IMPACTS OF CLIMATE VARIABILITY AND LAND USE CHANGE ON THE HYDROLOGY OF THE AMAZON RIVER BASIN By Omid Bagheri A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Civil Engineering – Doctor of Philosophy 2023 i ABSTRACT This dissertation investigates the intricate dynamics of hydrologic systems in the Amazon River basin (ARB) in the face of evolving climate patterns and human interventions. The ARB – a pivotal element of the global climate, hydrological, and biogeochemical systems – holds immense biodiversity and profoundly influences global water, energy, and carbon cycles. Climate variations and human activities, especially deforestation in the southern subbasins, have considerably altered the basin's functioning. Despite extensive research, critical scientific gaps remain regarding key processes that govern hydrologic dynamics and the resilience of the rainforest. This research disentangles the impacts of climate and land use/land cover (LULC) changes toward devising robust resource management strategies. Recent acceleration of the hydrological cycle of the ARB and the increase in the frequency of extreme events could be early indicators of the change in hydrological cycle in the region surpassing some irreversible thresholds. While some systematic tipping points are inferred over the ARB, no tipping points associated with dominant hydrological processes over the ARB are investigated. This inhibits the understanding of hydrological considerations needed for sustainable forest management under climatic change and growing human stressors. The dissertation employs high resolution (~2km), long-term simulations from a process-based hydrological model (LEAF-Hydro-Flood) to investigate the dominant hydrological processes across the ARB, their key roles in shaping basin functions, and the decadal evolutions therein. Further, by developing static and dynamics LULC scenarios, the impact of climate variability and LULC change are isolated. Finally, through a comprehensive area fraction analysis and using a corresponding tree cover dataset, the tipping points associated with dominants hydrological processes in the ARB are assessed. Results indicate that shallow groundwater (<5m deep) strongly modulates the seasonality of the surface ii fluxes across the ARB and at least 34% of the Amazonian Forest is supported by groundwater during the dry season. This study reveals a two-month lag between seasonal peak evapotranspiration (ET) and river discharge as a crucial mechanism in preventing rainforest tipping into savanna. The ARB is dominantly energy limited; however, the results suggest that in the absence of groundwater support, and with less than ~125 mm/month of precipitation, the ARB could have become water-limited over some regions. The long-term basin-averaged ET— dominated by transpiration—changed with a split pattern of ±9% in the past three decades. Similarly, water table depth (±19%) and runoff (±29%) changed with a heterogenous patterns across the ARB. The contribution of canopy interception loss and ground evaporation changed heterogeneously across the ARB in response to deforestation. River discharge did not change substantially due to the crucial buffering role of groundwater, but terrestrial water storage (TWS) decreased (increased) in the 2000s (2010s) compared to that in the 1990s. Although groundwater is the dominant contributor to total TWS, the dynamics of TWS over the major river channels are controlled by flood water, given relatively shallow groundwater. Despite extensive deforestation, climate variability remains the dominant influence on WTD dynamics; however, the impacts on ET varied across the basin. Runoff patterns were intricately tied to precipitation and water table dynamics, demonstrating regional variations influenced by both climate variability and LULC changes. The area fraction analysis of WTD seasonality confirms the existence of tipping points associated with groundwater dynamics in the ARB. This study provides crucial insights on (i) the dominant hydrological processes, (ii) isolated impacts of climate variability and LULC change on the water cycle of the ARB, and (iii) tipping points in the ARB that are associated with groundwater dynamics. These findings could be used to inform effective water resource management and sustainable environmental practices in this ecologically significant region. iii Dedicated to my beloved parents. iv ACKNOWLEDGEMENTS This Ph.D. journey has profoundly transformed my life. Through the inevitable highs and lows, it would have been insurmountable without the invaluable guidance and unwavering support of numerous individuals. Foremost among them is my deep gratitude to Dr. Yadu Pokhrel, my Ph.D. advisor, whose exceptional mentorship, support, enlightening guidance, and constant encouragement were instrumental throughout my doctoral program. This dissertation owes its existence to his indispensable assistance, and I am fortunate to have collaborated with him. I also extend sincere appreciation to my committee members, Dr. Shu-Guang Li, Dr. Phanikumar Mantha, and Dr. Nathan Moore, for their consistent assistance, meticulous comments, and insightful suggestions. The completion of this dissertation was made possible in part by the support of the National Science Foundation (CAREER, Award #1752729 and INFEWS, Award #1639115). I acknowledge with gratitude the provision of high-performance computing by NCAR’s Computational and Information Systems Laboratory (doi:10.5065/D6RX99HX). Special thanks go to Suyog Chaudhari and Mohammad Hassan Badizad for their technical insights in Python scripting, crucial for producing some of the figures in this dissertation. I express my thanks to dear friends and colleagues Farshid Felfelani, Sanghoon Shin, Tamanna Kabir, Amar Deep Tiwari, and Ahmed Elkouk for engaging in fruitful discussions and sharing joyful moments. My heartfelt appreciation goes to my parents, Ali and Fatemeh, for their boundless love and for guiding me in shaping my life. v TABLE OF CONTENTS Chapter 1 Introduction .................................................................................................................... 1 1.1 Research Motivation ........................................................................................................ 1 Global significance of the Amazon River basin ....................................................... 1 Water cycle and climate variability in the ARB ....................................................... 2 LULC change in the ARB......................................................................................... 6 Process-based analysis of climate variability and LULC change impacts ............... 8 Separating the LULC change impacts from CV impacts ........................................ 11 The ARB tipping points .......................................................................................... 14 1.2 Research Goal, Objectives, and Science Questions ....................................................... 16 1.3 Dissertation Outline........................................................................................................ 18 Chapter 2 Groundwater Dominates Terrestrial Hydrological Processes in the Amazon at the Basin and Subbasin Scales .......................................................................................... 19 Introduction .................................................................................................................... 19 2.1 2.2 Methods .......................................................................................................................... 25 Model Description .................................................................................................. 25 Atmospheric Forcing .............................................................................................. 27 Land Use/Land Cover and Leaf Area Index ........................................................... 28 Simulation Setup ..................................................................................................... 30 Trend Analysis ........................................................................................................ 31 2.3 Results and Discussion ................................................................................................... 32 Validation ................................................................................................................ 32 Dynamics of Key Hydrological Processes .............................................................. 39 Governing Hydrological Processes ......................................................................... 60 Early/Late Warning Signals .................................................................................... 68 Hydrological Implications for Forest Management ................................................ 69 2.4 Conclusions .................................................................................................................... 72 Chapter 3 Climatic and Anthropogenic Impacts on the Hydrology of the Amazon River Basin . 76 3.1 Introduction .................................................................................................................... 76 3.2 Methods .......................................................................................................................... 80 Model Description .................................................................................................. 80 Atmospheric Forcing .............................................................................................. 82 Land Use/Land Cover and Leaf Area Index ........................................................... 83 Simulation Setup ..................................................................................................... 84 3.3 Results and Discussion ................................................................................................... 85 Validation ................................................................................................................ 85 Impacts on Groundwater ......................................................................................... 85 Impacts on Evapotranspiration ............................................................................... 