In many ways, the tropical dry forest biome is the ‘invisible’ forest. Seasonally dry tropical forests (SDTFs) are found in the Americas, Africa, Asia and Australia, but they are not defined by a single physiography and the biogeographic origins of this biome are still under study (Dexter et al., 2015; Slik et al., 2018). In Latin America there is a wide array of names given to this forest type such as ‘bosque chiquitano’ in Bolivia, ‘chaco’ in Argentina or ‘caatinga’ in Brazil, while some SDTFs are referred to as miombo in Africa, and in Vietnam the deciduous dipterocarp forests are colloquially called ‘rừng khộp’ (‘rừng’ means forest and ‘khộp’ sounds like the dry leaves when you step on them). In many people’s imaginations and certainly in the popular press, tropical forests are dark, rainy and entangled ecosystems. In other words, many people equate tropical forests with rain forests. In fact, some people who live in lands that once supported SDTFs are unaware that their cities and towns used to be dry forests. Similarly, most young students from tropical and temperate latitudes alike who gravitate towards tropical forests get their start in rain forests. Compared to wet and moist tropical forests, SDTFs have lower annual rainfall and experience several months each year with little to no rainfall (Murphy & Lugo, 1986). As such, the structure, function, and diversity of these forests differs from other tropical forests with higher or lower seasonality (Fig. 1). SDTFs also contrast with savannas, which are characterized as mixtures of grass and trees with discontinuous canopy cover, and are maintained by fire (Dexter et al., 2018). Centuries of clearance for pastures and agriculture have greatly reduced the extent of SDTFs (Miles et al., 2006) to the point that they are considered one of the most threatened tropical biomes, with an estimated 45.8% of the biome converted to alternate land uses and a mere 7.6% of existing habitat under protected status (Hoekstra et al., 2005). However, as the dry forest biome is shrinking, interest in this diverse, threatened, and heterogeneous biome is growing (Pizano & García, 2014). On May 8, 2021, we organized an informal Virtual Symposium to share recent research and emerging themes related to the critically endangered SDTF biome. The Symposium was advertised through a flier and an email that we shared with our colleagues and professional networks. We tracked registrations and participants’ interests through a Google form. We were overwhelmed with the response. Our Symposium, which was free and conducted in Spanish, attracted 100+ participants from Colombia, Costa Rica, Brazil, the UK, Europe, Puerto Rico, Mexico, Germany, Ecuador, and elsewhere. Our participants also represented diverse career stages and institutions – from undergraduate students, post-baccalaureates and Emeritus faculty, to employees from land-management agencies and people who work for conservation-oriented nongovernmental organizations. We believe that this Symposium demonstrated a great interest in the science, management, and conservation of SDTFs. This meeting also highlights a promising new way to foster communication and collaboration among diverse groups of people, who may not necessarily participate in scientific meetings organized by professional societies, because of financial constraints, language barriers, concerns over the carbon footprint of travel, and/or the inability to get away for several days. The content and format of the meeting was guided by participant interests recorded on the registration forms. We used Zoom to host large sessions of 30 5-min lightning talks, break-out rooms where participants were assigned to small groups at random for ‘icebreakers’, and for thematic sessions. Participants were asked to record their talks in advance. Speakers found that having pre-recorded their talks allowed for time to answer questions directly in the chat when they arose. We selected 11 topics in advance for the thematic breakout sessions based on participants’ interest (Fig. 2). The topics were: (1) mitigation and adaptation to climate change; (2) biogeography; (3) disturbance; (4) biogeochemistry; (5) forest products and seed resources; (6) belowground ecology; (7) restoration; (8) functional ecology; (9) forest dynamics; (10) forest structure; and (11) management and conservation. Collectively, these activities provided a rapid assessment of the current understanding of dry forest ecosystems and point towards promising future directions. Later we describe the main themes and take-home messages that emerged from the Symposium. The talks highlighted that we do not yet understand how different manifestations of climate change (rain, drought, temperature, winds, fires, hurricanes, etc.) interact and influence community and ecosystem dynamics, or how to incorporate these concerns into current and future management plans for mitigation and adaptation to climate change. Six main