5). Water Scarcity heightens tensions, leading to theft, vandalism, and retaliation, which further reduce household income and elevate investment risks (reinforcing loop R5). Wildlife conflicts, especially with elephants seeking water, damage crops and infrastructure, amplifying vulnerability and discouraging technology adoption (reinforcing loop R6).

Ecological Restoration - CLD

Fig. 6

Thematic CLD 2 (Socio-economic variables marked in yellow, technological in orange, ecological in green and governance in purple).

The Thematic CLD 2 (Fig. 6) shows how river flow dynamics are closely linked to catchment health. Altered flows trigger both aquatic and terrestrial biodiversity decline, increasing flood events and erratic hydrological patterns (reinforcing loop R7). Riparian degradation from poor land allocation, agricultural expansion, and extensive flood irrigation further contributes to river´s pollution. Many CWP intakes lack filters, resulting in silted water that clogs irrigation systems, raising maintenance costs and discouraging adoption of drip systems. Continued use of flood irrigation further accelerates degradation (reinforcing loop R8). To counter this, WRUAs engage in riparian restoration through tree planting and awareness campaigns with landowners.

Downstream water scarcity drives pastoralist communities to move livestock upstream, searching water and grazing, often into forest areas. This results in overgrazing and direct use of springs, accelerating forest, wetland, and spring degradation and reducing base river flows. The ensuing catchment decline further destabilizes seasonal flows and intensifies dry-season scarcity, prompting additional pastoralist migrations and reinforcing forest and catchment degradation (reinforcing loop R9).

Forest and catchment decline is curbed by conservation and management activities of the Ontulili CFA, supported by the Timau WRUA. These include tree planting, monitoring of forest and riparian areas, and regulating livestock access to forests. These activities, funded by community enterprises authorized by the Kenya Forest Service - such as controlled farming, livestock rearing, and beekeeping - provide income that supports conservation but also exert pressure on water resources (reinforcing loop R10). The revenues finance forest monitoring, nursery management, and fire control, reducing illegal logging, charcoal burning, and wildfires. Improved catchment conditions stabilize environmental flows, reduce human–wildlife conflicts, and sustain downstream water availability (reinforcing loops R11, R12, R13, R14 and R15). The resulting benefits strengthen community engagement and financing for conservation (reinforcing loop R16), but also drive expansion of commercial activities (reinforcing loop R17). Restoration reduces water siltation, facilitating adoption of drip irrigation that enhances water-use efficiency and river flows, further improving catchment health and curbing water siltation (reinforcing loop R18).

Yet, forest commercial activities are undermined by illegal farming, often carried out by settlers operating with impunity due to corruption and weak governance capacity. These practices trigger land allocation conflicts and restrict the activities of registered CFA members, reducing community engagement in conservation. The decline in monitoring and management—already constrained by limited skills—further enables illegal abstraction, logging, honey harvesting, poaching, and charcoal burning, which frequently lead to wildfires and accelerate forest degradation.

Monitoring and Water Governance - CLD

Fig. 7

Thematic CLD 3 (Socio-economic variables marked in yellow, technological in orange, ecological in green and governance in purple).

WRUA activities, such as monitoring, catchment restoration, conflict resolution, and sensitization, are shaped by water availability and the frequency of water conflicts (CLD-3, Fig. 7). These functions are financed mainly through registration (one-time) and membership (annual) fees, which increase with population growth and expand WRUA’s spending capacity. Membership is voluntary, yet sensitization campaigns, typically via public meetings and word of mouth, encourage participation and generate additional revenues (reinforcing loop R19). As membership grows, peer-to-peer dialogue further amplifies sensitization, creating a reinforcing cycle of recruitment and awareness (reinforcing loop R20).

Sensitization campaigns also advocate for more efficient water use, including planned irrigation schedules, soil and water conservation, cultivation of drought-resistant crops, and adoption of climate-smart technologies. With more sensitization campaign, the water overuse reduces, while the number of farmers exposed to the benefits of efficient irrigation technology increases. The resulting reduction in withdrawals increases flows across the catchment, improving availability and easing conflicts. Over time, however, declining water-related conflicts reduce the intensity of WRUA activities and, consequently, the need for continued sensitization (balancing loops B3 and B4).

But membership and subscription fees alone fail to cover most WRUA conservation and management activities. Although monitoring and enforcement of unlawful abstraction is formally the duty of WRA, it can be delegated to WRUAs and financed through a portion of water fees. In practice, WRUA scouts engage in monitoring but receive no dedicated funding. Illegal abstractors often using diesel pumps near riparian areas face heavy penalties under WRA law, including equipment seizure, fines up to 1 million KSH, or imprisonment. By contrast, WRUA managers impose much smaller fines, which bolster their financial capacity (reinforcing loop R21) but also increase resentment and water entitlement among offenders, making them more resistant to WRA enforcement on one side, and more prone to over-abstract on the other (as they perceive the fine as a payment that allows them to do so). Given WRA’s limited security staff, laws against illegal abstraction cannot be enforced. Thus, while WRUA monitoring supports WRA’s capacity (balancing loop B5), the system of fines paradoxically fuels entitlement and further illegal withdrawals, worsening flows, scarcity, and conflicts, which in turn generate more WRUA activity and fines (reinforcing loop R22).

As noted in the R1 and R2 loops, water overuse in CWPs stems from weak monitoring. While individual abstractors must install meters, CWPs typically operate with a single intake meter, charging members a flat fee that encourages overuse. Household meters could address this, but their uptake is limited because they are not legally compulsory in rural areas and are further discouraged by poor water quality, as siltation clogs devices and raises maintenance costs. WRUA and CFA restoration efforts improve riparian health and water quality, making meter adoption more feasible. Wider use of household meters would reduce flat-fee systems, limit waste, and support adoption of climate-smart technologies. In turn, increased efficiency would boost river flows, improve availability, and reduce conflicts, lowering the need for WRUA engagement in restoration and conflict resolution (balancing loops B6 and B7).

