Lessons Learned from a North-South Science Partnership for Sustainable Development

Abstract

SDG goal 17 seeks to strengthen global partnerships, especially between Global North and South. However, in research and development, experiences indicate a mismatch in expectations with perceived power, funding and workload imbalances, a situation derogated as ‘parachute science’ or ‘helicopter research’. The research programme SPACES seeks to enhance North-South collaborations. As an inter- and transdisciplinary research programme focusing on the interactions between land, sea, atmosphere, biosphere and society, it aims to be a forum for fruitful partnerships. In this chapter, we carve out lessons learned from the nine projects involved in the programme’s second phase. Based on a survey amongst 66 SPACES II scientists, we explored their motivations for collaboration, their involvement in decision-making, the assigned resources and workloads as well as conflicts between Northern and Southern teams. Furthermore, we conducted bibliometric analyses and observed an intensification of the North-South co-authorship network over time. We conclude that SPACES can be considered a success as researchers acted largely as peers on an equal footing. Nevertheless, our insights show that (1) the asynchrony in funding is a threat for effective collaborations, (2) continuous project evaluation should incorporate a North-South component and (3) collaborative publications should be formalised as a tool for integration.

1 Introduction

A rapidly growing human population and global environmental change, particularly the impacts of climate change, are expected to put pressure on the natural resource base of the African continent (Gasparatos et al. 2017), on which the majority of the population depends for their livelihoods and which are important for economic growth and inclusive development (Gupta and Vegelin 2016). In order to meet the Sustainable Development Goals (SDGs), there is a need to deal with the multi-faceted and interlinked sustainability challenges faced by countries in Africa. This needs to be done from an inter- and transdisciplinary research approach with multiple perspectives to produce knowledge relevant to real-world problems (Pohl and Hirsch Hadorn 2007). In addition to working on pressing environmental challenges, respective research projects are now increasingly expected to consider the needs of society (Tress et al. 2005), whilst also having policy implications. Fulfilling these expectations requires collaboration of teams from the natural and social sciences, including non-academic stakeholders (e.g. institutional actors and local communities) to work on sustainability challenges (Patel et al. 2017). One key component for sustainable development is described in SDG 17, which calls for strengthening global partnerships (Waage et al. 2015). Respective partnerships can take the shape of South-South research collaborations, which are important for regional scientific and political integration (Boshoff 2010). Since the mid-twentieth century, however, North-South collaborations (NSCs) have received more attention particularly in the discourses around post-colonialism and the diverse strands of development aid (Gaillard 1994). In 1979, the United Nations adopted the ‘Vienna Program of Action’ stating key criteria to strengthen international research collaborations such as accounting for development priorities of the South, joint participation and control as well as capacity development (Gaillard 1994; UN 1979). Unfortunately, respective aspirations have still not been fully met as current critical perspectives on NSC confirm. Bradley (2017), for instance, explores the agenda-setting processes and power imbalances between Northern and Southern research partners and finds prevailing deficits (Bradley 2017). The term ‘parachute science’ or ‘helicopter research’ evolved in this regard to critically describe the way Northern researchers conduct research in the South without adding any benefit to the Southern science sphere (Giller 2020; Stefanoudis et al. 2021). Approaches to account for the deficits in NSC often refer to the necessity to follow a transdisciplinary mode of research (Schmidt and Pröpper 2017). More specifically, recent investigations particularly highlight the need to acknowledge the double-role of scientists from the South (acting as researchers and field facilitators), the need for joint publications to foster capacity building, as well as methodologically guided knowledge integration and co-production (Luetkemeier et al. 2021).

One such NSC initiative is the research programme Science Partnerships for the Adaptation to Complex Earth System Processes in Southern Africa (SPACES), funded by the German Federal Ministry for Education and Research (BMBF). The programme’s first phase was launched in 2012 (SPACES I), whilst the current second phase was initiated in 2018 (SPACES II). The programme can be classified as a type-I NSC (Bradley 2007:13) as it is characterizsed by research teams that are explicitly constructed to carry out a certain project. The BMBF funds collaborative German-African research projects in South Africa and Namibia that contribute to the formulation of policy recommendations for Earth system management. The programme is aligned with the national and international trends and initiatives on international collaboration, including SDG 17 and the BMBF’s Africa strategy of 2014. The programme was from the start defined as a joint endeavour with the Department of Science and Innovation (DSI) of South Africa and was intended to contribute to intensified cooperation with the Namibian Ministry of Education (MET). A core aim of the programme is to promote scientific networking between African and German research institutions. The programme’s research projects were intended to focus on jointly defined areas, consider both African and German interests, emphasise partnership and ownership and should enable continuity and reliability of collaborations. Scientific exchange between partner countries and international networking were stressed as expectations from BMBF. In this regard, the success indicators put strong emphasis on NSC in the form of, for example, the number of joint German-African publications, and jointly supervised students (BMBF 2017).

Against this background, this chapter examines the structure and quality of NSC within SPACES II to carve out both successful ways of collaboration, but also deficits in how Northern and Southern researchers worked together. We draw lessons learned to contribute to more efficient and equitable future research and development projects between Northern and Southern partners. Section 31.2 provides an overview on the methods we applied to explore the North-South collaborations. Section 31.3 presents the results of our investigations and Sect. 31.4 discusses our findings against the current state of knowledge in the scientific literature. Section 31.5 draws conclusions for future collaborative research projects.

2 Material and Methods

For exploring the structure and quality of NSC in SPACES II, we applied three distinct methods. First, we screened the research proposals of the projects involved to qualitatively assess how the research teams initially designed the North-South collaboration technically and conceptually. Second, we conducted a structured online survey amongst researchers involved in SPACES II to obtain their perspectives on how the collaboration actually manifested. Third, we quantitatively explored the evolution of co-authorships amongst Northern and Southern project partners over time. In general, we separate between Northern and Southern researchers based on their actual institutional affiliation and not based on their personal nationality. This means that a researcher from South Africa who is currently affiliated to a German institution is considered a Northern researcher in our investigations and vice versa.

2.1 Structured Online Survey

Most studies that seek to evaluate North-South collaborations follow a qualitative research design (Luetkemeier et al. 2021; Schmidt and Pröpper 2017; Casale et al. 2011), as the topic itself is highly diverse and case-specific as well as often subject to a small number of researchers involved in respective projects. As SPACES is a research programme with nine individual research projects in its second phase that partly build upon prior collaborations, a reasonable number of researchers were involved. Taking advantage of this large number of researchers, a quantitative empirical research approach was considered as a valuable way forward to representatively identify similarities and differences on how researchers consider NSC amongst German, Namibian, and South African partners.

