Focus on tropical dry forest ecosystems and ecosystem services in the face of global change

Tropical dry forests are distinct from wet and moist tropical forests by the presence of a strong dry season. This collection of papers explores the unique biodiversity, plant functional traits, coupling between carbon and water cycles, and threats to these important ecosystems. These studies have relevance for conservation and management of tropical dry forests.


Introduction
Historically, seasonally dry tropical forests (SDTF) have received less attention from the research community than tropical moist or wet forests (Meli 2003, Sánchez-Azofeifa et al 2005. The 23 papers in this ERL Focus Issue help close that gap and advance our understanding of these ecologically and economically important forests. SDTF are distinguished from tropical moist and rain forests by lower annual rainfall and more acute dry seasons (Murphy and Lugo 1986). These highly seasonal rainfall regimes contribute to the large diversity of leaf habits present in SDTF tree species (Eamus 1999), and the broad variation in canopy phenology ranging from forests dominated by evergreen trees to those dominated by deciduous trees (Vico et al 2017). Present in the Americas, Africa, and Asia, SDTF are threatened by a number of anthropogenic forces including land use (Portillo-Quintero and Sanchez-Azofeifa 2010, Dupin et al 2018) and climate change (Allen et al 2017). The topics addressed by the papers in this Focus Issue advance the state of knowledge on the structure, function, and threats to these forests, and can loosely be grouped by theme into patterns of biodiversity, plant functional traits, carbon cycling processes, and forest responses to anthropogenic pressures. The geographic scope of the papers in this Focus Issue is similarly broad, including Asia, Africa, and the Americas. Below we synthesize this diverse collection of papers and conclude by proposing an agenda for future research that builds upon the insights from this Focus Issue.

Patterns and drivers of biodiversity
An often-repeated generalization is that SDTF have lower biodiversity than tropical wet or moist forests, even though they are thought to contain many endemic species (Banda-R et al 2016). However, for many taxonomic groups in dry forests, we still lack basic distribution data for how species vary within habitats and along environmental gradients. Linking plant functional traits and species distributions across space and time The diversity of ecophysiological strategies and functional traits of the plant species within the SDTF biome has long fascinated ecologists (Sobrado 1991, Borchert 1994, Eamus 1999. While previous work has focused on

Carbon cycling processes driven by rainfall variability
The seasonality of rainfall in SDTF yields the prediction that carbon cycling processes such as net primary production and soil respiration should be closely coupled to water availability, and thus evince strong variation at rainfall event, intra-and inter-annual timescales. A number of papers in this Focus Issue address above and belowground carbon dynamics, with a particular focus on rainfall-induced mechanisms responsible for patterns across different timescales. These papers employ a variety of tools including remote sensing ( Collectively, these studies corroborate the strong role of rainfall in determining patterns and processes of carbon cycling in dry forests. At timescales of individual rainfall events, Waring and Powers (2016) used a field experiment that simulated rain events to investigate the mechanisms underlying the 'Birch effect', which refers to the pulse of carbon dioxide (CO 2 ) released to the atmosphere upon rewetting of dried soil. They found that the size of the CO 2 flux released upon simulated rainfall events was proportional to the dissolved organic carbon pool in the soil, which strongly suggests microbial mineralization of labile organic carbon accounts for much of the CO 2 pulse (Waring and Powers 2016). Similarly, net ecosystem production, or the balance between gross primary production and respiration is also driven by rainfall in SDTF. For example, Mulga vegetation in Australia can act as a sink or source of CO 2 to the atmosphere depending on annual rainfall, and the inter-annual differences in net ecosystem productivity can be an order of magnitude (Eamus et al 2016). Guan et al (2018) added to our understanding of the coupling between vegetation productivity and rainfall by using a simulation model to examine how different aspects of the rainfall regime (i.e. rainfall frequency, rainfall intensity, and rainy season length) affected gross primary production in Africa. In their simulation analyses, gross primary productivity was more sensitive to variation in rainy season length, whereas the borders among biomes reflected total annual rainfall (Guan et al 2018). Climate change is predicted to increase the frequency and severity of droughts (Dai 2013), and many tropical locations have already seen changes in rainfall seasonality (Feng et al 2013). Because of the apparent sensitivity of carbon cycling processes in SDTF to rainfall regimes that the papers in this Focus Issue document, we can expect profound changes in the structure and function of these forests in the future (Allen et al 2017). Many of these papers stress the urgent need for policies at both local, national, and international scales to better manage SDTF resources and safeguard their biodiversity for future generations.

Conclusions and future research directions
The papers in this Focus section advance the limited knowledge of the role of rainfall regimes in determining the boundaries, composition, structure, and function of SDTF, and provide key information on the response of those ecosystems to severe meteorological events such as droughts. Future research should build upon these findings in a number of ways. First, the variability in climates within the SDTF biome the enormous range in topo-edaphic conditions, and the large beta diversity documented by these studies suggests that future studies should focus on what links and differentiates SDTFs. The interactions among element cycles, including the coupling among water, carbon, and macronutrients such as nitrogen and phosphorus discussed in the paper by (Campo 2016) provides ample motivation for why research in STDF should consider multiple field sites.
Explicit consideration of this geographic variation is necessary to both inform process-based models and to design conservation policies. Second, the focus on 'soft' or easy-to-measure plant functional traits such as wood density and specific leaf area reflected in many of the studies in this collection demonstrates their utility in explaining patterns of species distribution and functional along environmental gradients. Future work needs to build upon these advances by measuring the so-called 'hard traits', i.e. plant hydraulic and anatomical traits described in (Eamus et al 2016), which may better reflect plant performance under changing climate conditions. Lastly, these papers underscore the close linkages between SDTF and the communities of people that live in around them (