research priorities emerged to move forward both research and management agendas: (1) investment in long-term monitoring to understand responses to extreme/rare events; (2) studying interactions with soil microorganisms, herbivores and seed dispersers; (3) incorporating ‘hard’ traits (i.e. physiological, biochemical) to understand species performance and ecosystem processes; (4) understanding the effects of changing disturbance regimes – especially fire and drought – on forest dynamics; (5) understanding how climate change may shift species’ distribution and hence alter the boundaries of SDTFs; and (6) understanding biogeochemical cycles and how nutrients mediate forest responses to climate change. Many of the talks and discussions centered around the related themes of reducing forest conversion, regeneration through secondary succession, and, most importantly, restoration. The focus on restoration likely reflects the reality that much of the original extent of SDTFs in Latin America is gone, and what remains are isolated fragments that face a multitude of threats. Moreover, countries such as Colombia now have legal mandates to restore from between 10 and 20 ha of dry forest for every hectare that is lost through deforestation. There are unique challenges to restoring SDTFs compared to wet forests, mainly related to rainfall seasonality as it can impose severe hydrologic stress on planted seedlings. Current knowledge gaps include basic seed biology and how to select viable species for restoration based on physiological and hydraulic traits. Beyond technical challenges, the audience acknowledged that successful restoration activities must be situated within human communities, and that the social context of restoration is as important as the biophysical one. The great diversity of plant forms and functions in SDTFs has given way to the evolution of a wide range of strategies for coping with drought. Characterized by a diverse array of phenological leaf habit strategies (e.g. evergreen, deciduous, brevi-deciduous and semi-deciduous), plant species in SDTFs have attracted lots of attention from eco-physiologists. Recent studies have highlighted how hydraulic traits underpin responses to acute drought, in contrast to ‘soft’ traits such as wood density (Powers et al., 2020; González-M et al., 2021). Also, we need to focus on intra-specific trait variation, rather than on inter-specific variation where only mean trait values for species are considered. As SDTFs encompass high climatic and soil heterogeneity (Allen et al., 2017; Waring et al., 2021), species’ responses to these gradients may depend more on individual's performances across the extent of species’ biogeographical ranges. Greater understanding of the covariation between statistical moments in trait distributions and life-history traits (e.g. height or leaf life span) will further inform how species respond to extreme weather events. Despite decades of neglect, the tide is turning for research in SDTFs. There are now robust networks of collaborations including DRYFLOR (who curate data on plant species distributions), the pantropical SECO project aiming to map and model carbon dynamics in SDTFs, and the Tropi-Dry project that leverages remote sensing to advance conservation. In particular, networks of forest plots that are monitored over time provide a platform for generating basic ecological knowledge of forest dynamics that can form the basis for comprehensive and sustainable management (Norden et al., 2021). This is the case of the Red BST-Col network established in 2014 in Colombia, for which permanent forest plots were established across the country, where SDTFs occur (González-M et al., 2019). While each of these networks has its own mandate and mission, strengthening the communication and collaboration among these networks and explicitly fostering links with practitioners (e.g. land managers, conservationists, foresters, restoration ecologists, and people involved in carbon markets) will help ensure that dry forests become more visible to society. This was our first attempt to unite dry forest enthusiasts, and we were delighted by the high turnout. This Symposium provided an opportunity for networking, and multiple collaborations have already emerged from it. That said, there were conspicuous omissions from the themes the Symposium covered, including very few talks on fauna and their role in dry forests, best conservation practices, environmental policy, ecosystem services, and the human dimensions of SDTFs. We hope that this event will inspire participants to replicate and extend this effort. We look forward to strengthening the collaborations to develop a joint agenda with the identification of priorities and standardization of protocols at different levels: research, policy, management and conservation. SDTFs are extremely fragmented, but our main interest is that the same does not happen to the people who study, manage, and care about them.