93 Impacts on Runoff................................................................................................... 98 3.4 Conclusions .................................................................................................................. 103 Chapter 4 Tipping Points Associated with Water Table Depth in the Amazon River Basin ..... 106 Introduction .................................................................................................................. 106 4.1 4.2 Methods ........................................................................................................................ 110 4.3 Results and Discussion ................................................................................................. 112 vi 4.4 Conclusion .................................................................................................................... 118 Chapter 5 Summary and Conclusion .......................................................................................... 120 REFERENCES ........................................................................................................................... 122 vii 1. Chapter 1 Introduction 1.1 Research Motivation Global significance of the Amazon River basin With a total area of ~7.3 million km2 (including the Tocantins) and diverse rivers, floodplains, and wetlands, the Amazon River basin (ARB) is home to the most extensive tropical forest biome on the planet (~40% of the global tropical forest) (L. E. O. C. Aragão et al., 2014; Junk et al., 2011; W. F. Laurance et al., 2001; Jose A. Marengo et al., 2018; Reis et al., 2019; Weng et al., 2018). It spans nine nations (i.e., Brazil, Bolivia, Peru, Ecuador, Colombia, Venezuela, Guyana, Suriname, and French Guiana) and hosts four of the 10 largest rivers in the world (i.e., Solimoes, Madeira, Negro, and Japura rivers) (Fassoni-Andrade et al., 2021). The Amazon River flows into the Atlantic Ocean with an average annual discharge of 206×103 m3/s (Callede et al., 2010), amounting to ~20% of the total global freshwater reaching the ocean annually (Nepstad et al., 2008) and transfers substantial amount of sediment to the ocean (1.1×109 tons/year) (Armijos et al., 2020). Further, ARB provides home to ~25% of all terrestrial species on earth, accounts for ~15% of global terrestrial photosynthesis and is referred to as “lungs of the earth” (Field et al., 1998; Malhi et al., 2008). Further, the ARB is an important component of global biodiversity as well as global water, energy, and carbon cycles, and plays a key role in the global climate system through high rates of precipitation recycling and atmospheric moisture transport, and large variations in freshwater storage and river discharge (Arvor et al., 2017; Cox et al., 2004; Y. Fan & Miguez- Macho, 2010; Fassoni-Andrade et al., 2021; Gash, 1979; Gatti et al., 2021; William F. Laurance et al., 2002; Malhi et al., 2008; Nagy et al., 2016; Carlos A. Nobre et al., 1991; B. Soares-Filho et al., 2010; B. S. Soares-Filho et al., 2006; Werth & Avissar, 2005). It receives high annual 1 rainfall, on average ~2,200 mm/year (Builes-Jaramillo et al., 2018; Espinoza Villar et al., 2009). This rainfall depends on advected oceanic moisture combined with moisture recycling. ET can contribute up to 30–40% of the atmospheric moisture during the dry season (Eltahir & Bras, 1994; Van Der Ent et al., 2010; Salati et al., 1979; Satyamurty et al., 2013; Staal et al., 2018; D C Zemp et al., 2014) and is important for the initiation of the seasonal monsoon over the southern Amazon (Wright et al., 2017). The ARB ecosystems host 10–15% of land biodiversity (Hubbell et al., 2008; Lewinsohn & Prado, 2005) and the basin stores an estimated 150 billion to 200 billion tons of carbon (Cerri et al., 2007; Malhi et al., 2006; Saatchi et al., 2011) and surface waters in the basin are a major source and sink of carbon dioxide (Abril et al., 2014; Amaral-Zettler et al., 2020; Guilhen et al., 2020; Raymond et al., 2013; Richey et al., 2002) and the largest natural geographic source of methane in the tropics (Kirschke et al., 2013; Melack et al., 2004; Pangala et al., 2017; Pison et al., 2013). Importantly, the ARB is one of the top fifteen tipping elements of the Earth system (L. E. O. C. Aragão et al., 2014; W. F. Laurance et al., 2001; Lenton et al., 2008; Schellnhuber, 2009; Weng et al., 2018). The population of the ARB is estimated at ~10 million, mostly concentrated in urban areas along the river and its main tributaries. The basin is home of the local, including some indigenous, people that rely on rivers as transportation corridors and utilize these environments for their subsistence(Anderson & Ioris, 1992; Campos-Silva et al., 2021; Endo et al., 2016). Amazon also serves the broader South American population in terms of energy, food, and other forest products. Water cycle and climate variability in the ARB The ARB is primarily characterized by lowlands with a warm and rainy climate; however, the upper basin, encompassing the eastern slope of the Andes, exhibits a diverse range 2 of mountain climates. The climate across the ARB varies from a wet northwest with minimal dry spells to a dry southeast marked by an extended dry season. Notably, alterations in precipitation, particularly during the dry season, play a pivotal role in determining the climatic trajectory of the ARB (Malhi et al., 2008). The hydroclimate system of the ARB operates across various spatial and temporal scales, and major climatic patterns are predominantly driven by large-scale processes. For instance, interdecadal and interannual variabilities (e.g., extreme events) in the ARB’s climate are primarily modulated by persistent sea surface temperature (SST) patterns in both the Atlantic and Pacific oceans. Specifically, those linked to El Niño/Southern Oscillation (ENSO). Concurrently, mesoscale processes (e.g., topography and land-atmosphere interactions) modulate localized circulations. Typically, ENSO is accompanied by droughts in the ARB, resulting in low river water levels, a heightened risk of forest fires, and impacts on river ecosystems (Malhi et al., 2008). Variations in Pacific SST, primarily influenced by ENSO, play a significant role in shaping wet-season rainfall patterns. This influence leads to the suppression of convection in the northern and eastern regions of the ARB during ENSO events. In contrast, the variability in dry- season rainfall is strongly tied to the north-south SST gradient in the tropical Atlantic. The intensification of this gradient results in a northward shift of the Intertropical Convergence Zone at interannual time scales, while also strengthening the circulation of the Hadley Cell over longer periods. Such intensification contributes to prolonged and more intense dry seasons, particularly affecting the southern and eastern regions of the ARB, as evidenced in 2005 (Malhi et al., 2008). The interannual variability of the Atlantic gradient is influenced by distant factors like ENSO and the North Atlantic Oscillation, along with variations in evaporation triggered by the strengthening or weakening of local trade winds. Over extended time scales, alterations in the 3 North Atlantic, such as changes in thermohaline circulation due to subpolar melting or a warmer North Atlantic associated with the warming of northern hemisphere continents, have the potential to enhance the Atlantic SST gradient. Dry-season rainfall, crucial for vegetation patterns, is often driven by locally generated convection and can be significantly impacted by deforestation (Malhi et al., 2008). Water in the ARB primarily originates from ocean evaporation, undergoing multiple moisture recycling processes before ultimately returning to the ocean through surface or aerial rivers. The region experiences abundant rainfall, averaging ~2,200 mm/year (Builes-Jaramillo et al., 2018; Espinoza Villar et al., 2009). This substantial precipitation is a result of various factors, including intense radiative heating, low-level convergence of oceanic water vapor, continuous infusion of water vapor into the atmosphere by the rainforest itself, aided by the mechanical uplifting of air by the Andes. Land surface processes play a pivotal role in distributing precipitation into ET (averaging ~1,220 mm/year), surface runoff, and changes in surface and subsurface storage (Malhi et al., 2008). The Amazon River exhibits a highly seasonal flow, with seasonal imbalances between rainfall and downstream river discharge leading to substantial flooding across a vast floodplain area. These flooding events have beneficial ecological and biogeochemical implications. The occurrence of extreme flood and drought events is linked to intense interannual precipitation variability, influencing forest fires and biogeochemical cycles (Malhi et al., 2008). Microclimate control at the forest edges, including temperature and humidity regulation, constitute fundamental aspects of the coupled biosphere-atmosphere system in the ARB. These mechanisms shape the climate not only within the rainforest but also in the surrounding regions. Moreover, these processes contribute to the resilience of the coupled system during the dry season along its southern borders. They ensure a consistent source of water vapor 4 to the ARB’s atmosphere, particularly crucial when Atlantic oceanic inputs weaken (Malhi et al., 2008). The recent acceleration of the hydrological cycle in the ARB can be attributed to the increased interannual variability (Bagheri et al., 2024; Barichivich et al., 2018; Chagas et al., 2022). In recent years, the ARB has faced numerous climate extremes, including droughts and floods, some classified as "once-in-a-century" events, underscoring the region's susceptibility to climatic shifts (Barichivich et al., 2018; José Antonio Marengo & Espinoza, 2016). Historical records of similar droughts and floods indicate a reduction in the flood return period from 20 years to 4 years since 2000, signifying an increased frequency of extreme flooding events (Barichivich et al., 2018; José Antonio Marengo & Espinoza, 2016). Additionally, regional discharges have shown a rise in high flow in the northwestern regions of the ARB and a decrease in low flow in the southwestern regions during the 1974-2009 period. The dry season has expanded by approximately one month in the southern regions of the ARB since the mid-1970s. Warming is evident over the ARB, with the warming