If the WRUA is provided with sufficient financial capacity, it can initiate the construction of a common water infrastructure (supervised by the WRA), consisting of a water intake located upstream, a treatment station and a distribution network. This infrastructure can optimize the provision of water to all the CWPs throughout the catchment. Water allocation to each CWP is more easily monitored, improving equitable water access and improving river flows (balancing loop B8). Also, more household water meters can be installed due to better water quality, reducing overall water wastes and improving water flows (balancing loop B9). Such infrastructure would reduce the adoption of portable water pumps, often utilized by illegal abstractors, furtherly improving water use monitoring (balancing loop B10) and reducing the pressure on surface water resources (balancing loop B11). As water becomes more accessible for people downstream, conflicts subside and the rationing programs reduce in frequency, together with the need for the WRUA to engage in catchment protection activities.

Enabling conditions for Technological Transition - CLD

Fig. 8

Thematic CLD 4 (Socio-economic variables marked in yellow, technological in orange, ecological in green and governance in purple)

Thematic CLD 4 (Fig. 8) highlights the factors that shape the transition to efficient irrigation technologies, displaying how access to finance, technical knowledge, and social cooperation are the most significant enablers of adoption.

Increased upstream production raises farmers’ incomes and improves their ability to invest in water infrastructure. Under ideal conditions, revenues from intensified farming would provide the assets and financial means - often through loans - to adopt precision irrigation practices such as sprinklers and drip systems, construct RWH infrastructure (e.g., gutters, pipelines, and ponds), or expand storage with additional tanks. Adoption of drip lines and sprinklers reduces overuse, lowering agricultural water demand and overall withdrawals (balancing loops B12 and B13). Similarly, SPIS can limit over-abstraction, as solar pumps have lower flow rates than diesel pumps and are often equipped with meters to monitor use (balancing loop B14), while RWH infrastructure and reservoirs further reduce reliance on surface water (balancing loop B15). However, small farm sizes constrain space for storage structures, and as production expands, available land for reservoirs declines, increasing dependence on river withdrawals (reinforcing loop R24).

While sensitization conducted by WRUAs and NGOs plays a crucial role in shaping perceptions, the adoption of climate-smart technologies such as RWH, SPIS, sprinklers, and drip lines is primarily hindered by low farmer incomes and structural financial barriers. Most smallholder farmers sell their produce at the farm gate, lacking the market linkages necessary for increasing their net profits from their practices. At the same time, financial service providers generally offer loans at high interest rates and require collateral that smallholders cannot provide, limiting their access to credit. As a result, small-scale farmers remain trapped in a cycle of low revenues and low investment capacity.

But market access is often facilitated through farmers’ self-help groups, which enable knowledge exchange and product bulking. However, these groups rarely have the capacity to engage with larger markets. By contrast, organizing small-scale farmers into cooperatives (Saving and Credit Cooperative Organizations, or SACCOs) provides significant advantages, including stronger bargaining power, economies of scale, and improved access to credit. Cooperatives can serve as guarantors for loans, offering lenders repayment security, while collective bulking opens market channels to larger buyers and increases revenues. Higher revenues, in turn, enable financial resource pooling that supports investments in transport, storage, and value addition facilities. Expanded market access raises farmers’ profit margins, and additional income is reinvested into improving market engagement through infrastructure and services (reinforcing loop R26). As incomes grow, farmers acquire assets that can serve as collateral for further investments in efficient water infrastructure and irrigation systems. These investments, together with increased income and assets, generate successful examples that encourage peer adoption, stimulate the creation of new farmer groups, and expand cooperative membership (reinforcing loop R27).

The main obstacle to this virtuous cycle is the scarcity of cooperative models in the region that enable financial resource pooling for small-scale farmers. One exception is the EMU-SACCO, a community credit initiative launched in 2018 by Fauna and Flora International and managed by the Mount Kenya–Ewaso Water Partnership (MKEWP). Beyond providing a platform for knowledge exchange and collective price negotiation with large buyers, this cooperative has established a “revolving fund,” financed through membership and registration fees, to support water-related investments among its members. The model functions by having farmers sell their produce directly to the SACCO, which then markets the bulk product to larger buyers at negotiated prices favourable to members. In turn, the SACCO offers loans at low interest rates, with repayment partially deducted automatically from farmers’ sales revenues, thereby reducing both financial risk and repayment burdens.

This system, however, can only succeed if scaled up to reach enough farmers, thereby expanding the cooperative’s financial capacity to offer low-interest loans. As more farmers witness the benefits experienced by their peers, additional farmer groups are formed, join the SACCO, and contribute memberships that are channelled into the revolving fund. This process enables more farmers to access credit for investments in efficient irrigation technologies and RWH infrastructure, which in turn enhances production, increases revenues, and generates further success stories that reinforce adoption (reinforcing loops R28 and R29).As more farmers access to drip and sprinklers, SPIS and RWH, the less is the amount of agricultural areas supplied with flood irrigation, less water is overused, and agricultural water demand and water abstraction decrease (reinforcing loop B16).

To support the upscaling of this strategy, WRUAs, civil society organizations, and NGOs promote awareness through demonstration farms, where extension officers and community members provide training on technology use, installation, operation, and maintenance, while also encouraging the formation of farmer groups and SACCO membership. These initiatives are often initiated by farmer groups themselves, creating a reinforcing dynamic in which more groups lead to more demo farms, and vice versa (reinforcing loop R30). Demonstration farms also build technical knowledge, which increases farmers’ willingness to invest and lowers perceived risks, including costs related to poor installation or maintenance. As know-how spreads, further investments and success cases emerge, reinforcing exposure and motivating the creation of additional farmer groups that, in turn, establish more demo plots (reinforcing loop R31).

The opportunities of SPIS - CLD

Fig. 9

Thematic CLD 5 (Socio-economic variables marked in yellow, technological in orange, ecological in green and governance in purple)

The adoption of SPIS and solar pumps (CLD-5, Fig. 9), alongside drip lines and sprinkler systems, plays a central role in the region’s technological transition by substantially reducing energy costs, which many farmers find prohibitively high when relying on the public power grid. By replacing fossil fuels and grid electricity, SPIS lowers the operation and maintenance costs of climate-smart irrigation systems, while also cutting household energy expenses through photovoltaic use in domestic activities. Such investments can facilitate the adoption, upkeep, and improvement of farm irrigation systems, reservoirs, and RWH infrastructure by reducing off-farm costs (reinforcing loop R32). Given the strong willingness to invest in photovoltaic solutions, many households pursue solar installations until they achieve full energy independence (balancing loop B17).