We initially compiled a list of researchers who were involved in ongoing research activities of SPACES II at the time of writing this chapter, based on the feedback of the respective project leads. Though the projects apply varying definitions in terms of who qualifies as a ‘project researcher’ (e.g. Southern researchers are not formally funded [cf. Sect. 31.3.1], so their actual affiliation to SPACES may be hidden), we consider the compiled contact list as reasonably comprehensive. All projects listed both Northern (124) and Southern researchers (87) with an average of 10 on the Southern and 14 on the Northern side. This results in a total statistical population of N = 211. The online survey was pre-tested amongst four researchers (two from the North, two from the South) and adapted somewhat to increase clarity in the language used in some questions. It was implemented and rolled out using the software LimeSurvey (2021) in July 2021.

In agreement with the project leads, we initially screened the nine project proposals for statements on how the collaboration between Northern and Southern researchers was designed. We specifically looked into the structural project setup in terms of (1) how the lead team was composed, (2) if prior collaborations existed, (3) what tasks Northern and Southern partners were assigned to and (4) which tools were planned to foster collaboration. This helped us to gain a basic understanding of how the nine projects were developed and what importance NSC received in initial considerations. Against this background, we built a structured questionnaire to gain quantitative insights into the researchers’ viewpoints on selected NSC issues. In order to receive as many responses as possible, we considered a time span of about 10 min for answering the questions as a suitable compromise between scope and applicability of the survey. The survey received an ethical clearance from the ethics board of Institute for Social-Ecological Research (ISOE). All respondents were guaranteed that their participation was fully anonymous and the results were solely used for the analysis of NSC as presented in this chapter.

As a basic structure of the survey, we made use of the ‘SDG Partnership Guidebook’ that suggests four pillars for effective partnerships in international collaborations for sustainable development (Stibbe and Prescott 2020). Though this framework is not specifically targeted towards research projects, we consider it as a comprehensive list of aspects to analyse the quality of NSC. The structured part of the survey was set up along the following topics:

• Fundamentals: What is the benefit for project partners to collaborate and how were the projects developed?

• Partnership relationship: Do researchers work transparently and trustfully with adequate involvement in decision-making processes?

• Structure and setup: Were resources, tasks and workload assigned fairly to all project partners?

• Management and leadership: Did tensions and conflicts occur between project partners and how was the communication implemented?

In addition to the structured questions, open questions on the impact of COVID-19 on project work and crucial issues to consider when developing future NSC projects were posed at the end of the survey. The structured questions were designed in a way to apply a four-point Likert-scale (Joshi et al. 2015). The respondents were hence asked to reveal either a tendency for approval or rejection of the questions and statements presented. As a fifth category, respondents were able to select ‘no answer’, which we did not treat as a neutral category in between approval and rejection, but rather as a non-response or missing value. The ordinal scale responses can be statistically evaluated, especially with respect to the differences between Northern and Southern partners. For doing this, the non-parametric Mann-Whitney U test (also known as Wilcoxon rank-sum test) for two independent samples was applied (MacFarland and Yates 2016). The null hypothesis that the two samples’ means do not significantly deviate from each other was rejected, if p < 0.05.

For analysing the open questions posed in the survey, we used the software MaxQDA (VERBI Software 2019) to code the respondents’ statements. The coding scheme evolved whilst working through the statements with a final grouping of similar codes.

2.2 Co-authorship Network Analyses

The purpose of the bibliometric analyses was to back up and verify statements and literature insights on the role of collaborative publications. The analyses served to visualise the evolution of the cooperative publication network, with a focus on Northern and Southern researchers, prior and during the SPACES I and II programmes. Publication data of SPACES II researchers (211 researchers as well as other academic staff in the nine research projects) was collected from the Scopus database. Scopus was selected due to the availability of full author affiliation data. The searches were limited to articles, conference papers, reviews, book chapters, books and data papers in English language. The scientific field was limited to Earth and Environmental sciences, Geosciences, Ecology, Marine sciences, computer sciences and related disciplines.

To study the evolution of collaborations, we analysed three 4-year periods: (1) immediately prior to SPACES (2009–2012); (2) during the first phase of SPACES (2013–2016); (3) immediately before and during SPACES II (2017–2020). These time brackets are indicative, as there were large differences in the start and end dates of the different projects (i.e. the first SPACES II project started in June 2018 and the last in February 2019). The final dataset for 2009–2012 consisted of 852 publications; 2013–2016 consisted of 1175 publications; and 2017–2020 of 1436 publications.

For visualising co-authorship networks by individual researchers and by institutes, we used the network visualisation function of VOSviewer (van Eck and Waltman 2010). For the analyses, different forms of author names were merged by building a dictionary (e.g. G.F. Midgley would be synonymous with G. Midgley). Similarly, different forms of spelling institute names were combined for the analyses, and furthermore, different departments of the same institute were merged to represent the parent institute as unit of analysis (e.g. ‘Stellenbosch University’). Only collaborating institutes of SPACES I or II were selected in the visualisations.

For the network visualisations, we included all individuals and institutes who had co-authored at least one publication with another SPACES II researcher or another SPACES II institute. We used fractional counting to control the disproportionate impact of multi-author documents (see Perianes-Rodriguez et al. 2016 for discussion on benefits). In fractional counting, the number of co-authors affects the weighting: for example a paper co-authored by 20 authors gets a weight of 1/20. We also excluded publications with more than 50 authors from the analyses, with the expectation that these outputs are less representative of true collaborative relationships. To construct the mapping, VOSviewer uses the ‘association strength’ similarity measure, which was normalised to control the large differences between author publishing activities. Clustering resolution was set to 1.0, as by default in VOSviewer; by changing the resolution, clusters can be made larger or smaller to explore the data (van Eck and Waltman 2009). Network visualisation was used to show the development of co-authorship links amongst individual researchers, and overlay visualisation was used to identify the main institutes contributing to NSC publications.

3 Results

For presenting the results of our study, we begin with a brief overview on how the proposals of the nine SPACES II projects addressed NSC. Subsequently, we present selected results of the online survey with a particular focus on common grounds and significant differences between the viewpoints of Northern and Southern researchers. Afterwards, we particularly shed light on the evolution of the collaboration between Northern and Southern researchers, based on co-authorships in scientific publications.