trend varying across different datasets. The warming trend becomes particularly pronounced from 1980 onwards, intensifying since 2000, with years like 2015-16 and 2020 ranking among the warmest in the last three decades (Almeida et al., 2017; Marengo et al., 2018). Determining the climate change fingerprint remains challenging due to the relatively short duration of climate records. However, climate modeling studies simulating the ARB’s deforestation predict significant reductions in rainfall over the ARB, impacting regional hydrology and increasing the vulnerability of ecosystem services for the local and regional population within and beyond the ARB. 5 Land use/land cover change in the ARB Deforestation, the complete removal of an area’s forest cover; and forest degradation, the significant loss of forest structure, functions, and processes; are the result of the interaction between various direct drivers, often operating in tandem (Barreto et al., 2021; Berenguer et al., 2014; Longo et al., 2020; Parrotta et al., 2012; Putz & Redford, 2010). Approximately 16% of the forest is deforested (Mapbiomas_Amazonia, 2022) with a concentration in Brazil where forest degradation has reached 17% (Bullock et al., 2020; Matricardi et al., 2020; Souza Jr et al., 2020). Deforestation in the Brazilian Amazon was less than 1% before 1975 but increased exponentially between 1975 and 1987 (Moran, 1993). The degraded forests are a persistent part of the landscape, as only 14% of them were later deforested (Bullock et al., 2020). Forest loss affects local temperature and precipitation, with increases in land surface temperatures and reductions in precipitation of up to 1.8% across the ARB. The deforestation is primarily a result of cattle ranching and replacement of forests with pasture and croplands at the agriculture frontier (“arc of deforestation”) in the southern subbasins (Bagley et al., 2014; Marcos H. Costa et al., 2007; Marcos Heil Costa & Pires, 2010a; Davidson et al., 2012; Guan et al., 2015; Mercedes & Montenegro, 2005; Moore et al., 2007; Morton et al., 2006). In addition, rapid population growth, timber extraction, mining, forest fires, construction of hydroelectric dams, urbanization and road network expansion are among other sources of land use/land cover (LULC) change in the ARB (Foley et al., 2005; B. S. Soares-Filho et al., 2006). In the past four decades, LULC changes occurred across the ARB, however, Tapajos (~31%), Xingu (~30%), Madeira (~21%), and Tocantins (~19%) are the four sub-basins of the ARB with big pockets of LULC change (Mapbiomas_Amazonia, 2022). Substantial LULC change happened around 1995, 1999 and 2004 (Mapbiomas_Amazonia, 2022; Smith et al., 2021). These changes in LULC in 6 conjunction with ongoing climate change impacted the terrestrial water cycle in recent decades (Sterling et al., 2013). Deforestation leads to local (e.g., changes in landscape configuration, climate change, and biodiversity loss), regional (e.g., impacts on hydrological cycle), and global impacts (e.g., increase of greenhouse gas emissions). Various anthropogenic drivers, including forest fires, edge effects, selective logging, hunting, and anthropogenic climate change can cause forest degradation (Andrade-Filho et al., 2017; Barlow et al., 2016; Bustamante et al., 2016; Phillips & Derryberry, 2017). Degraded forests have significantly different structure, microclimate, and biodiversity as compared to undisturbed ones. The degraded forests tend to have higher tree mortality, lower carbon stocks, more canopy gaps, higher temperatures, lower humidity, higher wind exposure, and exhibit compositional and functional shifts in both fauna and flora. Degraded forests can come to resemble their undisturbed counterparts, but this depends on the type, duration, intensity, and frequency of the disturbance event. In some cases, this may prohibit the return to a historic baseline. Deforestation and forest degradation are responsible for enormous quantities of CO2 emissions. The duration of the impacts of anthropogenic disturbances on Amazonian forests varies depending on the nature, frequency, and intensity of the disturbance; while logged forests may return to baseline carbon stocks within a few decades (Rutishauser et al., 2015), burned forests may never recover their original stocks (Silva et al., 2018). Recovery of degraded forests is also dependent on their landscape context, i.e., whether there are forests nearby that can act as sources of seeds and animals, thus speeding up recovery. Avoiding further loss and degradation of Amazonian forests is crucial to ensure they continue to provide valuable and life-supporting ecosystem services. 7 Process-based analysis of climate variability and LULC change impacts Over the past few decades, the global terrestrial water cycle has undergone unprecedented changes (Bosmans et al., 2017; Sterling et al., 2013), driven primarily by internal variability in the climate system, anthropogenic climate change (i.e., emission-driven) and direct human disturbances (Bosmans et al., 2017; Wohl et al., 2012). Human activities modulate the climate system at different scales through changes in LULC and components of water cycle to satisfy the growing need for food, fiber, water, and shelter for more than 7.8 billion people (Foley et al., 2005). Humans have changed more than 41% of natural landscape by anthropogenic land cover such as crop fields or pasture which impacts the evaporation-to-runoff ratio which, in general, has increased discharge and decreased evapotranspiration (ET) globally (Bosmans et al., 2017). Extensive LULC changes in watersheds has dramatic short- and long- term impacts on terrestrial hydrology and alters the occurrence and severity of extreme hydrological events (e.g. floods and droughts) which are the causes of the most human suffering among all climate-related events (Sterling et al., 2013). A significant portion of the changes in LULC are essential for agricultural and industrial development and the majority of other human interventions in the terrestrial hydrological cycle such as flow regulation and land development are requirements for ever-growing populations. Therefore, for sustainable development and for avoiding unintended consequences on land and water resources it is of crucial importance to understand the impacts of deforestation, afforestation and the collective LULC changes on terrestrial hydrological cycle. The system services in the ARB are altered due to climate variability and human disturbances with dominant form of deforestation as a result of replacing forests with pasture and agriculture specially across the “Arc of Deforestation” in the southern subbasins which is the 8 primary cause of LULC change (Marcos Heil Costa & Pires, 2010b; Davidson et al., 2012; Mercedes & Montenegro, 2005; Moore et al., 2007; Tropek et al., 2014). These changes in LULC in conjunction with ongoing climate change have impacted the terrestrial water cycle in recent decades (Sterling et al., 2013). Dependence of the hydro-ecological systems in the ARB on plentiful rainfall and the range of climatology across the basin highlights the importance of investigating the impacts of climate variability and LULC change on terrestrial hydrological cycle in ARB (Cook et al., 2012; Espinoza et al., 2015, 2016; Espinoza Villar et al., 2009; Nepstad et al., 2008). Previous interannual and interdecadal studies on hydrological alteration in ARB showed an overall long-term increasing trend in terrestrial water storage (TWS), however, the southern and southeastern sub-basins are experiencing significant decreasing trends in TWS, and LULC is known as the primary component contributing to the trend (Chaudhari et al., 2019). However, the processes which led to this alteration in the hydrological cycle of the ARB are not crystal clear yet. Impacts of human interventions in terms of LULC changes on terrestrial hydrological cycle are complex and depend on the initial LULC. The direct effects of the human induced LULC changes include the morphological and physiological variations in the landscape as reflected by altered aerodynamic roughness, leaf area index (LAI), stem area, surface resistance, albedo, and rooting depth (Bala & Nag, 2012; Bäse et al., 2012). The indirect effects of LULC changes on the soil and atmospheric boundary layer include the altered infiltration capacity and hydraulic conductivity in the shallow soil layer (Bonell et al., 2010; Ghimire et al., 2014; Hassler et al., 2011; Lanckriet et al., 2012; Muma et al., 2011; Neill et al., 2013), as well as varied net radiation, sensible and latent heat flux, and wind speed (Mishra et al., 2010). Both the direct and the indirect effects of LULC changes have strong implications particularly for energy, 9 momentum, and water balance in the atmospheric boundary layer and could affect hydrologic cycles and climate systems at different scales (e.g., locally, regionally, and globally) (Bala & Nag, 2012; Kumagai et al., 2013; Poveda et al., 2014; Dominick V Spracklen et al., 2012). The interaction of LULC in climate and land system, being scale-dependent, in addition to lack of comprehensive simulations of the system which include water cycle, ecology, abiotic- biotic linkages, and human interventions make study of impact of LULC change on terrestrial hydrological cycle a cumbersome task. Meanwhile, the study of the impacts of climate variability and LULC change on key hydrological variables (e.g., river discharge, ET, WTD and TWS) and quantification of their effects is currently feasible due to recent advances in fully- physics based hydrological models and emerging remotely sensed data and observations (Bosmans et al., 2017; Foley et al., 2005; Sterling et al., 2013; Wohl et al., 2012). Characterizing and understanding the dynamics of the ARB water cycle is of primary importance for climate and ecological research and for the management of water resources. Consequently, there is a need for comprehensive monitoring of the spatial-temporal dynamics of the ARB water cycle components and how they interact with climate variability and anthropogenic pressure. The region is now facing risks under climate and anthropogenic changes, and changes in Amazon hydrology could have substantial impacts globally (Jimenez et al., 2019). In the past decades, the basin experienced several intense climatic events, such