In addition, the lower flow rate of solar pumps and embedded water meters (which are a component of most solar pumps in the market) reduces hourly water withdrawals, limiting over-abstraction and thereby alleviating scarcity and water user conflicts. Fewer conflicts translate into fewer acts of vandalism and theft, which lowers investment risks (reinforcing loops R33 and R34). As water availability improves, elephants are less likely to enter farms in search of water, reducing damage to crops and irrigation infrastructure (reinforcing loop R35. Increased investment further drives adoption of solar pumps, reinforcing reductions in water-withdrawals. At the same time, access to solar energy enables the use of electrified fences and night lighting, which can deter both wildlife incursions and theft (reinforcing loops R36 and R37). However, widespread adoption of photovoltaic systems and SPIS remains constrained by limited technical know-how and, above all, by financial barriers. These challenges are further exacerbated by government taxation on solar products, which raises initial costs and therefore reduces accessibility.

Discussion

This study reveals that the dynamics of water governance, technological transformation, and ecological feedbacks in the Mount Kenya region are deeply intertwined within a complex adaptive system.

The participatory system dynamics approach, operationalized through CLDs, demonstrates that both resource scarcity and the sociotechnical conditions that shape farmer decisions emerge from mutually reinforcing feedback loops that link human behaviour, institutional incentives, environmental variability and ecological processes. From a theoretical standpoint, this analysis advances the Social-Technical-Ecological System (STES) perspective, which conceptualizes technology as both a mediator and product of social–ecological interactions (Ahlborg et al., 2019; Wei et al., 2018).

The Objectives Misalignment

At the heart of these interactions lies a persistent misalignment between the institutional, infrastructural, and cognitive dimensions of the local irrigation regime. This partially mirrors the “service regime” misalignments identified by Wainaina (2021), who argued that transitions toward drip irrigation in Kenya are constrained by incoherence across five key regime dimensions: infrastructure, organizational mode, rationale, social interaction, and temporal-spatial configurations. Rainwater harvesting and SPIS, together with drip systems, are promoted as efficiency solutions, yet the social and institutional capabilities needed to sustain them lag far behind. Moreover, flat-rate water tariffs, weak enforcement and lack of subsidies and economic incentives eventually end up reinforcing the gap between farmers and policymakers objectives (i.e. profit maximization and environmental protection, respectively) (Wainaina, 2021).

As central argument established by Pannell et al. (2006) on rural innovations, adoption occurs when the landholder perceives that the innovation will enhance the achievement of their personal goals. This expectation depends on three broad sets of issues: the learning process, the landholder's characteristics, and the characteristics of the practice itself. While this study highlights the importance of extension services and building know-how through initiatives such as demo farms to facilitate the learning process of a significantly complex innovation (CLD-4), the most crucial factor determining adoption in the long run is the “Relative Advantage”. This the degree to which an innovation is perceived as superior to the practice it supersedes, particularly in economic terms (Pannell et al., 2006). Conservation practices such as RWH, micro irrigation systems and SPIS, are characterized by high up-front costs and benefits that occur in the future, reducing their attractiveness. The Timau case shows how governance can severely reduce the perceived relative advantage encouraging excessive withdrawals (overuse) and dis-incentivizing upstream farmers from investing in efficient technologies like RWH or micro-irrigation.

Moreover, the government taxation and import duties on solar technology components reduces small scale farmers´ capacity to access to SPIS even further. Over recent years in Kenya, some exemptions on VAT or import duty for stand-alone solar (SAS) products were removed, making solar equipment 10–24% more expensive and reducing sales by up to 20% (ACE TAF., 2019). These fiscal pressures effectively shift an otherwise promising technology toward a higher-cost or less feasible category for many farmers. Meanwhile, inconsistent application of tax exemptions, bureaucratic delays, and customs classification ambiguity further increase transaction costs for firms and users (ibid). The reinstatement of VAT/import-duty exemptions for solar and renewable energy equipment in some Kenyan policies suggests acknowledgment of this barrier, but lingering institutional uncertainty and revenue pressures may slow further reform (Githinji, 2021). This situation typifies the “multi-level tension” discussed in the sustainability transitions literature (Smith & Stirling, 2010; Geels, 2002), where niche innovations like SPIS are constrained by incumbent policy structures.

Institutional Lock-in

The over reliance on the permit system to monitor water use cost-effectively and the flat-rate water fee to overcome the socio-economic (costs and acceptance), institutional (reshaping the management processes) and technical (high siltation of water) barriers associated to household water meters, generates what Smith and Stirling (2010) describe as “institutional lock-in,” where the very governance structures intended to manage resources reinforce unsustainable practices.

Such lock-in perpetuates a CPR (common pool resource) system characterized by lack of congruence between rules and local conditions, both social-ecological and technical. As Ostrom argues (1990), this is a condition essential for the successful management of common pool resources. Moreover, low excludability capacity (especially of seasonal appropriators), no effective monitoring due to poor financing (both of the WRUA and the WRA) and lack of a graduated sanctioning scheme (for both illegal water withdrawal and over-abstraction) contribute in driving the common resource system to failure (Ostrom, 1990). This was also observed by Dell´Angelo et al. (2016) in their analysis of Mount Kenya’s CWPs, where, despite the clear progress made towards a more participatory governance model, the CWPs under study displayed clear signals of poor consistency with Ostrom´s design principles for successful Community Based Governance of CPRs, and low level of adaptation to change. This study show how this inconsistency is reproduced also at the WRUA and WRA level.

Reactive Adaptation

The CLDs also show how ecological feedbacks amplify institutional weaknesses within the local context. When rivers experience low flows, WRUAs often intensify sensitization activities, urging farmers to ration water, adopt conservation measures, or invest in efficient irrigation. In these periods of visible scarcity, farmers become more aware of the need for alternative technologies and are more receptive to adopting climate-smart options such as RWH or drip irrigation. Conversely, during years of relative abundance, WRUA campaigns tend to decline, and farmer incentives to adopt efficiency-enhancing technologies weaken. Periods of drought trigger short-lived cooperation and emergency rationing, yet these responses dissipate once rainfall returns. This dynamic underscores how ecological variability directly shapes the institutional pressure of WRUAs and, in turn, affects the pace of technological transition. Such episodic mobilization, followed by a return to the status quo, reflects a limited adaptive capacity, or a “reactive adaptation” (Folke et al., 2005). In the Mount Kenya context, adaptation is largely reactive, sustaining short-term stability at the expense of long-term transformation. The system therefore oscillates between crisis and complacency, never reaching the structural reorganization that would ensure resilience. This highlights the need for sustained sensitization and support mechanisms that go beyond immediate scarcity signals, ensuring that adoption decisions are not entirely contingent on short-term hydrological fluctuations.