3.1 Designing Effective Collaborations in Project Proposals

Screening the proposals confirmed that the projects cover a broad range of topics from Earth system sciences and likewise differ strongly in their scientific setup ranging from clearly focused interdisciplinary to extensive transdisciplinary approaches. In this sense, the projects also deviated in terms of (1) formal assignment of responsibilities to African partners, (2) allocation of grants and resources for African institutions as well as (3) tools planned to enhance collaboration between Northern and Southern researchers.

The responsibilities of African partners varied strongly across the projects. Whilst the work packages of some projects were solely led by German partners, most projects explicitly implemented a tandem structure in which both a Northern and a Southern partner were formally considered as work package leads. For some projects, this even held true for the structure of the overall project lead. Though the lead structures were rather heterogeneous, most proposals clearly outlined the responsibilities and tasks of Northern and Southern partners.

In SPACES II, the BMBF funded the German partners, whereas the African partners were expected to apply for funding from their country or provide evidence of counter-financing on their own, according to prior agreements with the South African and Namibian ministries. Therefore, African partners were not formally allowed to be funded in the same way as the German partners. Some projects sub-contracted their African collaborators and hence made financial means available for researchers to carry out particular tasks. However, funds provided through sub-contracts are limited to a certain proportion of the total project budget and to service-based rather than research contracts.

In terms of collaborative tools, the projects formally referred to the SPACES-associated programme of the German Academic Exchange Service (DAAD) to fund mutual visits of researchers. Some projects went further and explicitly mentioned publications as a key tool for collaboration as well as data-sharing protocols. One project even introduced a steering committee composed of Northern and Southern researchers to guarantee joint decision-making.

3.2 Researchers’ Perspectives on North-South Collaboration

In total, 66 researchers of the SPACES II programme participated in the online survey (response rate of 31%) during the month of July in 2021. About 36% of them came from a South African or Namibian institution, whilst the remaining 64% were affiliated to a German institution. These shares correspond to the distribution in the statistical population of Southern (87) and Northern (124) researchers. About 38% of the respondents were female, 59% were male and about 3% preferred to not answer this question. Most of the respondents (12 Southern and 23 Northern researchers) described themselves as holding a leading position in their respective projects, either as overall project leads, work package leads or principal investigators. Only about one-fourth of the respondents declared themselves as PhD students. About 48% of the respondents (11 Southern and 21 Northern partners) stated that they did not have any prior collaborations with their current Northern or Southern project partners, respectively.

Fig. 31.1

Motivations for researchers and research groups to join the North-South collaboration. Results are based on question 5 of the online survey with a sample size of n = 66. According to the Mann-Whitney-U-test, no significant differences could be found between Northern and Southern researchers

3.2.1 Fundamentals

When considering the motivations of project partners to join the SPACES II programme, the differences between Northern and Southern researchers were not statistically significant (question 5). They expected benefits (1) in expanding their professional network, (2) in sharing knowledge and mutual learning as well as (3) from gaining novel research insights, which conventional projects might not provide (Fig. 31.1). ‘Access to a new study area’, ‘tapping into new research funds’ and ‘gaining access to resources to carry out research’ were less important than the aforementioned ones but still more than 50% of the respondents consider them relevant. Looking at the fulfilment of these motivations and expectations, the majority of the Southern researchers indicated that they were ‘absolutely satisfied’ or at least ‘to a large degree’ (question 6). The expectations for improved networking scored highest with a consent of about 78%.

In terms of designing the overall research question, the participants’ responses revealed that this process was organised in a collaborative manner in most cases. Often, SPACES II projects could build on prior collaborations to develop the project proposal, and researchers carried out an integrating process, for example, via workshops (question 8). As Fig. 31.2 presents, a high proportion of Northern partners (81%) compared to Southern partners (54%) were either ‘completely’ or ‘to a larger degree’ involved in the initial drafting of the overall research questions. This difference is statistically significant.

Fig. 31.2

Researchers’ evaluation regarding their integration into the initial process of defining/drafting the overall research question(s). Results are based on question seven of the online survey with a sample size of n = 66. The responses of Northern and Southern partners are significantly different with p < 0.05

Aside from the initiating process, a partnering mindset amongst the researchers is a key attribute of successful North-South collaborations (Stibbe and Prescott 2020:70). Our respondents used terms like ‘communication’, ‘transparency’, ‘knowledge sharing’ and ‘respect’ to illustrate what they consider a partnering mindset. In this regard, about 69% had a positive view on the partnering mindset without a significant difference between Northern and Southern researchers (question 10).

3.2.2 Partnership Relationship

To look at how the researchers evaluated their partnership in more detail, we asked the respondents if they consider their respective Northern or Southern counterparts as acting transparently (question 13) and as committed in the project (question 12). The responses showed an appreciation for each side with no significant differences between the North and the South. Both groups evaluated the commitment and the transparent way of acting as largely positive (>90%). However, when looking deeper into specific issues of project collaboration, some critical aspects appeared to be of concern. One issue was that according to formal funding requirements, official project reports had to be written exclusively in German. This may pose a challenge to a collaborative project, in which working and publication language is English. Figure 31.3 shows that about 40% of the researchers considered this requirement as a serious obstacle for project collaboration, whilst no significant difference could be found between the two groups (question 14).

Fig. 31.3

Researchers’ evaluation on whether the fact that official project reports to the funder had to be written exclusively in German language was an obstacle to collaboration. Results are based on question 14 of the online survey with a total sample size of n = 57. Differences in responses from Northern and Southern partners are not statistically significant at p < 0.05

A discrepancy between Northern and Southern researchers became obvious in terms of involvement in decision-making processes with respect to project administration and management (question 16) as well as practical research conduction (question 15). In both cases, the Southern researchers felt that they had been less involved in decision-making processes than their Northern counterparts had. For both questions, the differences between the groups were statistically significant (Fig. 31.4).

Fig. 31.4

Perceived degree of involvement in (a) decision-making process with respect to project administration (e.g. funding formalities, reporting, representation) and (b) decision-making processes with respect to practical research conduction (e.g. field work planning, case study selection, method selection). Results are based on questions 15 and 16 of the online survey with a total sample size of n = 62. The responses of Northern and Southern partners are significantly different with p < 0.05

Moving on to even more practical and tangible issues in research collaboration, the question of authorship was assessed. We asked the researchers if they considered the way they have been involved in publishing collaborative research results was adequate or not (question 17). Figure 31.5 shows that overall, 84% of the respondents had a rather positive view on their involvement in publishing collaborative research results. None of the researchers chose to answer that ‘absolutely no’ involvement took place. However, in contrast to all other answer categories more Southern researchers answered ‘somewhat involved’, though this difference is not statistically significant due to the small sample size. Section 31.3.4 looks deeper into the evolution of co-authorship amongst Northern and Southern partners throughout the course of the programme.

