The Anthropocene requires a more holistic, critical and systems understanding of our relationship to and with the earth surface. We now live in an era of shifting baselines, where all components of the Earth surface – from black carbon enhancing snowmelt and runoff on mountain tops to levees accelerating erosion and transfer of sediments off continents into the deep ocean – are responding to humans as geomorphic agents and to the cascading effects of anthropogenic climate change. We also now live in an era of shifting assumptions: the more we look to reconstruct or restore a ‘pristine’ landscape free of the imprint of human activity, the more we find such a place has not existed for as long as we have been observing and cataloguing the land, due to centuries of activities such as controlled burning of forests, grazing, or use of the plow. As so eloquently put by Tsing et al. (2017), if one looks closely ‘every landscape is haunted by past ways of life.’ There was a time when searching for pristine landscapes made sense in the context of finding a baseline against which to check the effectiveness of our conceptual models of accelerated process regimes, or to find engineering solutions to human-induced events. However, as Lave et al. (2018) argue “most landscapes are now deeply shaped by human actions and structural inequalities around race, gender, and class [which] incorporate the materiality of nature, creating inextricably eco-social systems. Thus, it no longer makes sense (if it ever did) to concentrate natural science research on pristine systems.” Can our focus on mechanistic/reductionist thinking keep up with the added complexity we need to embrace to help us understand these evolving human-earth systems? 

Moreover, is prediction even viable in the Anthropocene? Given all the uncertainties in both magnitude and timing of forcings and feedbacks between human activity, climate and the earth surface, can we be confident that our understanding of process-responses in the past are indeed the key to the present, let alone the future?  What is the role of models in this era of exceeding planetary boundaries and tipping points (Steffen et al., 2015)? There are limits to the principle of uniformitarianism in this era of shifting baselines. What we need now is a paradigm shift, moving away from predictive, inductive logics and positivist approaches, and towards a fuller embrace of non-equilibrium, emergence and systems thinking. Chaos theory, for example, provides a radically different framework for studying system dynamics, highlighting the limitations inherent in reductionist (towards a steady state) and mechanistic (linear cause and effect-based) analysis of complex earth systems (e.g., Phillips, 1992; 2003, 2011; Murray et al., 2009, Kieler, 2011; Shuster & Just, 2005). What would happen if we were to flip our conceptual models of landscape function and start from an assumption of open systems in disequilibrium, rather than from an assumption of (dynamic) equilibrium and conditions of steady state?

To thoughtfully grapple with considerations of both complexity and causality in landscapes in this new age of uncertainty, we believe the geomorphology community should continue to cooperate with and learn from other disciplines like evolutionary biology, ecology, even neuroscience and epidemiology. Where we once looked to geologists and geographers for ontological framings, followed by leaning on physicists and engineers to distill and quantify the world, we should now be looking to ecologists and climatologists who are throwing out notions of steady state and embracing complexity and post-modern synthesis as foundational frameworks for understanding. For instance, the field of evolutionary biology has started to embrace a radical rethinking of symbiosis, beyond the Darwinian notions of discrete, bounded units, to explain the interactions of animals and their microbiomes, algae and reef-building corals, mycorrhizal fungi and plants (e.g., Margulis, 2008; Gilbert et al., 2012; McFall-Ngai et al., 2013). Perhaps we should start to think of a landscape in the same vein, as a sympoietic organism of interconnected processes and histories (as defined by Haraway (2016)). One arena in which this type of systems thinking is happening is in Critical Zone Observatories (CZOs), in which geomorphologists incorporate interdisciplinary and complex systems approaches that include geochemistry, ecology and systems biology to study the interactions of surface processes and the biosphere (e.g., Anderson et al., 2008; Dethier & Lazarus, 2006; Riebe et al., 2015). Understanding the complex interactions between processes that operate in the Critical Zone, from the canopy to the water table, requires integrated and interdisciplinary approaches that span the granular to the watershed scale. These reframings of inquiry should be applied more broadly across landscapes, beyond the ‘critical zone’ defined by the contemporary weathering front. We should also be asking ourselves if we have the right statistical frameworks to quantify our understanding of landscape complexity. For instance, in the field of ecology there has been renewed interest in using Bayesian approaches and multiple model pathways to analyze systems in which multiple causative factors lead to real-world complexity that is difficult to reduce to a single, isolated mechanism (e.g., Hilborn & Mangel, 1997; Elliott & Brook, 2007). Our mental models of landscape change should embrace these new forms of synthesis and statistical inference, and are beginning to (e.g., Fox et al., 2015; Chandra et al., 2019).

 Other, more holistic, theories are needed to inspire the field, including indigenous and place-based knowledges of the land and our role in shaping it. We need to revisit how we are privileging western science over other ways of seeing and being, particularly as many of the communities that are most impacted by the changing landscapes of the Anthropocene are currently the least involved in guiding the scientific effort. Here, it would be nice to provide a compelling list of examples of what this might look like. However, with the exception of recent calls for the practice of ethno-geomorphology and roadmaps for respectful, bicultural engagement between geoscientists and indigenous knowledge holders in Canada, Hawaii and New Zealand (e.g. Brierley et al., 2018; Kochan, 2015; Swanson, 2008; Wilcock et al., 2013; Wilkinson et al. 2020), much of this work remains to be done. Instead, we invite the earth surface. community to imagine how we might frame new questions that require integrating new (and old) ways of seeing the land and considering landscape change. What is the geomorphic imprint of colonialism, for example (Koch et al., 2019)? Or of white supremacy in North America (Pulido et al., 2019)? What if we start from an indigenous First Nations or Maori perspective that rivers and glaciers are kin and have agency (Cruikshank, 2012; Suzuki & Knudtson, 1992; Wilkinson et al., 2020)?  We are not advocating for any approach in particular. We simply draw attention to the multiplicity of worldviews possible (and necessary) for achieving a more holistic understanding of the Anthropocene, and the ways in which these worldviews have been excluded over the last 70 years of data-centered disciplinary evolution. 

 To truly expand the breadth of the geosciences, we need to interrogate the hierarchies of knowing that our current scientific institutions sustain (King & Tadaki, 2018). This involves acknowledging that the current science of landscapes does not prioritize the perspectives of the Global South, and that we have known this to be true for decades (Kiage, 2013; Stocking, 1995).  Could it be that a fixation on more of the same data at higher and higher resolution is not compatible with the questions relevant to the communities most impacted by landscape change in the Anthropocene, or to the methods most suitable or accessible for conducting research that is relevant there? Optimistically, we hope that asking a greater breadth of questions might also help improve on the diversity crisis that plagues the earth sciences (Bernard & Copperdock, 2018; King et al. 2018; Popp et al., 2019). 

We are now more aware than ever that every landscape is a complex concept with a complex history, that requires both global and local frameworks of inquiry (Garcia-Ruiz, 2015). We need to revisit how we understand landscapes from more than one perspective in time and space. As Sir Ron Cooke suggests in the preface to Roy and Trudgill (2014): “Process is generic – water really does usually, but not always, flow downhill, and the pebble in a stream is battered by predictable forces. But location is unique, and specific landscapes always require interpretation that depends in part on the unique consequences of unique historical events: it is for this reason that what is observed so often conflicts with what is known, and new questions that inspire research come flooding in.”  To bring about these new questions that will inspire our research in the next century, perhaps we need to consider shifting some resources away from the data-gathering effort and towards interdisciplinary and international collaborations, seeking to integrate perspectives not only from the international, heterogeneous scientific community, but from local and indigenous understandings of the landscapes we are working in.