Ecological feedbacks on adoption decisions

Low income farmers tend to be much more sensitive to risk and incline to short term survival strategies (with a short management horizon), and to not adopt innovation that are perceived to increase risk (Pannell et al., 2006). Moreover, crop losses due to droughts or erratic hydrological patterns foster perceptions of fragility that discourage investment in infrastructure. The relative advantage of an innovative land use is reduced if it is perceived to be more subject to various risks compared to the current land use, including establishment failures (ibid). The high frequency of thefts and vandalism acts during drought periods on one side, and wildlife incursion on the other, elevate investment concerns, thereby reducing the perceived relative advantage of investments, in the absence of insurance schemes. But also drought, water scarcity and recurrence of crop failures play a major role in shaping vulnerability, and therefore risk aversion attitudes. Without mechanisms to buffer against these risks, short-term survival strategies often take precedence over longer-term efficiency, but innovative insurance mechanisms tailored to smallholders could help break this barrier (Menapace et al., 2013). In this sense, instruments such as Index-based crop insurance (triggered by rainfall or water availability thresholds) has been piloted in parts of East Africa with promising results (Ntukamazina et al., 2017).

Ecological processes also affect the compatibility of both the innovation (specifically the drip line component) and monitoring technology (household water meters) within the region. Compatibility refers to the innovation’s fit with a landholder’s existing set of technologies, practices, and resources (Kaine and Lees 1994). A technology that is readily adoptable by a farmer requires to fit the existing technological components, environmental conditions and practices, besides existing beliefs and values (Pannell et al., 2006). Similarly to Wainaina´s (2021) and Kanda & Lutta (2022) findings, this study shows how poor water quality indeed affects drip lines adoption. In order to make the technologies compatible with the resource in use, it would be necessary to install high quality filters. This would affect the innovation´s complexity, and, as Roger (2003) argues, the efforts required for ongoing management, and the risk for the innovation to fail any given year. Complexity ultimately reduces significantly the innovation´s relative advantage (Rogers, 2003) and willingness to invest. Moreover, Farmers who repeatedly encounter failures from neighbours grow skeptical of innovation, slowing diffusion further. Same process applies to household water meters as well. Yet, evidence show that adoption of (smart) water meters indeed significantly improves CWPs performance, especially in low-resource, geographically remote areas (Molu & Luketero, 2025).

In this sense, as displayed in CLD-2, CFA efforts (jointly with the WRUA) to restore the catchment through tree planting, forest conservation and wetland protection represents an important piece in reducing the innovation´s complexity and compatibility by acting on the resource system itself. Yet, as mentioned by Koech et al. (2009), CFAs members need to attain a reasonable level of socio-economic development through the use of forest resources for their basic needs. Land tenure conflicts represent a serious barrier towards their objective, and the lack of benefits eventually results in community dis-engagement in management and conservation practices. But also internal conflict of interests can arise within the forest communities itself, where conservation can clash with exploitation, and existing capacities fail to meet necessary capacities (Koech, 2009).

SACCOs

SACCOs are promising collective-action mechanisms to pool capital, reduce risk, and facilitate technology uptake. Indeed, savings and credit cooperatives in Kenya have mobilized substantial capital and offer lower-cost loans to members compared to conventional lenders (Olando et al., 2013) However, SACCOs are not immune to weaknesses, particularly free-rider problems and limited capital contributions from poorer members (Anania & Gikuri, 2015).

In collective action theory, free-riding occurs when individuals enjoy the benefits of a public good or service without contributing proportionately (i.e., some farmers might benefit from cooperative investments without participating or contributing capital) (Olson, 1971). If many members lack liquidity to contribute, they may opt out or under-contribute while still benefiting from the collective infrastructure, thereby weakening incentives for wealthier or more active members to invest. Empirical studies of Kenyan and East African cooperatives often observe that low-income members contribute less, leadership challenges and coordination costs impede performance, and some members drop out when returns or services are delayed (Anania & Gikuri, 2015). This dynamic can erode trust, discourage investment, and reduce overall participation, particularly among those with capital, who may see disproportionate risk or burden (Ibid). Therefore, while SACCOs can be a strong enabling mechanism, their design must internalize incentives, accountability, and mechanisms to limit free-riding, such as minimum contributions, monitoring, differential benefits, or membership rules.

Enabling conditions and leverage points

Despite these constraints, the analysis points toward promising leverage points. Strengthening SACCOs and cooperative arrangements remains central: if well designed, they can pool resources, improve access to markets, and mitigate individual risk. Positive feedback loops generated by demonstration farms, peer-to-peer learning, and cooperative-led investments can convert scarcity-driven reinforcing loops into virtuous cycles of innovation and resilience.

At the grassroots level, empowering WRUAs and CFAs with funding and enforcement authority can improve monitoring, reduce illegal activities, create capacity and strengthen conservation. The Water Act 2016 does envisage directing a portion of water-fee revenues toward WRUA activities (GoK, 2016), but this is rarely operationalized in practice. Strengthening legal and financial pathways to ensure that funding reaches WRUAs is essential. On the farm level, on one hand, scaling extension services, demonstration plots, and capacity training remains vital to reduce technical risk and build farmer confidence. On governance, on the other, shifting from flat-rate to volumetric pricing (with household metering and appropriate safeguards) could align incentives with efficiency and equity. The construction of a public water infrastructure for the provision of rural households can facilitate the operationalization of the water meters, besides making the common resource more easily excludable and compatible with the drip technology through filtering. Yet its establishment remains hindered by low financial resources and political interference, as many leaders have interests in keeping the system under the following conditions due to conflicts of interest.