Fig. 31.5

Evaluation of researchers from the North and the South, whether they have been adequately involved in the publication of joint research. Results are based on question 17 of the online survey with a total sample size of n = 64. The responses from Northern and Southern researchers do not differ statistically significantly (p < 0.05)

3.2.3 Structure and Setup

From a structural perspective, the survey results revealed insights into the researchers’ focus of work (question 18). Figure 31.6 shows to which activities the respondents allocated their personal workforce. More than 60% of all respondents indicated that they spent most of their work on empirical research, followed by student supervision and field logistics. Overall, the differences between Northern and Southern researchers were not statistically significant, except for the communication with the funding agency, for which the Northern partners allocated more work than their Southern partners did.

Fig. 31.6

Workload assigned to certain project activities by researchers. Results are based on question 18 of the online survey with a total sample size ranging between n = 62 and n = 65. The responses from Northern and Southern researchers do not differ statistically significantly (p < 0.05), except for ‘funder communication’

Fig. 31.7

Researchers’ evaluation of if Southern partners have been adequately supported with resources (e.g. funds, staff, equipment). Results are based on question 20 of the online survey with a total sample size of n = 63

In terms of funding and resources endowment, the online survey assessed respective viewpoints. With respect to the Southern partners’ in-kind or non-monetary contributions, such as the provision of working hours and equipment (question 19), the results show that 58% of the respondents consider these as adequately acknowledged. Here, no significant differences could be found between the North and the South.

Considering the endowment of Southern partners with adequate resources for conducting project activities, Fig. 31.7 shows the perception of Northern and Southern researchers (question 20). The responses showed a critical perspective as about 71% of the respondents considered the Southern partners as only ‘somewhat supported’ or worse. The views between Northern and Southern researchers did not vary. Interestingly, though not statistically significant, a larger share of the Southern partners considered their own resources endowment as being adequate, whilst more Northern partners considered it the other way round.

3.2.4 Management and Leadership

As a key component of formal project collaboration and integration, sharing data and results was highlighted by most projects in their proposals. In this regard, 86% of our respondents confirmed respective processes were implemented as intended (question 23). Similarly, the researchers had a positive perception of how responsibilities and expectations were shared amongst the involved researchers. The majority declared that only minor tensions and conflicts were registered during the project period (question 21). In both cases, no significant differences could be found between the two groups.

Figure 31.8 shows how the researchers evaluated different tools in their potential to keep Southern or Northern colleagues, respectively, up to date about ongoing research activities (question 22). Regular project meetings and bilateral exchange were considered as the primary means for mutual updates. However, a considerable number of respondents revealed that respective tools have either not been used in the projects, or have been implemented poorly so that no continuous mutual update was possible.

Fig. 31.8

Evaluation of researchers, which communication channels/tools helped them during the course of the project to stay informed about the activities of their Northern/Southern partners. Results are based on question 22 of the online survey with a sample size ranging from n = 61 to n = 63. The responses from Northern and Southern researchers do not differ significantly

3.3 Qualitative Insights from the Survey

Alongside the structured survey part, the respondents were invited to comment on two issues in an open, unrestricted way. First, we assessed how the COVID-19 pandemic affected NSC and second, the researchers were asked to mention the key issues in which they see a need for action to enhance NSC in a hypothetical follow up funding programme.

3.3.1 Impact of COVID-19

About 91% of our respondents took the opportunity and reported about outcomes for project collaboration. In general, the researchers agreed that the COVID-19 related restrictions of public life like travel restrictions (Devi 2020) led to impairments of research activities. The following consequences were highlighted in more detail.

• Delay of research activities: Due to travel restrictions, empirical fieldwork (e.g. data acquisition) was delayed or even cancelled. Some projects missed essential periods to assess relevant data to accomplish their research goals (e.g. missed growing season). Experiments had to be cancelled or restarted, which led to higher costs. Overall, researchers assume that this may have led to reduced scientific output.

• Exchange between researchers suffered: Travel restrictions prohibited joint fieldwork sessions, which are usually seen as a key element for productive exchange between Northern and Southern researchers. No conferences and project meetings could be held in person and thus, important forums for exchange and networking could not be implemented. However, some respondents observed an improved communication between Northern and Southern partners via the increased use of digital communication technologies (cf. Fig. 31.8). Long-established research collaborations were considered as more effective in coping with the COVID-19-induced research restrictions.

• Higher workload for Southern partners: Though the quantitative survey results did not indicate a significant difference between the workload of Northern and Southern researchers, some respondents indicated that COVID-19 put more pressure on the South. As travel restrictions prohibited German researchers to travel, Southern partners had to take over essential research tasks (e.g. data acquisition, stakeholder communication).

• Negative psychological impacts: The delay of research activities was considered by some of our respondents as having negative impacts on the motivation of researchers to conduct project tasks. This may have been particularly true for student researchers who only had a limited period available to conduct their research tasks and who may have cancelled their activities.

3.3.2 Fields of Action to Improve North-South Collaborations

We asked our respondents to state crucial aspects that should be considered for future research projects to enhance NSC (question 24). In this regard, 55 out of 66 shared their opinions on this issue and we condensed these into the following dimensions.

• Improved funding opportunities for Southern partners: Most respondents from both the North and the South indicated that they see the necessity to improve the funding opportunities of Southern partners. The recommendations covered a wide range of topics, but frequently touched upon the points of enhancing financial endowment for fieldwork and laboratory analyses, more options for Southern researchers and students to visit their Northern counterparts (e.g. for mutual learning and utilising research infrastructure) and improved funding modalities for Southern students, in general. One female African scientist boiled it down to ‘funding directed to Southern partners reduces the “parachute science” feeling and empowers the Southern partners’.

• More equitable project design and management: Respondents often highlighted the necessity to better integrate Southern partners into overall project management. This includes both the design phase for proposal writing (aligning research activities to local needs) as well as continuous project management and decision-making processes. One key issue raised was the request to more explicitly declare certain research objectives and clearly assign tasks and responsibilities to certain actors of the project team.

• Enhanced exchange between researchers and students: The respondents saw the need to improve the exchange between Northern and Southern researchers on both the Post-Doc and the student level. Personal exchange with one another was seen as a key element to enhance NSC, which could, for instance, be fostered in the field of empirical research, collaborative teaching, student supervision and exchange programmes.