as extreme droughts and floods, with no equivalent in the last 100 years (Barichivich et al., 2018; Marengo & Espinoza, 2016). Severe droughts can lead to environmental disturbances, from increased fire occurrence (Zeng et al., 2008) to abrupt shifts in fish assemblages (Röpke et al., 2017). Moreover, the accumulated negative impacts of increased human interventions across the region, such as damming (Forsberg et al., 2017; Latrubesse et al., 2017), deforestation (Arias et al., 2020; Coe et al., 2009; Gutierrez- 10 Cori et al., 2021; Leite-Filho et al., 2020; Leite et al., 2012), fires (Aragão et al., 2008; Libonati et al., 2021; Xu et al., 2020; Zeng et al., 2008), and mining (Abe et al., 2019; Lobo et al., 2015), will possibly trigger major modifications that could affect the ARB’s water cycle. Separating the LULC change impacts from CV impacts Hydrological impacts of LULC change are difficult to discern at large-scale basins with gradual changes and difficult to isolate from climate variability impacts either through observations or experiments (Arias et al., 2018; Dey & Mishra, 2017; Levy et al., 2018; Davidson et al., 2012; Cavalcante et al., 2019). Having a variety of LULC types in various stages of protection and regeneration and possibility of occurring positive feedback adds to the difficulty level (Nepstad et al., 2001; Nobre et al., 2016; Lima et al., 2013; Costa & Pires, 2010; Knox et al., 2015; Costa et al., 2003; Panday et al., 2015; Rodriguez et al., 2010). Because climate variability is typically high, any underlying changes due either to climate change or LULC changes can be effectively obscured (Kundzewicz & Robson, 2004). This complexity is noted as a potential reason for the disagreement in hydrological response to LULC change among macrocatchment studies in the tropical rainforest (Cavalcante et al., 2019). For instance, Trancoso (2006) reported a predominant decline in runoff within the Xingu, Tapajos, and Madeira subbasins due to reduction in precipitation; Arias et al. (2018) observed a reduced river flow across much of the Tapajos, despite no significant trend in annual precipitation; and by masking the deforestation impacts, Panday et al. (2015) estimated the decrease in river flow in Xingu due to climate variability. Conversely, Marengo et al. (1998) did not observe significant trends in discharge within the ARB and Tocantins. However, Costa et al. (2003) noted an uptick in mean discharge in the upper Tocantins, where savanna was the primary original land cover. In the majority of these subbasins, forest predominantly converted to pasture, except in Tapajos 11 where forest converted to soybean croplands which might explain the reduced river flows in Tapajos (Cavalcante et al., 2019). Global climate change affects the ARB through temperature increase and alters precipitation patterns and climate extremes, leading to increased tree mortality and terrestrial and aquatic biodiversity loss. Coupled with land-use change through deforestation and degradation, this reduces ET, changes carbon cycling dynamics, decreases the resilience of the ecosystems, and leads to further biodiversity loss and tree mortality, emitting greenhouse gases that impact not only regional, but the global climate. In this way, deforestation in the ARB enhances climate change. Because major drivers of the hydrological system are not stationary in time, isolating the hydrological impacts of human activities from climate variability is challenging. Separating the impacts of climate variability and anthropogenic impacts on the water cycle of the ARB is important for several reasons, and a large number of methods and theories have been widely used. First, the impacts of climate variability and LULC can be accumulative, subtractive, intensifying, and mitigating, therefore, there is a need for separating the impacts to identify the drivers of the change. Second, separating the impacts of climate change and anthropogenic impacts can help us develop more effective strategies for managing the ARB. For example, if the drivers behind the changes in water availability are primarily due to climate change, prioritizing strategies such as water conservation, drought-resistant crops, and water storage to adapt to those changes might be effective. On the other hand, if changes in water availability are primarily due to human activities, prioritizing strategies such as land use planning, water use efficiency, and pollution control to mitigate those impacts might be effective. Overall, understanding the interplay between climate variability and land cover is fundamental to the conservation and 12 sustainable management of tropical river basins, where forests play an important role in regional hydrological alterations at regional, continental, and global scales (Coe et al., 2013; Davidson et al., 2012; Malhi et al., 2008). Various techniques have been employed to disentangle the influences of climate variability and human activities on hydrological processes. These methods include hydrological modeling, conceptual, analytical, and experimental approaches (Cavalcante et al., 2019; Dey & Mishra, 2017; Wang, 2014; Tomer & Schilling, 2009; Schaake, 1990; Wang & Hejazi, 2011). Specific methodologies dedicated to isolating the impacts of climate variability and human activities on streamflow have been devised, such as the Tomer and Schilling framework (Tomer & Schilling, 2009), the elasticity-based method (Schaake, 1990), and the decomposition of the Budyko-type curve method (Wang & Hejazi, 2011), among others (Dey & Mishra, 2017; Wang, 2014; Wei et al., 2013). Acknowledging that each method/technique possesses its unique strengths and weaknesses, Wei et al. (2013) proposed that employing a combination of methods would constitute a more robust research strategy than relying on any single method alone. They also emphasized the need for additional case studies. Paired catchment studies, commonly utilized to assess the impact of vegetation changes on water yield, typically involve small catchments and can be cost-prohibitive (Brown et al., 2005). Alternatively, model simulations are often used, necessitating time-consuming calibration and validation processes, large datasets dependent on model assumptions, and statistical methods like time series analysis (Zégre et al., 2010; Zhao et al., 2010). However, the need for calibration is obviated in physics-based hydrological models, and advancements in remote sensing and computational systems have alleviated many associated limitations. 13 The ARB tipping points Tipping points (unstable equilibrium states) are defined as phenomena that, beyond a certain threshold, runaway change propels a system to a new state (van Nes et al., 2016; Scheffer et al., 2001). For example, due to deforestation and replacement of forest with pasture ET decreases and water table becomes shallower owing to extra recharge. Then, groundwater causes a positive feedback mechanism (DeAngelis et al., 2012) in further decreasing ET and recharging groundwater and propelling the forest system to an alternative tree species system. Therefore, once a threshold is passed, the dynamics of the system can accelerate dramatically to cause a ‘runaway change’. Two fundamental different ways in which a system can move to another stable state: (i) a change in external conditions (disturbance; e.g., climate change) which in models are represented by parameters, or (ii) a change in the state of the system itself (perturbations; e.g., human activities) which in models is represented by state variables (van Nes et al., 2016). The first type of tipping points are detected by warning signals or resilience indicators because of the gradual erosion of the resilience of the previous state of the system (van Nes et al., 2016). The ongoing changes in the ARB’s forest system may result in a loss of resilience and surpassing tipping points, triggering a persistent shift to an alternative state within the ecosystem. Five systemic tipping points are inferred over the ARB including four associated with climate and one associated with human-induced changes (Science Panel for the Amazon, 2021). These tipping points include (1) receiving annual precipitation below 1,000 mm/yr, as inferred from satellite observations of tree cover distributions (Hirota et al., 2011; Staver et al., 2011) or 1,500 mm/yr, as inferred from global climate models (Malhi et al., 2009), (2) a dry season lasting more than seven months, determined from satellite observations of tree cover distributions (Staver et 14 al. 2011), (3) maximum cumulative water deficit values exceeding than 200 mm/yr (Malhi et al. 2009) or 350 mm/yr (Zelazowski et al., 2011) over the ARB lowlands, inferred from various analyses with global climate models, (4) a 2oC increase in the Earth’s equilibrium temperature, identified through a coupled climate–vegetation model (Jones et al., 2009), and (5) 20-25% accumulated deforestation of the entire basin, determined through a combination of environmental changes (e.g., increased dry season length), climate projections aligned with the most pessimistic pathway of the Intergovernmental Panel on Climate Change (IPCC), and human-induced degradation via deforestation (Lovejoy & Nobre, 2019; Carlos A. Nobre et al., 2016). Existing evidence indicates that, depending on diverse combinations of stressing conditions, disturbances, and feedback mechanisms, the current forest configurations at the local scale, could be replaced by: (i) a seasonally dry, closed-canopy tropical forest with an increasing abundance of deciduous tree species (Dexter et al., 2018; Malhi et al., 2009); (ii) a tropical savanna state dominated by native grass and tree species (Cox et al. 2004; Jones et al. 2009; Hirota et al. 2011; Staver et al. 2011; Lovejoy and Nobre 2019); (iii) an open-canopy degraded state, dominated by invasive alien grasses and native fire-tolerant tree species (Barlow & Peres, 2008; Brando et al., 2012; Flores et al., 2016); and (iv) a closed-canopy secondary forest, dominated by native early successional tree and other plant species (Poorter et al., 2016; Rozendaal et al., 2019). Local-scale forest collapses could initiate cascading effects on rainfall recycling, intensifying dry seasons and wildfire occurrence, potentially leading to massive forest loss at continental scales, particularly in the southwest of the basin. The probability of crossing these tipping points largely depends on heterogeneities across the system, including geological, physical, chemical, and cultural processes that influence connectivity and the likelihood of contagious disturbances. The primary concern is that beyond these potential tipping points, the 15 system might enter a loop of reduced rainfall, increased fire, and heightened forest mortality. Over the past six decades, the temperature in the ARB has risen by 1-1.5℃ (Nobre et al., 2016), approximately 18% of the forest area has been deforested (Mapbiomas_Amazonia, 2022), forest degradation has reached 17% (Bullock et al., 2020; Matricardi et al., 2020), forest fires have significantly increased (Aragão et al., 2018), dry season lengths (measured as the number of consecutive days with less than 50mm rainfall) are three to four weeks longer compared to six decades ago (Fu et al., 2013), and dry season water storage deficit is on a divergent trend (Chaudhari et al., 2019). Some studies suggest that the escalating frequency of unprecedented droughts, such as those in 2005, 2010, 2015-16, and 2020, could be signaling the imminent arrival of a tipping point (Bagley et al., 2014; Lovejoy & Nobre, 2019; Walker, 2020). Consequently, there is an imperative need to curtail deforestation in the ARB, rehabilitate the lost forest in its southern and eastern regions, and provide science-based guidelines to inform forest management policies (Lovejoy and Nobre, 2019; Walker, 2020). 