Finally, while RWH at the farm level is widely recognized as a promising adaptation strategy (Oguge & Oremo, 2017; Mati et al., 2022; Barron & Salas, 2009) this study highlights that its potential is constrained by physical and spatial limits. Smallholder farms often lack adequate space for large ponds or tanks, and the storage that can be constructed depletes quickly during extended dry spells. As a result, individual RWH systems provide only temporary relief and cannot by themselves ensure water security throughout the season. To overcome this limitation, investment in larger-scale and collective storage infrastructure is necessary. Strengthening and facilitating initiatives led by CWPs could play a key role in complementing farm-level systems, ensuring that water captured during wet periods is available over longer dry seasons. Linking farm-scale RWH with collective infrastructure would enhance resilience, reduce pressure on rivers during low-flow periods, and support the broader transition toward sustainable irrigation practices.

Conclusions

This paper advances a Socio-Techno-Ecological Systems (STES) perspective that integrates technology as a core driver within socio-economic, water governance and ecological dynamics, challenging the traditional separation of socio-technical and socio-ecological research.

Through participatory CLDs, the study demonstrates how stakeholder-informed modelling can capture context-specific feedbacks while retaining analytical rigor, offering a methodological contribution to qualitative system dynamics. For policy, the findings highlight that promoting climate-smart irrigation technologies requires more than technical dissemination. Adoption depends on institutional change, particularly strengthening WRUAs with financial and enforcement capacity, scaling cooperative and credit mechanisms, and aligning incentives to reduce over-abstraction. Addressing these governance and financial barriers is essential to transform reinforcing cycles of scarcity, conflict, and dis-adoption into virtuous loops of innovation, conservation, and resilience. By linking scientific frameworks with actionable policy levers, this research provides a pathway toward more equitable and sustainable water governance in smallholder farming systems. Nevertheless, the case study presents some limitations: it focuses only on small-scale farming and surface water resources, and omits groundwater use, large-scale farming, and agronomic innovations. The model is a representation of stakeholders´ perceptions (not predictive), and the sample size might not fully represent the heterogeneity of actors operating in the catchment. Future research should include a broader range of stakeholders, while also comparing experiences across catchments with different conditions and exploring intra-household decision making. Finally, a pilot cooperative-insurance hybrid should be designed to its effectiveness in enabling technology adoption among vulnerable farmers.

Author Contribution

A.C. designed the study, conducted fieldwork and interviews, developed the CLDs, performed the analysis, and wrote the manuscript. M.C.B., S.L.S., G.U., B.M.M., and S.S. contributed to study design, provided methodological and contextual guidance, supported interpretation of findings, and critically revised the manuscript. All authors approved the final version.

Funding Statement

The research leading to this results received funding from the Elsa Neumann Stiftung (NaFöG Stipendium) under the Grant Agreement N° H77005.

Conflicts of interest

All authors certify that they have no affiliation with or any involvement in any organization or entity with any financial interest or non-financial interest in the subject matter or materials discussed in this manuscript. The authors have no conflicts of interest to declare that are relevant to the content of this article.

Ethical approval

and accordance

The research protocol was reviewed and approved by the National Commission for Science, Technology and Innovation (NACOSTI) in Kenya (Research Permit No.

692880) in accordance with the Science, Technology and Innovation (STI) Act, 2013, and National Scientific and Ethics Committee (NSEC) guidelines.

Consent to participate

Informed consent was obtained from all individual participants included in the study.

Both verbal and formal informed consent (by signing a consent form) was obtained prior to the interview. All participants were adults, and participation to the research was entirely voluntary.

Consent to publish

The authors affirm that human research participants provided informed consent for publication of the findings elaborated through interview and participatory modelling rounds. No identifying information are presented in the study.

Data Availability

The qualitative interview datasets generated and analysed during the current study are not publicly available due to confidentiality commitments under the NACOSTI permit and consent agreements. Anonymised versions of the transcripts and individual CLDs may be made available from the corresponding author upon reasonable request.

Acknowledgement

The authors gratefully acknowledge the support and collaboration of the Timau WRUA, the participating community members, and the local civil-society and institutional stakeholders who generously contributed their time and insights throughout the interviews, participatory modelling sessions and validations. Appreciation is extended to MKEWP, the field assistants and facilitators who supported data collection, and to colleagues who provided methodological guidance and constructive feedback during the development of this study.

Reference

Adebiyi, J. A., & Olabisi, L. S. (2022). Participatory causal loop mapping of the adoption of organic farming in Nigeria. Environmental management, 69(2), 410–428. DOI:10.1007/s00267-021-01580-w

Agnew, S., Smith, C., & Dargusch, P. (2018). Causal loop modelling of residential solar and battery adoption dynamics: A case study of Queensland, Australia. Journal of cleaner production, 172, 2363–2373. DOI:10.1016/j.jclepro.2017.11.174

Ahlborg, H., Ruiz-Mercado, I., Molander, S., & Masera, O. (2019). Bringing technology into social-ecological systems research—motivations for a socio-technical-ecological systems approach. Sustainability, 11(7), 2009. https://doi.org/10.3390/su11072009

Alamanos, A., Xenarios, S., Assubayeva, A., Landis, C. F. M., Dellis, K., & Koundouri, P. (2025). Systems-thinking innovations for water security. Frontiers in Water, 6, 1492698. DOI:10.3389/frwa.2024.1492698

Alemu, F., Kumie, A., Medhin, G., Gebre, T., & Godfrey, P. (2017). A socio-ecological analysis of barriers to the adoption, sustainablity and consistent use of sanitation facilities in rural Ethiopia. BMC public health, 17(1), 706. DOI: 10.1186/s12889-017-4717-6

Aliyu, M., Hassan, G., Said, S. A., Siddiqui, M. U., Alawami, A. T., & Elamin, I. M. (2018). A review of solar-powered water pumping systems. Renewable and Sustainable Energy Reviews, 87, 61–76. https://doi.org/10.1016/j.rser.2018.02.010

Anania, P., & Gikuri, A. (2015). SACCOS and members’ expectations: Factors affecting SACCOS capacity to meet members’ expectations.