3.4 Role of SPACES in Fostering Collaborative Publications

In addition to the insights into the role of co-authorships between Northern and Southern researchers, we explored this issue in a quantitative way to reveal if SPACES was able to enhance collaboration as measured by this indicator. For this purpose, the following subsections first look into the co-authorship networks of individual researchers, then zoom out onto the institutional level and finally assess the relative proportion of collaborative papers against the total number of papers published by SPACES researchers.

3.4.1 Co-authorship Networks Between Individual Researchers

During the years 2009–2012, one-third (69 out of 211) of SPACES II network researchers were involved in publications co-authored with another SPACES researcher, irrespective of his/her regional affiliation. During 2013–2016, the proportion had increased to 45% (95 researchers), and during the last time bracket, 2017–2020, the majority 82% (147 researchers) had co-authored publications with another SPACES II researcher. All authors contributing most to the total co-authorships were German: during 2009–2012 F Jeltsch (total link strength, i.e. the total strength of the co-authorship links of the researcher with others in the network 14.0) and N Blaum (11.0); during 2013–2016 S Higgins (17.0), T Haberzettl and R Mäusbacher (14.0), and during 2017–2020, H Kunstmann (27.0) and P Laux (22.0). The strongest co-authorship links (i.e. those with most co-authored publications; by fractional counting, publications with higher number of co-authors are given less weight) were between two German authors (during 2009–2012 Jeltsch and Blaum, 9.3; during 2013–2016 Higgins and Scheiter 8.5; and during 2017–2020 Kunstmann and Laux 21.5) or two South African authors (during 2009–2012 Chirwa and Syampungani, 7.0).

Similarly, an expanding network is seen when limiting the focus to North-South co-authored publications (i.e. links between German and southern African authors). Figure 31.9 shows the co-authorship network constructed from the publications of SPACES II researchers during the three selected time brackets. Each circle represents a researcher, and the size of the circle indicates ‘total link strength’. Lines amongst researchers represent co-authorship links: researchers connected with a line have at least one co-authored publication, and those with thicker lines have a stronger link. In addition, authors are clustered together according to collaboration strength, i.e. those authors that publish more with each other are located closer together. To highlight the evolution of NSC, only respective collaborative linkages are shown in the visualisations. Note that the vertical and horizontal coordinate locations of the authors in the visualisation reflect the full collaboration network: this way, natural collaboration clusters (marine-terrestrial clusters; remote sensing cluster etc.) were maintained.

Fig. 31.9

Networks of SPACES (II) researchers with North-South co-authorship links (i.e. publishing with a Northern or Southern SPACES (II) research partner) on the 2009–2012 (above); 2013–2016 (middle) and 2017–2020 (below) publications. Size of circles indicates total link strength (i.e. authors with more co-authored publications show larger). South-South and North-North collaboration links have been removed from the visualisation

During the years 2009–2012, only 9% (20 authors) were involved in North-South co-authored publications with other SPACES II researchers. Within 2013–2016, the network had broadened to 41 researchers (19%), and in 2017–2020, nearly half (103) of the researchers had published with a Northern or Southern SPACES II co-author. Nearly all authors contributing most to NSC co-authorships were Namibian or South African: during 2009–2012 B Strohbach (3.0) and H Verheye (3.0); during 2013–2016 M Meadows (3.8), A van der Plas (3.6) and K Kirsten (3.5); and during 2017–2020, I Smit (6.5) and I Grass (5.0). The strongest co-authorship links were between H Verheye and W Ekau (1.5) and H Verheye and W Hagen (1.5) (2009–2012); B Bookhagen and T Smith (2.0) and V Morholz and A van der Plas (1.5) (2013–2016) and B Bookhagen and T Smith (4.0) and I Grass and S Weier (2.5) (2017–2020) (Fig. 31.9). Notably, most of these are senior researchers and none are early-career researchers. When considering all NSC collaborative publications, German collaborators dominated lead authorship during all three time brackets. During the years 2009–2012, 52% of collaborative papers were led by a German author, whereas 26% had a southern African lead author. During the years 2013–2016, the difference was proportionally slightly smaller, with 51% German-led, and 34% southern African led publications. During the last bracket 2017–2020, 47% of the collaborative publications were German-led and 27% by a southern African lead author.

Specific observations can be made when focusing on the clusters that indicate topical collaborations of SPACES II researchers (based on broad definition of the researchers’ main field). The oceanography researchers cluster publishing particularly on the Benguela upwelling system, including A Van der Plas and D Louw (Namibia), W Ekau (Germany), H Verheye (South Africa) and others, exists before SPACES and stays relatively stable throughout the research programme. In the last time bracket (2017–2020), this cluster is larger and linked to others, with P Brandt (Germany) and M Rouault (South Africa) as significant contributors (see violet, orange, light yellow and pink clusters). Furthermore, there is a topical cluster with focus on earth observations and related topics, composed of prominent researchers like B Strohbach (Namibia), C Schmullius and U Gessner (Germany), R Mathieu, and I Smit (South Africa), with others. This cluster grows and gets connected to (agri-)ecosystems, in the last time bracket (see red, light red and bright yellow in 2017–2020). The vegetation and ecosystems (modelling) cluster, involving G Midgley and J Slingsby (South Africa), with S Higgins, S Scheiter, T Hickler and F Jeltsch (Germany), exists before SPACES but extends further during the observed periods, linking with other terrestrial research. The paleo environmental researchers collaboration, including especially M Meadows, K Kirsten and L Quick (South Africa) with T Haberzettl, R Mäusbacher (Germany) becomes prominent during the second time period (2013–2016) and grows into a set of clusters with more collaborators during the last time period (see turquoise and green clusters in 2017–2020 graph). Few clusters appear only when SPACES II collaborations are elaborated. However, the agroforestry-related cluster (blue in 2017–2020), including P Chirwa (South Africa), J Sheppard, L Borrass, H.-P. Kahle, C Morhart (Germany) and others, seems to appear only on the third studied bracket, indicating that with this specific topic, the NSC was initiated as result of SPACES II.

3.4.2 Institute Co-authorship Network

Figure 31.10 presents the institute co-authorship overlay visualisation constructed from all publications authored by SPACES II researchers during 2009–2020. Each circle represents an institute, and its size indicates the total link strength (strength of the co-authorship links of the institute with other institutes in the network). Lines amongst institutes represent co-authored publications; thicker lines indicate more co-authored publications, however, multi-author publications are given less weight. The colour of the circles indicates the average publication year of the institute. To highlight the evolution of NSC within SPACES II, only linkages between institutions from Germany and Africa are shown in the visualisations. Coordinate locations of the institutes reflect the full collaboration network.