1.2 Research Goal, Objectives, and Science Questions As discussed above, the hydrology of the ARB has been extensively studied; however, critical scientific gaps remain regarding key processes that govern hydrologic dynamics and the resilience of the rainforest. This inhibits the understanding of hydrological considerations needed for sustainable forest management under climatic change and growing human stressors. This dissertation aims to examine the changes in basin-wide water and energy balances under large- scale climate variability and LULC changes and the resulting shifts in system thresholds toward a new equilibrium. The goal is to quantify the impact of climate variability and LULC change in the past four decades, identifying the dominant hydrological processes at the basin and subbasin scales, and identifying the tipping point associated with the dominant hydrological processes 16 using a multi-scale assessment of the basin based on the results of high-resolution simulations using LEAF-Hydro-Flood (LHF). This dissertation is driven by the following overarching scientific questions: (1) How have the major components of water and energy balances in the ARB evolved due to changes in hydrological drivers? (2) What dominates terrestrial hydrological processes at the basin and subbasin scales in the ARB? (3) What are the impacts of climate variability and LULC change in the ARB over the past four decades? (4) Are there tipping points in the ARB associated with WTD dynamics? These overarching questions are addressed by answering the following specific science questions under different chapters. Chapter 2. Analysis of the hydrologic dynamics of the ARB and governing processes Q1. How did the fundamental hydrological processes in the ARB evolve over the past three decades? Q2. What key factors govern the seasonality of the ARB at the basin and subbasin scales? Chapter 3. Quantifying the contribution of climate variability and LULC change in shifting the ARB to the current equilibrium at basin and subbasin scales Q3. What are the contributions of climate variability and LULC change in shifting the hydrology of the ARB to the current equilibrium at the basin and subbasin scale? Q4. At what temporal and spatial scale should the impacts of climate variability and LULC change be assessed? Chapter 4. Investigating tipping points associated with dominant hydrological processes in the ARB Q5. Are there tipping points associated with dominant hydrological processes in the ARB? Q6. How resilient is the hydrological system of the ARB against the tipping points? 17 To investigate the dominant hydrological processes over the ARB at the basins and subbasin scale, the hydrology of the ARB is simulated over the past four decades using the LHF model. Then, the contribution of climate variability and LULC change in shifting the ARB to the new equilibrium is isolated through two sets of separate simulations with static and dynamics LULC. Further, area fraction analysis of water table depth (WTD) and theory of dynamics systems are used to investigate the tipping points associated with WTD in the ARB. 1.3 Dissertation Outline The research questions are tackled in separate chapters (Chapters 2 through 4), and the key findings are summarized in Chapter 5. The following provides a summary of the remaining chapters. Chapter 2. Groundwater Dominates Terrestrial Hydrological Processes in the Amazon at the Basin and Subbasin Scales. Chapter 3. Impacts of climate variability and LULC change on hydrological cycle of the Amazon River basin. Chapter 4. Tipping points associated with water table depth in the Amazon River basin. Chapter 5. Summary and Conclusion. 18 2. Chapter 2 Groundwater Dominates Terrestrial Hydrological Processes in the Amazon at the Basin and Subbasin Scales Based on: Bagheri, O., Pokhrel, Y., Moore, N., Mantha, S.P., (2024). Groundwater Dominates Terrestrial Hydrological Processes in the Amazon at the Basin and Subbasin Scales. Journal of Hydrology, 628, p.130312. 2.1 Introduction The Amazon River basin (ARB) is home to the most extensive tropical forest biome on the planet (e.g., 40% of the global tropical forest area) and is also one of the tipping elements of the Earth system (L. E. O. C. Aragão et al., 2014; W. F. Laurance et al., 2001; Lenton et al., 2008; Schellnhuber, 2009; Weng et al., 2018) The basin is an important component of global biodiversity as well as global water, energy, and carbon cycles, and plays a key role in the global climate system through precipitation recycling and atmospheric moisture transport (Arvor et al., 2017; Y. Fan & Miguez-Macho, 2010; Fassoni-Andrade et al., 2021; William F. Laurance et al., 2002; Malhi et al., 2008; B. S. Soares-Filho et al., 2006; Werth & Avissar, 2005). The basin functioning (e.g., carbon storage, maintenance of biodiversity, and climate regulation) in the ARB has been altered substantially over the past few decades due to climate variability and human disturbances with deforestation as the dominant form; the deforestation is primarily a result of cattle ranching and replacement of forests with pasture at the agriculture frontier (“arc of deforestation”) in the southern subbasins (Bagley et al., 2014; Marcos H. Costa et al., 2007; Marcos Heil Costa & Pires, 2010b; Davidson et al., 2012; Guan et al., 2015; Mercedes & Montenegro, 2005; Moore et al., 2007; Morton et al., 2006). In addition, rapid population growth, timber extraction, mining, forest fires, and road network expansion are among other sources of land use and land cover (LULC) change in the ARB (Foley et al., 2005; B. S. Soares- Filho et al., 2006). 19 The hydrological cycle in the ARB is strongly modulated by evapotranspiration (ET) and frequent (up to 7 recycling per water molecule) (Salati et al., 1979; Staal et al., 2018; Weng et al., 2018) and substantial (25-50% of total Amazonian rainfall) moisture recycling (L. C. E. O. Aragão, 2012; Eltahir & Bras, 1994; Van Der Ent et al., 2010; D. V. Spracklen et al., 2012; Staal et al., 2018; Delphine Clara Zemp et al., 2017). Therefore, the basin’s hydrologic system is highly susceptible to widespread deforestation and forest degradation because it can substantially reduce moisture availability for recycling by increasing surface runoff (Lovejoy & Nobre, 2019; Malhi et al., 2008; Schellnhuber, 2009; Delphine Clara Zemp et al., 2017). In addition, the possibility of having positive feedback due to tree loss might exacerbate deforestation impacts (Delphine Clara Zemp et al., 2017); tree loss reduces both ET and rainfall, lengthening dry season, reducing humidity, and potentially increasing forest fire (Delphine Clara Zemp et al., 2017). Over most of the deforested areas in the ARB, land use is beyond moderate intensity and the hydrologic system has evolved under climate variability and anthropogenic disturbances, especially land use change (Chagas et al., 2022). For example, wet (dry) season is becoming wetter (drier) during the past decades in around one-third of the ARB (mainly in southern and eastern regions of the basin) (Leite-Filho et al., 2019) and the seasonal storage deficit has increased over time (Chaudhari et al., 2019). In addition, vapor pressure deficit (VPD) is increasing over South America (Barkhordarian et al., 2019) and mortality rate of wet-climate tree species where dry season is becoming longer is increasing (Esquivel‐Muelbert et al., 2019). Moreover, due to the increase in the frequency of extreme droughts, higher temperatures and increased forest degradation, the rainforest is becoming more vulnerable to fires (L. E. O. C. Aragão et al., 2018). Therefore, over many of the deforested regions, especially in southern and 20 eastern ARB, the hydrological system is likely being transformed with some changes being potentially irreversible (Lovejoy & Nobre, 2019). Several studies have predicted that with the current rate of deforestation and biodiversity loss, the ARB may have two tipping points, which could lead to savannization of the bistable regions of the tropical forest through loss of moisture recycling as a result of crossing the 40% deforestation threshold (change in internal state of the system due to anthropogenic impacts) or a 4℃ increase in temperature (global/regional climatic drivers) (Cox et al., 2004; van Nes et al., 2016; Carlos A. Nobre et al., 2016; Carlos Afonso Nobre & Borma, 2009; Sampaio et al., 2007; Schellnhuber, 2009; B. S. Soares-Filho et al., 2006; Staver et al., 2011; Walker, 2020; Zak & Nippert, 2012). However, based on the Assessment Report 5 of the Intergovernmental Panel on Climate Change (IPCC) and by going beyond a single-factor in explaining the forest degradation and considering the combined roles of global warming, deforestations and wildfires, the threshold for deforestation has been suggested to be as low as 20-25% instead of 40%, which could push the ARB toward an open-canopy degraded