Anderies, J. M., Janssen, M. A., & Ostrom, E. (2004). A framework to analyze the robustness of social-ecological systems from an institutional perspective. Ecology and society, 9(1). https://doi.org/10.5751/ES-00610-090118

Andersson, J., Lennerfors, T. T., & Fornstedt, H. (2024). Towards a socio-techno-ecological approach to sustainability transitions. Environmental Innovation and Societal Transitions, 51, 100846. DOI: 10.1016/j.eist.2024.100846

Andersson, J., Plummer, P., Lennerfors, T. T., & Hedberg, C. (2024). Socio-techno-ecological transition dynamics in the re-territorialization of food production: the case of wild berries in Sweden. Sustainability Science, 1–16. DOI:10.1007/s11625-023-01461-7

Arthur, B.W. The Nature of Technology. What It Is and How It Evolves; Free Press: New York, NY, USA, 2009. DOI:10.1016/j.futures.2010.08.015

Asif, M., Inam, A., Adamowski, J., Shoaib, M., Tariq, H., Ahmad, S., ... & Nazeer, A. (2023). Development of methods for the simplification of complex group built causal loop diagrams: A case study of the Rechna doab. Ecological Modelling, 476, 110192. https://doi.org/10.1016/j.ecolmodel.2022.110192

Barron, J., & Salas, J. C. (2009). Rainwater harvesting: a lifeline for human well-being. UNEP/Earthprint.

Berkes, F., Colding, J., & Folke, C. (Eds.). (2008). Navigating social-ecological systems: building resilience for complexity and change. Cambridge university press. DOI:10.1017/CBO9780511541957.020

Bouchet, L., Thoms, M. C., & Parsons, M. (2022). Using causal loop diagrams to conceptualize groundwater as a social-ecological system. Frontiers in Environmental Science, 10, 836206. DOI:10.3389/fenvs.2022.836206

Budiman, I., Wagemakers, A., Derks, M., Sahara, & Moore, S. (2024). Socio-environmental systems in technology adoption in animal husbandry in South-East Asia: A framework synthesis approach. NJAS: Impact in Agricultural and Life Sciences, 96(1), 2349518. https://doi.org/10.1080/27685241.2024.2349518

Bureš, V. (2017). A method for simplification of complex group causal loop diagrams based on endogenisation, encapsulation and order-oriented reduction. Systems, 5(3), 46. DOI:10.3390/systems5030046

Coletta, V. R., Pagano, A., Pluchinotta, I., Zimmermann, N., Davies, M., Butler, A., ... & Giordano, R. (2024). Participatory causal loop diagrams building for supporting decision-makers integrating flood risk management in an urban regeneration process. Earth's Future, 12(1), e2023EF003659. DOI:10.1029/2023EF003659

Cornwall, A., & Jewkes, R. (1995). What is participatory research?. Social science & medicine, 41(12), 1667–1676. https://doi.org/10.1016/0277-9536(95)00127-S

Dell’Angelo, J., McCord, P. F., Gower, D., Carpenter, S., Caylor, K. K., & Evans, T. P. (2016). Community water governance on Mount Kenya: an assessment based on Ostrom’s design principles of natural resource management. Mountain Research and Development, 36(1), 102–115. DOI:10.1659/MRD-JOURNAL-D-15-00040.1

Elsawah, S., Pierce, S. A., Hamilton, S. H., Van Delden, H., Haase, D., Elmahdi, A., & Jakeman, A. J. (2017). An overview of the system dynamics process for integrated modelling of socio-ecological systems: Lessons on good modelling practice from five case studies. Environmental Modelling & Software, 93, 127–145. http://dx.doi.org/10.1016/j.envsoft.2017.03.001

Filatova, T., Polhill, J. G., & Van Ewijk, S. (2016). Regime shifts in coupled socio-environmental systems: Review of modelling challenges and approaches. Environmental modelling & software, 75, 333–347. https://doi.org/10.1016/j.envsoft.2015.04.003

Folke, C., T. Hahn, P. Olsson, and J. Norberg. 2005. “Adaptive Governance of Social–Ecological Systems.” Annual Review of Environment and Resources 30, no. 1: 441–473. DOI:10.1146/annurev.energy.30.050504.144511

Gichuki, F. N. (2002). Water scarcity and conflicts: A case study of the Upper Ewaso Ng’iro North Basin. The changing face of irrigation in Kenya: Opportunities for anticipating change in eastern and southern Africa, 113 − 34.

Grupper, M. A., Horne, A. C., Webb, J. A., & Olden, J. (2025). Identifying and Approaching Barriers to Environmental Flow Implementation Using Social–Ecological Systems Thinking. Wiley Interdisciplinary Reviews: Water, 12(1), e1764. DOI:10.1002/wat2.1764

Hanf, F. S., Ament, F., Boettcher, M., Burgemeister, F., Gaslikova, L., Hoffmann, P., ... & Wickel, M. (2025). Towards a socio-ecological system understanding of urban flood risk and barriers to climate change adaptation using causal loop diagrams. International Journal of Urban Sustainable Development, 17(1), 69–102. https://doi.org/10.1080/19463138.2025.2474399

Haraldsson, H. V. (2004). Introduction to system thinking and causal loop diagrams (pp. 3–4). Lund, Sweden: Department of chemical engineering, Lund University.

Hardin, G. (1968). The tragedy of the commons: the population problem has no technical solution; it requires a fundamental extension in morality. science, 162(3859), 1243–1248.

Helmrich, A., Kuhn, A., Roque, A., Santibanez, A., Kim, Y., Grimm, N. B., & Chester, M. (2023). Interdependence of social-ecological-technological systems in Phoenix, Arizona: Consequences of an extreme precipitation event. Journal of Infrastructure Preservation and Resilience, 4(1), 19. doi: 10.1186/s43065-023-00085-6

Hossain, M. S., Ramirez, J., Szabo, S., Eigenbrod, F., Johnson, F. A., Speranza, C. I., & Dearing, J. A. (2020). Participatory modelling for conceptualizing social-ecological system dynamics in the Bangladesh delta. Regional environmental change, 20, 1–14. DOI:10.1007/s10113-020-01599-5

Inam, A., Adamowski, J., Halbe, J., & Prasher, S. (2015). Using causal loop diagrams for the initialization of stakeholder engagement in soil salinity management in agricultural watersheds in developing countries: A case study in the Rechna Doab watershed, Pakistan. Journal of environmental management, 152, 251–267. https://doi.org/10.1016/j.jenvman.2015.01.052

Kaine GW, Lees JW (1994) Patterns in innovation: An analysis of the adoption of practices in beef cattle breeding. Occasional paper no. 190. Rural Development Centre, University of New England, Armidale.