Fig. 31.10

Network of SPACES institutes (both SPACES and SPACES II) with North-South co-authorship links (i.e. publishing with a Northern or Southern SPACES II research partner) during 2009–2020. Size of circles indicates total link strength (i.e. institutes with more co-authored publications show larger, however fractional counting reduces the influence of multi-author publications). The colour of the circles indicates the average publishing year. South-South and North-North collaboration links have been removed from the visualisation

The vast majority of the SPACES II institutes (63 out of 66, 95%) were involved with at least one NSC collaborative publication (Fig. 31.10). The strongest collaborating institutes were the UCT (South Africa) (total link strength 56.2), FSU Jena (26.9), and the Senckenberg Research Centre (20.0) (Germany). The strongest links were between UCT and GEOMAR (collaboration link strength 14.8), UCT with Goethe University Frankfurt (6.5) and UCT with FSU Jena (6.5). Notably, although several Namibian authors came up as main collaborators in the individual analyses, no Namibian institute made it in the group of the 10 strongest collaborating institutes. The first Namibian institute, the MFMR, was the 12th strongest collaborating institute, with a total link strength of 12.

The average publication year score indicates that publications from FSU Jena, CSIR, and the University of Hamburg came out on average a bit earlier than those from e.g. the UCT, Goethe-University, University of Göttingen, Senckenberg Research Institute, Stellenbosch University, and the University of Kwa-Zulu Natal.

3.4.3 Relative Proportion of Collaborative Publications

We further expected that the impact of SPACES I and II would show as an increased proportion of North-South co-authored publications over time. Using the same set of SPACES I and II authors publications and counting the total number of publications per country, those publications where at least one author’s affiliation was a German university or institution were considered ‘German’. Those publications where at least one author’s affiliation was an African university or institution (including South Africa, Namibia, Angola, Botswana, Lesotho, Mozambique, Swaziland/Eswatini, Zambia, Zimbabwe, and SASSCAL) were considered ‘African’. Figure 31.11 shows that the proportion of collaborative publications peaked in 2018, probably indicating the higher number of project publications that came out at the end of the SPACES I projects.

Fig. 31.11

Number of SPACES II researchers’ total publications, North-South ‘collaborative’ publications and the relative proportion of collaborative publications for the years 2009–2020 as obtained from Scopus

4 Discussion

Having presented a wide spectrum of results above, in the following subsections, we take a closer look at certain issues uncovered or raised that we consider as essential to address when designing future NSC. In this respect, we first reflect upon the (financial) resources endowment of the Southern partners, as the majority of our respondents indicated the need for improved funding schemes. Second, we discuss the way Southern partners were involved in project design and continuous management as a significant difference in viewpoints between the North and the South could be uncovered. Third, we explore the results of the co-authorship analysis to discuss the potential of collaborative publications as a key tool for effective NSC. Finally, we critically reflect upon the applied methods and the representativeness of our results.

4.1 Resources Endowment of Southern Partners

Both the quantitative survey results as well as the qualitative statements of our respondents are critical of the resources endowment of the Southern partners, especially with respect to monetary means (cf. Sects. 31.3.2.2 and 31.3.3.2). The observation of an imbalance in funding is shared by researchers from the North and the South. Comparable results were found in other meta-studies that looked into NSC programmes elsewhere (Giller 2020; Skupien and Rüffin 2020; Schneider et al. 2019).

Formally, the SPACES II programme was set up as a cooperative funding scheme together with the South African Department of Science and Innovation (DSI) and the Namibian Ministry of Education (MET) (BMBF 2017). With both countries, Germany had developed pre-existing arrangements to enhance collaborative research. According to the funding regulations, which are legally bound by the German federal budget law, only German institutions were entitled to apply for research grants awarded by BMBF. African partners had to apply for respective linked grants at their home country’s collaborating partner institution (DSI or MET), had to be subcontracted by their German partners for particular service-based activities or had to contribute in-kind or non-monetary means (BMBF 2017). This model applied to all project research activities, whilst support for student exchange and mutual research visits by partners was enhanced through an exchange and researcher mobility programme established by the German Academic Exchange Service (DAAD) for which both African and German students and researchers could apply.

These funding formalities do not appear to have been sufficient to establish an adequate resources endowment of the Southern partners to conduct research and ensure sufficient researcher mobility, as the experiences of our respondents indicate. This imbalance became exacerbated during the COVID-19 pandemic, when travel restrictions (Devi 2020) inhibited German researchers from carrying out activities in South Africa and Namibia. According to the respondents, African partners stood in to replace these functions, without being fully compensated for their efforts, though the situation was somewhat alleviated by flexible responses in funding arrangements. Overall, the COVID-19 pandemic can be considered a caesura for project collaboration, but it might also provide a window of opportunity to enhance NSC (Jeppesen and Miklian 2020). Despite the drawbacks that our respondents state with respect to the limitations of meetings in person, we particularly see the potential of digital communication tools to enhance the quality of collaboration in a post-pandemic world. Respective tools like video conferencing, data sharing and project management can act complementary to standard project tools such as annual meetings, phone conferences and research visits. In this regard, Sowe et al. (2021) found an increased utilisation of digital communication and collaboration tools amongst Northern and Southern research partners during lockdown periods with certain challenges in particular for Southern researchers like the availability of reliable power and internet connections (Sowe et al. 2021).

In order to improve the funding situation of Southern researchers in NSC programmes, potential improvements can be explored in (1) more concrete multilateral agreements for securing research funds prior to programme start, (2) equal treatment of Southern and Northern researchers in terms of (staff, material, travel) funding to conduct research as well as (3) the reduction of structural institutional barriers for Southern researchers which prohibit southern partners from getting the same incentives as their northern counterparts from third-party grants (e.g. lack of research incentivising policies in some Southern universities). Furthermore, we consider multilateral independent institutions such as the Southern African Science Service Centre for Climate Change and Adaptation (SASSCAL, Helmschrot and Jürgens 2015) as a blueprint. It is intended to be capable of funding research projects based on its own agendas and with its own monetary means. If such an institution is successfully set up and operated, efforts in this direction could provide an effective model to overcome respective funding imbalances.

4.2 Involvement in Project Design and Management

Our survey revealed significant differences amongst the views from Northern and Southern researchers with respect to their involvement in project design and ongoing management, as well as fieldwork decisions (cf. Sects. 31.3.2.1 and 31.3.2.2). As these findings might indicate a top-down approach from the Northern to the Southern partners—possibly caused by funding imbalance—we consider the following insights as crucial in order to enable effective, balanced and equal NSC.