state, a very likely near future scenario (Lovejoy & Nobre, 2019; Carlos A. Nobre et al., 2016; Walker, 2020). Other studies have shown that temperatures in the region rose by 1-1.5℃ over past six decades (Carlos A. Nobre et al., 2016), ~18% of the forest area is deforested (Mapbiomas_Amazonia, 2022), forest degradation reached 17% (Bullock et al., 2020; Matricardi et al., 2020), forest fires significantly increased (L. E. O. C. Aragão et al., 2018), dry season lengths (number of consecutive days with less than 50mm rainfall) are three to four weeks longer in comparison to six decades ago (Fu et al., 2013), and dry season water storage deficit is on a divergent trend (Chaudhari et al., 2019). Some studies suggest that the increasing frequency of unprecedented droughts such as those of 2005, 2010, 2015-16 and 2020 could be signaling that the tipping point is at hand (Bagley et al., 2014; 21 Lovejoy & Nobre, 2019; Walker, 2020). Therefore, there is a need to reduce deforestation in the ARB, rebuild the lost forest in its southern and eastern regions and to provide science-based guidelines to assist forest management policies (Lovejoy & Nobre, 2019; Walker, 2020). Globally, passive and active approaches have been used to alleviate environmental stressors and to restore the forest through a secondary succession (Morrison & Lindell, 2011; Poorter et al., 2021). To measure the success of forest restoration, typical characteristics such as forest structure and diversity and ecosystem functioning are compared between the old-growth forest and the secondary forest where hydrological functioning is often neglected (Poorter et al., 2021). While tree restoration has been recognized as an effective way to store carbon and mitigate the impacts of climate change, not many studies have considered the hydrological effects of tree restoration (Hoek van Dijke et al., 2022). A recent study on the impacts of large- scale tree restoration showed that restoration can significantly alter terrestrial water cycle at different spatial scales and the impacts are non-linear and complex (Hoek van Dijke et al., 2022). Traditional management policies in the ARB were commonly developed focusing on maximizing economic benefits and neglecting hydrological roles of the forest (Carlos A. Nobre et al., 2016). Such omission of the hydrological roles arises partly from the lack of a comprehensive understanding of the short- and long-term impacts of management practices over varying temporal and spatial scales. As such, it is imperative that we better understand the dominant hydrological processes across the ARB that govern forest resilience and are crucial for improved management practices. In addition, since forest management can have long-term implications on the future of the ARB, it is important that such studies investigate the decadal evolution of the dominant processes under climate variability and human disturbances. Lastly, 22 identifying warning signals can help better monitor the impacts of management policies on the terrestrial hydrological cycle. However, observational data—even those based on remote sensing—for such long-time scales and all relevant hydrological variables are lacking, especially for the entire ARB, or are available at short temporal scales, which make hydrological modeling the only viable option to study the terrestrial hydrology of the ARB. Early hydrological modeling studies in the ARB were conducted to uncover the underlying processes involved in moisture recycling and to study the impact of land use/land cover change on the water cycle (Marcos Heil Costa & Foley, 1999; Eltahir & Bras, 1994; Carlos A. Nobre et al., 1991; Shukla et al., 1990; Zeng, 1999; Zeng et al., 1996). These studies have emphasized the importance of land-atmosphere feedback in hydrological modeling to reduce the uncertainty in the results as some earlier studies found contradictory outcomes associated with neglecting the feedback (Eltahir & Bras, 1994). The limitations in required data and computational resources to run distributed hydrological land surface models in the past lead to significant growth of lumped hydrological models and data- based studies in the ARB. Lumped models are valuable tools to understand the big picture of hydrology in the ARB and to address wide range of research questions, however, they do not fully account for the heterogeneity in biomes and are simplistic in parameterizing various storage and fluxes, making them inappropriate for studies on process characterization (Heerspink et al., 2020; Maeda et al., 2017). Advances in process-based hydrological modeling and remote sensing methods have provided new opportunities to simulate basin hydrology and study the dominant terrestrial hydrological processes (Clark et al., 2015; Frappart et al., 2019; Getirana et al., 2012; De Paiva et al., 2013; Pfeffer et al., 2014). Such models have been used to simulate groundwater dynamics 23 across the ARB, leading to fundamental advances in the understanding of the role of groundwater and providing opportunities to disentangle research questions that were not possible to address before (Chaudhari et al., 2021; Miguez-Macho and Fan, 2012a, 2012b; Pokhrel et al., 2014, 2013). For example, Miguez‐Macho and Fan (2012a) investigated the role of groundwater on the surface water dynamics of the ARB and the buffering role of groundwater during the dry season based on the results of LEAF-Hydro-Flood (LHF) simulations. They found that the dynamics of WTD dominates streamflow in the headwater catchments and the two-way exchanges of surface and subsurface water over the large floodplains. In addition, shallow WTD supports large areas of waterlogged wetlands that are rarely flooded. In a following study, Miguez‐Macho and Fan (2012b) investigated the role of groundwater in mitigating water stress on related processes to soil moisture and ET. Further, Pokhrel et al. (2013), studied the influence of groundwater on terrestrial water storage (TWS) using LHF model, finding that subsurface storage dominates the dynamics of TWS over a major part of the ARB; however, they reported that where WTD is shallow, the dynamics of TWS is governed by floodwater. In another study and based on the results of the LHF model, Chaudhari et al. (2019) investigated the dominant mechanisms modulating the dynamics of TWS and droughts over the ARB. They suggested that the ARB is getting wetter overall, but the southern and southeastern subbasins are getting drier with the dry season water storage deficit on a divergent trend. A recent study suggests that the double stress of waterlogging and drought is the primary driver of forest-savanna coexistence with alternating drought and waterlogging at the seasonal scale favoring savanna over forests (Mattos et al., 2023). Despite the findings in recent studies, to the best of our knowledge, there are no comprehensive studies that investigated the key processes governing the hydrologic 24 dynamics at the basin and subbasin scales across the ARB, the linkages therein, and their historical evolution. The present study addresses the aforementioned research and knowledge gaps by answering the following science questions. (1) How did the fundamental hydrological processes in the ARB evolve over the past three decades? (2) What key factors govern the seasonality of the ARB at the basin and subbasin scales? (3) To what extent can hydrological variables in the ARB serve as viable early or late warning signals of alterations in the terrestrial water cycle during secondary succession? (4) What are the implications for forest management that can be derived from the findings of studies such as ours? We hypothesize that the shallow water table depth (WTD<5m) was a key attribute that supported the ARB’s hydrologic regime during the past three decades against climate variability and anthropogenic disturbances. As such, shallow groundwater fraction area could be taken as a proxy to monitor the impacts of human activities on the basin’s hydrology and ecosystem functioning. Our second hypothesis is that the changes in the spatial distribution of ET serve as a direct measure of the hydrological impact of large- scale LULC changes in the basin. In addressing these questions and hypotheses, we first identify the dominant hydrological mechanisms by using the results from a basin-scale, fully process- based hydrological model. Then, we investigate how the key hydrological processes have evolved over the last three decades. Finally, we examine the role of the governing hydrological processes for sustainable forest management in the ARB. 2.2 Methods Model Description The model used in this study is LHF (Miguez-Macho and Fan, 2012a, 2012b; Pielke et al., 1992; Pokhrel et al., 2014, 2013; Walko et al., 2000). As described in detail in Miguez- 25 Macho and Fan (2012a), LHF is a fully process-based hydrology model capable of resolving coupled surface and subsurface hydrological processes at the continental-scale. The model was developed at two stages building on the Land-Ecosystem-Atmosphere Feedback (LEAF), the land-surface component of Regional Atmosphere Modeling System (RAMS) (Walko et al., 2000). The physics in the original LEAF model is described in detail in Walko et al. (2000). Turbulent and radiative exchange of the atmosphere with multilayer soil and snow water and thermal energy, surface storage, vegetation canopy, canopy air are inherited features of LEAF in LHF (Miguez-Macho and Fan, 2012a, 2012b). These include the parameterizations for simulating ET, which are similar to those used in state-of-the-art land surface models (e.g., (Lawrence et al., 2019); details are available in Walko et al. (2000). However, LEAF parameterizations including representation of sub-grid hydrologic heterogeneity, lateral soil water movement based on TOPMODEL (Beven & Kirkby, 1979) and groundwater flow processes have been replaced with new schemes or largely improved. The new developments and enhancements have been particularly tested over the ARB as described in the following. At the first stage of LHF development over North America (Miguez-Macho et al., 2007, 2007), LEAF-Hydro was adapted from LEAF by adding a prognostic groundwater module to allow (1) the rise and fall of water table or shrinkage and growth of the vadose zone, (2) the recharged water table to reach a new equilibrium following a rain event by discharging