Kimwatu, D. M., Mundia, C. N., & Makokha, G. O. (2021). Monitoring environmental water stress in the Upper Ewaso Ngiro river basin, Kenya. Journal of Arid Environments, 191, 104533. DOI:10.1016/j.jaridenv.2021.104533

Király, G., & Miskolczi, P. (2019). Dynamics of participation: System dynamics and participation—An empirical review. Systems Research and Behavioral Science, 36(2), 199–210. https://doi.org/10.1002/sres.2580

Kiteme, B. P., & Gikonyo, J. (2002). Preventing and resolving water use conflicts in the Mount Kenya highland–lowland system through water users' associations. Mountain research and development, 22(4), 332–337. DOI:10.1659/0276-4741(2002)022[0332:PARWUC]2.0.CO;2

Koech, C. K., Ongugo, P. O., Mbuvi, M. T. E., & Maua, J. O. (2009). Community Forest Associations in Kenya: challenges and opportunities. Kenya Forestry Research Institute.

Lesrima, S., Nyamasyo, G., & Kiemo, K. (2021). Unresolved water conflicts by water sector institutions in Ewaso Ng’iro North River Sub-Basin, Kenya. Journal of Applied Sciences and Environmental Management, 25(2), 269–275. DOI:10.4314/jasem.v25i2.21

Liniger, H., Gikonyo, J., Kiteme, B., & Wiesmann, U. (2005). Assessing and managing scarce tropical mountain water resources: The case of Mount Kenya and the semiarid Upper Ewaso Ng'iro basin. Mountain Research and Development, 163–173. DOI:10.1659/0276-4741(2005)025[0163:AAMSTM]2.0.CO;2

Lybbert, T., & Sumner, D. (2010). Agricultural technologies for climate change mitigation and adaptation in developing countries: policy options for innovation and technology diffusion. https://doi.org/10.1016/j.foodpol.2011.11.001

Marshall, N. A., Park, S. E., Adger, W. N., Brown, K., & Howden, S. M. (2012). Transformational capacity and the influence of place and identity. Environmental Research Letters, 7(3), 034022. DOI: 10.1088/1748-9326/7/3/034022

Martinez-Fernandez, J., Banos-Gonzalez, I., & Esteve-Selma, M. A. (2021). An integral approach to address socio-ecological systems sustainability and their uncertainties. Science of the Total Environment, 762, 144457. DOI:10.1016/j.scitotenv.2020.144457

Mati, B. M., Sijali, I. V., & Ngeera, K. A. (2022). Investigating pathways for agricultural innovation at scale: Case studies from Kenya.

McCord, P., Dell'Angelo, J., Baldwin, E., & Evans, T. (2017). Polycentric Transformation in Kenyan Water Governance: A Dynamic Analysis of Institutional and Social-Ecological Change. Policy Studies Journal, 45(4), 633–658. https://doi.org/10.1111/psj.12168

Menapace, L., Colson, G., & Raffaelli, R. (2013). Risk aversion, subjective beliefs, and farmer risk management strategies. American Journal of Agricultural Economics, 95(2), 384–389. DOI:10.1093/ajae/aas107

Mertzani, A., & Pitt, J. (2023, September). Sustainable self-organisation of socio-techno-ecological systems. In 2023 IEEE International Conference on Autonomic Computing and Self-Organizing Systems Companion (ACSOS-C) (pp. 92–97). IEEE. DOI:10.1109/ACSOS-C58168.2023.00046

Miller, J. H., & Page, S. E. (2009). Complex adaptive systems: an introduction to computational models of social life: an introduction to computational models of social life. Princeton university press.

Molu, G. G., & Luketero, S. (2025). Influence of prepaid smart water metering on performance of community water projects in Marsabit County, Kenya. International Academic Journal of Information Sciences and Project Management, 3(7), 351–369.

Musyoki, J. K., Mugwe, J., Mutundu, K., & Muchiri, M. (2016). Factors influencing level of participation of community forest associations in management forests in Kenya. Journal of Sustainable Forestry, 35(3), 205–216. DOI:10.1080/10549811.2016.1142454

Mutiga, J. K., Mavengano, S. T., Zhongbo, S., Woldai, T., & Becht, R. (2010). Water allocation as a planning tool to minimise water use conflicts in the Upper Ewaso Ng’iro North Basin, Kenya. Water resources management, 24, 3939–3959. DOI:10.1007/s11269-010-9641-9

Neely, K., & Walters, J. P. (2016). Using causal loop diagramming to explore the drivers of the sustained functionality of rural water services in Timor-Leste. Sustainability, 8(1), 57. DOI:10.3390/su8010057

Noor, R., Inam, A., Zahra, S. M., Shoaib, M., Riaz, R., Sarwar, A., ... & Ahmad, S. (2022). A methodological framework for modeling sustainability visions: A case study of groundwater management in Faizpur distributary, Pakistan. Agricultural Water Management, 271, 107822. DOI: 10.1016/j.agwat.2022.107822

Ntukamazina, N., Onwonga, R. N., Sommer, R., Rubyogo, J. C., Mukankusi, C. M., Mburu, J., & Kariuki, R. (2017). Index-based agricultural insurance products: challenges, opportunities and prospects for uptake in sub-Sahara Africa. Journal of Agriculture and Rural Development in the Tropics and Subtropics (JARTS), 118(2), 171–185.

Oguge, N., & Oremo, F. (2017). Fostering the use of rainwater for off-season small-scale irrigation in arid and semi-arid areas of Kenya. In Rainwater-Smart Agriculture in Arid and Semi-Arid Areas: Fostering the Use of Rainwater for Food Security, Poverty Alleviation, Landscape Restoration and Climate Resilience (pp. 159–174). Cham: Springer International Publishing. DOI:10.1007/978-3-319-66239-8_8

Olando, C. O., Jagongo, A. O., & Mbewa, M. O. (2013). The contribution of SACCO financial stewardship to growth of SACCOS in Kenya.

Olson Jr, M. (1971). The Logic of Collective Action: Public Goods and the Theory of Groups, with a new preface and appendix (Vol. 124). Harvard university press.

Olsson, P., Moore, M. L., Westley, F. R., & McCarthy, D. D. (2017). The concept of the Anthropocene as a game-changer: a new context for social innovation and transformations to sustainability. Ecology and Society, 22(2). DOI:10.5751/ES-09310-220231

Ongugo, P. O., Mogoi, J. N., Obonyo, E., & Oeba, V. O. (2008). Examining the roles of community forest associations (CFAS) in the decentralization process of Kenyan forests.