The initial steps in project design are known to be particularly crucial for a fruitful and collaborative atmosphere and thus an overall successful project. Referring to transdisciplinary research in this regard, the process of ‘problem framing’ is essential in which perceptions of scientific disciplines and non-scientific stakeholders are brought together (Jahn et al. 2012). Almost half of the respondents from the South indicated that they were not adequately involved in these processes, which may relate to the long time lag of more than a year between the participatory process that involved workshops with scientists and funding representatives, and the announcement for grant applications and their award. On the contrary, these perceptions are also likely due to ongoing staff turnover and the subsequent lack of awareness of these original discussions by those who joined the projects later or were not in a position to participate in respective processes (e.g. PhD students). Nonetheless, this result confirms to a degree previous findings on the difficulties of Southern researchers to participate in the processes of agenda setting (Bradley 2017). Ideas to overcome an inadequate involvement of Southern researchers in the agenda-setting process could be found in an explicit initial project phase for collaborative problem framing via ‘quick-initiation-funding’ schemes (Luthe 2017).

In terms of involvement in overall project administration and management, the discrepancy between Northern and Southern researchers continues. In both fields, Southern partners indicated lower levels of involvement, as confirmed by other studies previously (Luetkemeier et al. 2021; Schmidt and Pröpper 2017), which might be a direct consequence from being less involved in project design. Though we could assume that heavy involvement in project administration and management is somewhat time-consuming and even burdensome, it affords power to influence overarching strategies. Hence, being not involved at these levels limits the subjective feeling of ownership. Likewise, Southern researchers indicated that they had not been involved in fieldwork planning or practical project decisions, largely. This is again crucial, because it could even be counterproductive, as important Southern knowledge stocks remain untapped (i.e. environmental and institutional setting). Excluding them from decision-making processes therein may hamper the effectiveness of actual research tasks.

Overall, the evaluated projects within SPACES II show a heterogeneous degree of involvement. Whilst some projects explicitly applied a ‘tandem structure’ in lead positions, in other projects, the role of African collaborators was less obvious. Tandem-structuring the leadership of work packages within a project can be a tool to balance opportunities and responsibilities for Northern and Southern partners. However, having two leaders might also complicate the decision-making process. Furthermore, especially if funding comes from the North, the actual leadership might remain in the hands of the Northern partner. One option to circumvent these challenges could be to keep the leadership of each work package to one person only, either coming from the South or North and in any case equally balanced across all work packages. Alternatively or in addition, a Steering Committee can be an adequate tool to foster joint decision-making within the collaborative research project.

The aspect of under-valuation may also be reflected in the answers of our respondents to the question, if in-kind or non-monetary contributions were adequately acknowledged (cf. Sect. 31.3.2.3). Here, 42% of the Southern respondents indicated that their contributions were not adequately appreciated. This could possibly lead to dissatisfaction and even frustration amongst Southern partners, as found in a qualitative evaluation of two German-Namibian collaborative projects by Luetkemeier et al. (2021). The imbalance in involvement and appreciation of contributions could likely result in unclear responsibilities and research tasks as well as skewed workloads.

Interestingly, our survey does not support a qualitative finding from Luetkemeier et al. (2021) that Southern researchers were less involved in actual empirical research, but rather in activities such as field logistics, stakeholder communication or student supervision. Our results do not show a significant difference between the tasks and workloads of Northern and Southern researchers, indicating that Southern partners did not implicitly have to act as ‘field facilitators’ of Northern researchers. Both Northern and Southern researchers have been equally involved in empirical research, which we consider a positive sign—apart from the considerations above—for a truly cooperative research project.

4.3 Co-authorship as a Key Tool for Effective Collaborations

When observing all SPACES II researchers’ collaborative publications via bibliometric networks, the main contributors were German (senior) researchers, and strongest links were formed within organisations. These relationships remained stable, probably indicating that the SPACES II projects were largely established on existing collaborations within Germany. The total network of collaborating authors grew rapidly with time, as researchers were taking up collaborative publication activities with their SPACES II colleagues. Similar growth was observed when limiting focus to NSC co-authored publications. Prior to SPACES, less than 10% of authors were publishing together with their Northern or Southern partners, whilst immediately before and during SPACES II (2017–2020), this had increased to a half. This indicates that SPACES II was successful in meeting its objective of improving the networks of NSC, when using co-authored publications as a measure. On the other hand, whilst during the last time bracket, the vast majority of researchers (82%) co-published papers with SPACES II-colleagues, only a half published with their Northern or Southern partners.

In the light of earlier studies, the proportion of NSC publications amongst SPACES II researchers (12% during the ‘best’ year of 2018) appears relatively low. Pouris and Ho (2014) found that in 2011, 58% of papers published on the African continent were co-authored with international partners; Germany was the fourth most important international collaboration partner. Access to funding and equipment were assumed to be the major reasons behind the high proportion of South-North co-authored papers in Africa, and also main factors explaining the comparatively low numbers of collaborative publications between southern African countries (Pouris and Ho 2014; Zravkovic et al. 2016). In the case of SPACES II, we can assume that a higher proportion of NSC papers would have been recorded especially from the German side if only those publications officially published ‘under SPACES II’ were counted. In particular, the senior researchers were undoubtedly involved in various other (non-NSC) projects. Furthermore, the COVID-19 pandemic and the cancellations of the associated researcher mobility programme may have decreased the overall number of co-authored NSC publications from SPACES II. For example Zravkovic et al. (2016) found that southern African researchers, who left their home country temporarily for scientific education and training (e.g. master/PhD studies), also published more NSC papers.

In contrast to the overall network, the NSC co-authored publications network was dominated by Southern researchers as the most productive, best-linked researchers. It appeared that the NSC papers were more focussed on fewer, very active collaborators particularly on the Southern side. Similarly, a few South African institutes contributed disproportionately to the NSC co-authorship network. The strongest collaborator, UCT, is both the highest-ranked South African university, and the highest-ranked African university on the main global university rankings (rank 269 on the Center of World University Rankings 2021–2022 Edition; rank 155 on the Times Higher Education University Rankings 2021; and rank 226 on the QS Top Universities Ranking of 2022). This trend may have two distinct reasons: First, most fully funded project researchers (i.e. PhD and post-doctoral researchers) are German-affiliated, including southern African PhD students funded with 4-year PhD grants, each publishing some NSC papers. Second, the largest and wealthiest institutes on the Southern side may have longer history and better preparedness for international collaborations and thus outperform smaller institutions in the South. The question of how this may be a result of or may even consolidate power asymmetries in research amongst Southern partners goes beyond the scope of this study but can be a valid entry point for further investigations of South-South collaborations.