into rivers within a grid cell and convergence and divergence of lateral flow among adjacent cells, (3) two-way exchange between surface water and groundwater to represent both gaining and losing streams, (4) river routing to the ocean using the kinematic wave method and (5) sea level to influence coastal drainage by assigning the sea level as the groundwater head boundary condition. During the second stage over the ARB (Miguez-Macho and Fan, 2012a, 2012b), LHF 26 was developed through further enhancement of LEAF-Hydro by incorporating a river-floodplain routing scheme to estimate streamflow more realistically by solving the full momentum equations of open channel flow, also considering back water effect and the inertia of deep flow, which are both significant in the ARB (Bates et al., 2010; Miguez-Macho and Fan, 2012a, 2012b; Yamazaki et al., 2011). The incorporation of flood dynamics also enabled an explicit simulation of floodwater-groundwater interactions, a dominant process in the ARB. The initial LHF studies over the ARB provided an extensive evaluation of many hydrologic variables across the basin (Miguez-Macho and Fan, 2012a, 2012b). The model was subsequently used in numerous studies that presented further evaluations using observational and satellite-based data on various hydrologic fluxes and stores, and by using different atmospheric forcing datasets demonstrating robust model performance over the ARB (Brown et al., 2022; Chaudhari et al., 2021, 2019; Chaudhari and Pokhrel, 2022; Miguez-Macho and Fan, 2012a, 2012b; Pokhrel et al., 2014, 2013). Atmospheric Forcing LHF model in this study is forced with ERA5 reanalysis (Hersbach et al., 2020) available from 1950 to present at the spatial resolution of 0.25 degree and hourly time steps. The availability period and the spatial resolution were the main reasons for using EAR5 data. In previous studies, LHF results forced by WATCH Forcing Data methodology applied to ERA- Interim (WFDEI) reanalysis were successfully validated over ARB, however, the dataset is not available after 2019 (Chaudhari et al., 2019). Staal et al. (2020) used ERA5 over the ARB to conduct hydrological and atmospheric moisture tracking simulations and their results showed that ERA5 performs better than ERA-Interim in estimating wind fields and rainfall, especially in tropics (Staal et al., 2020). A total of eight variables from ERA5 dataset are used: precipitation, 27 surface pressure, surface solar (i.e., shortwave) radiation downwards, surface thermal (i.e., longwave) radiation downwards, air temperature, dewpoint temperature, u- and v-components of wind speed. Specific humidity is calculated from dewpoint temperature and surface pressure. The 3-hourly data at the coarser resolution noted above are spatially interpolated within LHF to the model grid resolution (~2km) using a bilinear interpolation (Chaudhari et al., 2021; Miguez- Macho & Fan, 2012, 2012; Pokhrel et al., 2013, 2014). Land Use/Land Cover and Leaf Area Index The annual land use/land cover (LULC) maps are derived from the European Space Agency (ESA) Climate Change Initiative’s Land Cover project; the original data are reclassified and aggregated to match the land use categories used in LHF, following our previous study (Chaudhari et al., 2019). Specifically, the 22 classes from the ESA land cover maps are reclassified into the 30 classes of LHF (Table S3). The datasets comprise an annual time series land cover maps with 300-meter spatial resolution for the 1992 to 2020 period. The baseline maps in the ESA dataset are generated using the Medium-spectral Resolution Imaging Spectrometer (MERIS) instrument based on the UN Land Cover Classification System (LCCS) and the maps were further modified based on the detected changes in land use and land cover by AVHRR (1992-1999), SPOT-Vegetation (1999-2012), and PROBAV (2013-2020) instruments. The LHF model updates LULC on an annual basis to account for year-to-year LULC changes; in the ARB, these annual changes are largely caused by human activities. The lookup table for leaf area index (LAI) is derived by overlaying the ESA land use map over the LAI maps from Moderate Resolution Imaging Spectroradiometer (MODIS) for the period of 2000 to 2020 and using a pixel-by-pixel analysis and the monthly values (Table S4) are calculated from the long- term 4-day mode of LAI for each LHF land cover class. 28 Table 2-1. Reclassification of ESA land use/land cover classes into LHF classes. LHF Classes ESA Classes Evergreen needleleaf forest Evergreen broadleaf forest Deciduous needleleaf forest Deciduous broadleaf forest Mixed woodland Woodland Tree cover, needleleaved, evergreen, closed to open (>15%) Tree cover, needleleaved, evergreen, closed (>40%) Tree cover, needleleaved, evergreen, open (15-40%) Tree cover, broadleaved, evergreen, closed to open (>15%) Tree cover, mixed leaf type (broadleaved and needleleaved) Tree cover, flooded, fresh or brakish water Tree cover, flooded, saline water Tree cover, needleleaved, deciduous, closed to open (>15%) Tree cover, needleleaved, deciduous, closed (>40%) Tree cover, needleleaved, deciduous, open (15-40%) Tree cover, broadleaved, deciduous, closed to open (>15%) Tree cover, broadleaved, deciduous, closed (>40%) Tree cover, broadleaved, deciduous, open (15-40%) Mosaic natural vegetation (tree, shrub, herbaceous cover) (>50%) / cropland (<50%) Mosaic tree and shrub (>50%) / herbaceous cover (<50%) Tree or shrub cover Wooded grassland Mosaic herbaceous cover (>50%) / tree and shrub (<50%) Closed shrubland Open shrubland Grassland Crop/mixed farming Shrubland Shrubland evergreen Shrubland deciduous Shrub or herbaceous cover, flooded, fresh/saline/brakish water Sparse vegetation (tree, shrub, herbaceous cover) (<15%) Sparse tree (<15%) Sparse shrub (<15%) Sparse herbaceous cover (<15%) Lichens and mosses Grassland Mosaic cropland (>50%) / natural vegetation (tree, shrub, herbaceous cover) (<50%) Irrigated crop Cropland, irrigated or post-flooding Cropland Bare ground Urban and built up Lakes, rivers, streams (inland water) Cropland, rainfed Herbaceous cover Bare areas Consolidated bare areas Permanent snow and ice Urban areas Water bodies 29 Table 2-2. Reclassification of ESA land use/land cover classes into LHF classes. LHF Land Cover Classes Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec Lakes, rivers, streams Mixed woodland Crop/mixed farming Irrigated crop Evergreen needleleaf forest Evergreen broadleaf forest Deciduous broadleaf forest Woodland Wooded grassland Closed shrubland Open shrubland Grassland Cropland Bare ground Urban and built up 0.01 1.29 1.24 1.10 1.15 3.20 1.74 1.21 0.66 1.20 0.20 0.71 1.60 0.20 0.69 0.01 1.23 1.20 0.76 1.05 1.60 1.71 1.18 0.64 1.21 0.20 0.69 1.61 0.20 0.69 0.01 1.20 1.28 0.75 0.74 2.15 1.88 1.24 0.65 1.29 0.22 0.69 1.62 0.22 0.70 0.01 1.23 1.71 0.75 0.69 4.10 2.18 1.58 0.68 1.63 0.22 0.71 1.68 0.21 0.73 0.01 1.29 2.11 0.75 0.70 5.05 1.81 1.57 0.73 1.58 0.20 0.71 1.60 0.20 0.73 0.01 1.26 2.12 1.17 0.82 5.61 1.69 1.57 0.85 1.26 0.18 0.71 1.20 0.20 0.72 0.01 1.20 1.66 1.26 1.18 5.69 1.39 1.27 0.85 1.18 0.17 0.69 1.05 0.19 0.70 0.01 1.17 1.28 1.33 1.21 6.00 1.24 1.21 0.84 1.13 0.18 0.72 0.76 0.19 0.68 0.01 1.17 1.29 1.33 1.23 6.16 1.21 1.21 0.82 1.13 0.20 0.74 0.76 0.20 0.68 0.01 1.17 1.26 1.20 1.21 5.70 1.23 1.20 0.77 1.14 0.20 0.69 0.85 0.20 0.67 0.01 1.19 1.24 1.18 1.17 5.06 1.57 1.21 0.75 1.15 0.20 0.69 1.14 0.20 0.67 0.01 1.33 1.25 1.18 1.14 4.19 1.71 1.20 0.71 1.20 0.20 0.71 1.30 0.20 0.68 Simulation Setup The LHF model is set up for the entire ARB (∼7.1 million km2) including the Tocantins River basin (Figure 2-1). Simulations are conducted for the 1979–2020 period at a spatial resolution of 1 arcmin (∼2 km) with a time step of 4 minutes as in previous studies (Chaudhari et al., 2019; Miguez-Macho & Fan, 2012; Pokhrel et al., 2013, 2014), and the output is saved at daily time steps. To capture the hillslope processes at up to the first-order stream valleys, very fine spatial scale for the simulations is desired, however, due to the computational costs and the coarse resolution of input data (such as soil characteristics) the spatial resolution of 1 arcmin is chosen as a tradeoff, as in previous studies. Reservoirs are not considered in the simulation, however, based on previous studies the impact of the reservoirs in the ARB are not substantial in the downstream reaches (Chaudhari & Pokhrel, 2022). As the focus of this study is to investigate the terrestrial hydrological processes at basin and subbasin scales, the impacts of reservoirs would not alter the findings. Starting with the equilibrium water table (Y. Fan et al., 2013) for 30 1979, the model is spun up for ~200 times for the year 1979 to stabilize WTD and the results for the 1992–2020 period (28 years) are analyzed. As the primary goal of this study is to examine the dominant processes in the ARB on a decadal scale and since the land cover and LAI datasets are available after 1992, simulations for 1979 to 1992 are discarded as additional spin-up. Moreover, as the model simulates land surface, hydrologic, and groundwater processes on a full physical basis, no calibration was performed (Chaudhari et al., 2019). Trend Analysis The Mann-Kendall (MK) test (Kendall, 1948; Mann, 1945) which serves as a prevalent method for detecting alterations in time-series data (Li et al., 2014) is used to detect the long- term trend. In this study, the detected trend is deemed statistically significant when the p value is less than 0.05 (i.e., 95% confidence level). Moreover, we employ the Theil-Sen slope estimator (Sen, 1968; Theil, 1950) to calculate the slope of change which computes the median slopes of lines fitted through pairs of data points in the dataset. Importantly, it exhibits greater robustness against outliers compared to simple linear regression methods (Lavagnini et al., 2011). The outcomes of the MK test are interpreted utilizing the z-score metric, wherein the sign of the z- score denotes the direction and magnitude of the trend. To comprehensively address the heterogeneity observed in the changes across key hydrological variables in our study, we separately calculate the mean slope separately for areas exhibiting negative and positive slopes. Additionally, we compute the basin-averaged slope to provide a more intricate understanding of the transformations occurring over the past three decades