Ostrom, E. (1990). Governing the commons: The evolution of institutions for collective action. Cambridge university press.

Ostrom, E. (2002). Common-pool resources and institutions: Toward a revised theory. Handbook of agricultural economics, 2, 1315–1339. https://doi.org/10.1016/S1574-0072(02)10006-5

Ostrom, E. (2007). A diagnostic approach for going beyond panaceas. Proceedings of the national Academy of sciences, 104(39), 15181–15187. https://doi.org/10.1073/pnas.0702288104

Ostrom, E. (2009). Collective action theory. DOI:10.1093/oxfordhb/9780199566020.003.0008

Ostrom, E., Gardner, R., & Walker, J. (1994). Rules, games, and common-pool resources. University of Michigan press.

Pannell, D. J., Marshall, G. R., Barr, N., Curtis, A., Vanclay, F., & Wilkinson, R. (2006). Understanding and promoting adoption of conservation practices by rural landholders. Australian journal of experimental agriculture, 46(11), 1407–1424. DOI:10.1071/EA05037

Partelow, S. (2018). A review of the social-ecological systems framework. Ecology and Society, 23(4). DOI:10.5751/ES-10594-230436

Polaine, X. K., Dawson, R., Walsh, C. L., Amezaga, J., Peña-Varón, M., Lee, C., & Rao, S. (2022). Systems thinking for water security. Civil Engineering and Environmental Systems, 39(3), 205–223. DOI:10.1080/10286608.2022.2108806

Ram, S. A., & Irfan, Z. B. (2021). Application of System Thinking Causal Loop Modelling in understanding water Crisis in India: A case for sustainable Integrated Water resources management across sectors. HydroResearch, 4, 1–10. DOI:10.1016/j.hydres.2021.02.001

Richards, N., & Syallow, D. (2018). Water resources users associations in the Mara Basin, Kenya: Pitfalls and opportunities for community based natural resources management. Frontiers in Environmental Science, 6, 138. DOI:10.3389/fenvs.2018.00138

Rogers EM (2003) ‘Diffusion of innovations.’ 5th edn. (Free Press: New York, NY). DOI:10.1016/j.jmig.2007.07.001

Rural Focus Ltd. (2022). Abstraction & Pollution Survey of Timau River: Final. Timau River Project.

Scheffer, M., Bascompte, J., Brock, W. A., Brovkin, V., Carpenter, S. R., Dakos, V., ... & Sugihara, G. (2009). Early-warning signals for critical transitions. Nature, 461(7260), 53–59. DOI:10.1038/nature08227

Schmitt Olabisi, L., Liverpool-Tasie, S., Rivers, L., Ligmann-Zielinska, A., Du, J., Denny, R., ... & Sidibé, A. (2018). Using participatory modeling processes to identify sources of climate risk in West Africa. Environment Systems and Decisions, 38, 23–32. DOI:10.1007/s10669-017-9653-6

Scoones, I. (2010). Dynamic sustainabilities: Technology, environment, social Justice. Routledge. DOIhttps://doi.org/10.4324/9781849775069

Selin, H., & Selin, N. E. (2023). The human–technical–environmental systems framework for sustainability analysis. Sustainability Science, 18(2), 791–808. DOI:10.1007/s11625-022-01177-0

Smith, A., & Stirling, A. (2008). Social-ecological resilience and socio-technical transitions: critical issues for sustainability governance.

Sohns, A., Ford, J. D., Adamowski, J., & Robinson, B. E. (2021). Participatory modeling of water vulnerability in remote Alaskan households using causal loop diagrams. Environmental Management, 67(1), 26–42. DOI:10.1007/s00267-020-01387-1

Sorge, S., Mann, C., Schleyer, C., Loft, L., Spacek, M., Hernández-Morcillo, M., & Kluvankova, T. (2022). Understanding dynamics of forest ecosystem services governance: A socio-ecological-technical-analytical framework. Ecosystem Services, 55, 101427. DOI:10.1016/j.ecoser.2022.101427

Sterman, J. (2002). System Dynamics: systems thinking and modeling for a complex world.

Timau WRUA. (2023). Timau WRUA Sub Catchment Management Plan (2023–2033). Timau WRUA.

Virapongse, A., S. Brooks, E. C. Metcalf, et al. 2016. “A Social–Ecological Systems Approach for Environmental Management.” Journal of Environmental Management 178: 83–91. https://doi.org/10.1016/j.jenvman.2016.02.028

Wazed, S. M., Hughes, B. R., O’Connor, D., & Calautit, J. K. (2018). A review of sustainable solar irrigation systems for Sub-Saharan Africa. Renewable and Sustainable Energy Reviews, 81, 1206–1225. https://doi.org/10.1016/j.rser.2017.08.039

Wei, Y., Wu, S., & Tesemma, Z. (2018). Re-orienting technological development for a more sustainable human–environmental relationship. Current Opinion in Environmental Sustainability, 33, 151–160. https://doi.org/10.1016/j.cosust.2018.05.022

White, S. M. (2015, April). Systems theory, systems thinking. In 2015 Annual IEEE Systems Conference (SysCon) Proceedings (pp. 420–425). IEEE. DOI:10.1109/SYSCON.2015.7116787

Zhang, T., Tan, Q., Zhang, S., Zhang, T., & Zhang, W. (2021). A participatory methodology for characterizing and prescribing water-energy-food nexus based on improved casual loop diagrams. Resources, Conservation and Recycling, 164, 105124. https://doi.org/10.1016/j.resconrec.2020.105124

ACE TAF. (2019). Importation guidelines for solar PV products and systems in Kenya 2019. Africa Clean Energy Technical Assistance Facility. Retrieved from https://www.ace-taf.org/kb/importation-guidelines-for-solar-in-kenya/

Githinji, E. M. (2021, November 5). Tax Incentives on Renewable Energy. Clean Energy 4 Africa. Retrieved from https://cleanenergy4africa.org/tax-incentives-on-renewable-energy/

GoK. (2002). The Water Act, 2002. Government Printer, Nairobi. Nairobi: Government Printer. Retrieved from https://www.wra.go.ke/

GoK (2016). The Water Act, 2016. The Kenya Gazette Supplement. Nairobi: Government Printer. Retrieved from https://www.wra.go.ke/

Yes