Besides bibliometrics of published articles, it is relevant to look at the process leading to co-authored papers. The question of authorship is often a difficult and disputed topic in research collaborations (see, e.g., Luetkemeier et al. 2021). It would be an unwanted outcome if Southern collaborators felt themselves not invited to co-author in publications where they would have had a clear role. Previous studies have raised concern over North-South collaborations where Southern partners were confronted with implicit expectations of acting as facilitators of communication to stakeholders and policy-makers, or even field managers (Luetkemeier et al. 2021).

4.4 Critical Reflection on Methodology

Against the background of our experiences in carrying out the survey, we consider a structured online survey as a suitable method to collect perspectives from researchers in the field of Earth system sciences. We consider them as being open to such a survey format. Deliberately, we assembled an author team of Northern and Southern researchers to keep a balanced view on relevant NSC issues that should be part of the assessment. Furthermore, we designed the survey in a way to make the barriers for participation as small as possible. In this regard, we consider the available period of about a month and the required time for completion of only 10 min as a good compromise. In addition, we assume that the inclusion of both quantitative and qualitative questions offered the researchers a suitable tool to share their views on NSC in a comfortable and anonymous way.

Although we consider the response rate of about 31% as being adequate, the critical question is: why did not more researchers take the opportunity to participate? We can only make assumptions but reasons could be found in an incomplete or a non–up-to-date list of contact details, holiday season in the North in July, low interest in the subject, low involvement in practical NSC (e.g. on a student level) or frustration and resignation about the topic itself or current framing conditions (e.g. COVID-19 situation).

To better evaluate the representativeness of the survey, it should be noted that the respondents evaluated their collaboration with all their Southern or Northern partners, respectively. However, it could likely be the case that they actually had different experiences with different partners. Hence, their actual responses either may be an averaged assumption about all their partners, or may be biased by a certain single case. Furthermore, the results we presented are averaged over all SPACES II projects. It is likely the case that the expectations towards NSC differed amongst the projects, depending on the actual research subject, the required methodology and, hence, (dis)satisfaction might differ.

With regard to the co-authorship analysis, we must note that due to COVID-19, the publishing of project results was, in many cases, delayed, and due to the normal publication time lag, the proportion of NSC authors may still significantly increase after the publishing of this chapter. In addition, the Scopus database does not necessarily cover all publications, but was selected here because of the availability of institute affiliations data.

5 Policy Message

Handling Global Change phenomena such as climate change, biodiversity loss and environmental degradation, on the one hand, and approaching the challenges of poverty reduction and enhancing water and food security in the developing world are amongst the most crucial challenges today. SDG Goal 17 highlights the important role of international collaborations to develop applied solutions for these problems. Effective NSC in science is one key component therein.

In this study, we assessed the quality and structure of NSC in the research programme SPACES II based on a structured online survey that was carried out amongst the involved researchers from South African, Namibian and German institutions as well as an extensive co-authorship network analysis. In synopsis of our results, we consider the SPACES programme as a good step forward in carrying out NSC, because Northern and Southern researchers acted largely as peers on an equal footing. Despite this success, room for improvement exists, in particular, with respect to the integration of Southern researchers in the processes of project design and continuous management as well as the equal availability of funds for research. In particular, the latter point turned out to be crucial for successful NSC as most challenges, we identified, likely, have their origin in unequal resources endowment. Against this background, we recommend the following key points to enhance NSC for future research programmes:

• Asynchrony of funding: Research requires funds to cover costs for staff, equipment and mobility for all actors involved in a project. The persisting asynchrony of funds available to Northern and Southern partners is, however, a fundamental area of conflict and frustration and hence a potential threat to effective NSC. If collaborative research initiatives are an explicit intention by funders and research consortia, appropriate and sufficiently timed monetary means are required for all parties involved to carry out the intended research. We see four options to getting closer to this desirable state: First, multilateral negotiations between funders prior to an NSC programme should put emphasis on legally securing sufficient funds from all countries involved. Second, multilateral funding agencies based ‘in the South’ should be (financially) supported to enhance the opportunities for independent research funding. Third, funding regulations should acknowledge the financial requirements of Southern researchers, for example, via direct provision of funds or explicit subcontracting opportunities. Fourth, remaining institutional barriers in Southern institutions that prohibit researchers from benefitting equally from third-party grants should be identified and reduced.

• Continuous NSC evaluation: We consider an investigation like ours not just of interest for final project evaluation. If a large number of scientists are involved in a collaborative research project, especially comprehensive quantitative metrics could well be used to periodically monitor the quality of NSC within a research programme or project. Hence, we suggest evaluating partnership expectations and perceptions in the project beginning as well as in the mid-term to immediately respond to dissatisfaction amongst partners that might negatively influence the achievement of project goals and lead to the abortion of existing or rejection of future collaborations. Therefore, we suggest monitoring the number of involved researchers as an important metric, especially from the South, where partners are often not formally funded. In addition, conducting surveys on the researchers’ changing perceptions throughout the project concerning levels of involvement and opportunities to engage can quickly uncover negative developments and enable countermeasures. Furthermore, monitoring of continued co-publishing activities of the involved researchers (even after formal NSC programmes ended) may provide an effective indication of the quality and long-term impact of collaborative projects. Although we did not investigate the ratio between female and male project members within this study, we strongly recommend including gender balance in the evaluation of collaborative research projects. This includes the share of women in the overall project as well as their position (e.g. as leaders) and differences or similarities between Northern and Southern partners. Furthermore, knowledge gained over the last decades on transdisciplinary science with respect to formally setting up interdisciplinary projects in a methodologically guided way can be fruitful to structure future activities.

• Collaborative publications: Though alternative outputs and outcomes of research, such as societal impacts, are increasingly considered relevant when evaluating the quality of research, scientific publications remain the key currency, due to their quantifiability and associated metrics of researchers’ performance. In this regard, publications are a central tool to enhance collaboration between Northern and Southern researchers and our analysis shows that an intensification did emerge over time. However, we see a need to enhance this tool, as in particular on the Southern side, co-authored publications tend to be disproportionately focused on a few established, senior partners. Here, respective successes are strongly bound to available funding for Southern researchers to be able to invest time and money in respective authorship processes.