within the ARB. 31 2.3 Results and Discussion Validation The simulated streamflow from LHF is compared with observations obtained from the Agência Nacional de Águas (ANA) in Brazil (http:// hidroweb.ana.gov.br, last accessed: 10 September 2022). In this regard, 55 stream gauging stations from a wide range of river discharge magnitudes with at least 30 years of record are considered across the ARB. Figure 2-1 presents the results of three performance metrics, namely the Pearson correlation coefficient (PCC), modified Kling–Gupta efficiency (KGE) and Nash–Sutcliffe efficiency (NSE) (Siqueira et al., 2018). High values for PCC, KGE and NSE metrics can be observed for most stations, indicating overall good performance for various topographic locations and river discharge values. However, there are some stations with relatively lower PCC, KGE and NSE, which are situated mostly in streams with low annual mean flow and steep slopes, including the headwaters across the Tapajos and Madeira subbasins and along the streams in the northeastern regions of the ARB. The lower accuracy at those stations is likely related to high topographic gradient, where precipitation drains quickly, causing rather erratic patterns of seasonal streamflow, which adds challenges to resolving hillslope processes for low order streams at 2km resolution. In addition to the above statistical measures, the long-term seasonality of streamflow for 12 major gauge stations is compared (Figure 2-1). R-squared and RSR (a standardized version of the root mean square error (RMSE) that takes the standard deviation of the observed data at different stations into account (Legates & McCabe, 1999)) indicate good model performance in predicting the streamflow seasonality. 32 Figure 2-1. River discharge validation based on monthly average values derived from daily river discharges at 55 gauge stations across the ARB. The size of the circles in the top panel indicates flow magnitude with each of the three portions of the circles showing a model performance metric: PCC (red), KGE (green), and NSE (blue). The background image shows simulated river discharge indicated by line thickness (~2km grids). The grid panels at the bottom depict the long-term seasonal cycle of monthly river discharge for 12 major gauge stations indicated on the top panel. 33 As shown in previous studies (Chaudhari et al., 2019; Felfelani et al., 2017; Pokhrel et al., 2013) and corroborated in this study (Section 3.2), groundwater is the major component of TWS in the ARB because over most parts of the basin the water table is relatively shallow. Therefore, given the lack of systematic water level observations in the ARB, TWS validation can be used as a proxy for groundwater validation. Comparison of TWS anomalies between LHF simulations and GRACE data (Figure 2-2) shows a high level of agreement for the basin- averaged anomalies and for most of the eight subbasin-averaged anomalies. However, at some of the subbasins there are some differences which are likely caused by the biases in forcing data, imperfect model parameterizations, and potential biases in GRACE data for small subbasins (Chaudhari et al., 2018, 2019; Felfelani et al., 2017; Longuevergne et al., 2010) such as Tocantins. However, in all subbasins the simulated TWS follows the patterns of precipitation anomalies (grey bars in Figure 2-2), further suggesting that some of the discrepancies could be attributed to precipitation biases. The model performs better in the first half of the simulation period in comparison to the second half, especially in the western subbasins including Solimoes and Japura, which could be partially attributed to the decreasing trend in precipitation in the first half of the simulation period (Figure 2-3). 34 Figure 2-2. Validation of TWS anomalies obtained from LHF simulations against GRACE anomalies (CSR Mascons) for the entire ARB and its eight subbasins for the period of 2002-2020. Basin and subbasin-averaged precipitation anomalies are obtained from ERA5 dataset (grey bars). Seasonal cycles of GRACE and simulated TWS and its components are shown in the right panel of each time series. GRACE results are shown as the mean of mascon solutions and simulated TWS anomalies are calculated with respect to the anomaly window of 2004-2009 for consistency with GRACE. 35 Figure 2-3. The absolute change in decadal mean of the spatial distribution of annual precipitation (Panels A to C) and net radiation (Panels D to F) over the preceding three decades (Hersbach et al., 2020). The absolute changes in decadal mean in the 2000s (2001 to 2010) from the 1990s (1992 to 2000) and in 2010s (2011 to 2020) from 2000s are showcased in the respective panels. The model simulates the seasonality of TWS very well in comparison to GRACE (low RSR and high R2; Figure 2-2), which adds more confidence to the results of this study as the seasonality is a key focus area in this study (Figure 2-2). The seasonal cycle of TWS components shows the dominant role of groundwater storage in governing TWS changes in the majority of the subbasins, especially in the subbasins with relatively deep groundwater (WTD>2m) such as Tocantins, Tapajos, and Xingu. However, in the subbasins where the groundwater is relatively shallow, flood water storage plays an equally prominent role in modulating TWS anomalies (e.g., Solimoes, Purus, and Negro). We note that because soil moisture storage in LHF is defined as the moisture above WTD, the seasonal cycles of groundwater and soil moisture storage have an inverse relationship. Decadal mean of the spatial distribution of ET is validated against MODIS MOD16A3GF Version 6.1 (Running et al., 2021) product, a year-end and gap-filled yearly composite dataset 36 produced at 500m resolution for the period of 2000-present (Figure 2-4A). The comparison of long-term annual mean of ET between the LHF simulation results (Figure 2-4B) and MODIS data shows a good agreement (Figure 2-4D). However, there are notable differences over certain areas which include flood-dominated regions, grasslands, and shrublands (Figure 2-5). These differences could be attributed to the differences in the way ET is estimated. The annual MODIS ET was derived based on the Penman-Monteith equation (Monteith, 1965), which includes inputs of daily meteorological reanalysis data along with MODIS remotely sensed data products such as dynamic vegetation properties, albedo, and land cover (Running et al., 2021). ET in LHF is, however, calculated based on energy balance approach (Miguez-Macho et al., 2007; Walko et al., 2000). Another source of discrepancy is the different in spatial resolution between the two products; model results could have higher uncertainties in regions within waterbodies including, river channels, lakes, and wetlands, where MODIS product might have accurately captured the ET dynamics. In addition to MODIS ET, the long-term annual mean of ET is compared with The Global Land Evaporation Amsterdam Model (GLEAM; Figure 2-4C) V3.8a (Martens et al., 2017; Miralles et al., 2011) product, a set of algorithms to estimate daily components of land evaporation at 0.25 degree grid cell from satellite and reanalysis data for the period of 1980- present based on the Priestly and Tylor equation (Priestley & Taylor, 1972) and Gash’s analytical model (Gash, 1979). The comparison of long-term annual mean of ET between the LHF simulation results and GLEAM data shows a better agreement than MODIS over most of the ARB (Figure 2-4E). However, along the northern boundary of the ARB, the comparison shows more discrepancy in comparison to MODIS. A further investigation by comparing MODIS and GLEAM datasets shows that there are notable discrepancies even between the two datasets (Figure 2-4F), making it difficult to draw a clear conclusion on model performance. In 37 general, given that MODIS and GLEAM ET are also estimates—not true observations—that is known to include uncertainties (Xu et al., 2019), these comparisons demonstrate that the simulated ET is not out of bounds (Figure 2-4) and add further confidence to the results of this study. Figure 2-4. Decadal mean of the spatial distribution of annual ET in the past two decades from LHF model (A), MODIS (Running et al., 2021) dataset (B), and GLEAM (Martens et al., 2017) (Miralles et al., 2011) dataset (C). The relative errors (%) in LHF results compared to MODIS and GLEAM are shown in panels D and E, respectively. Panel F shows the relative difference between MODIS and GLEAM. 38 Figure 2-5. Land use and land cover (LULC) maps for 1992, 2000 and 2010 are shown in the top panels (Santoro et al., 2017). The changes in LULC in the 1990s (1992 to 2000), 2010s (2001 to 2010) and 2010s (2011 to 2020) are shown in bottom panels. The lower colorbar shows the initial and final LULC type and in the colorbar. F, S, C, and W stands for Forest, Shrubland, Cropland and Water, respectively. Overall, the evaluation of river discharge, TWS, and ET with multiple independent products, in addition to a substantial validation presented in multiple previous studies (Chaudhari et al., 2019; Miguez-Macho & Fan, 2012; Pokhrel et al., 2013, 2014), presents sufficient basis on the usefulness of the model to study the dominant hydrological processes and their spatial and temporal variability across the ARB. Further, given that the model is fully physically based, not calibrated with observations, and applied over a large domain of the ARB, we consider the model performance to be satisfactory for our application. Dynamics of Key Hydrological Processes In this section we address the first research question by presenting an in-depth analysis of the dynamics of key hydrological processes within the ARB over the last three decades with a focus on various water and energy balance components and associating the changes in these 39 components with major hydrologic drivers. The dynamics of the hydrological functioning of the basin, driven by climatic and anthropogenic factors, have substantial implications for forest and ecosystem management of the ARB. Exploring the shifts in these processes provides insights into the basin's sensitivity to climatic change, the impacts of human activities, and overall forest resilience. The subsequent sub-sections unravel the details of each hydrological component, shedding light on the interplay among them and implications for forest management. 2.3.2.1 Dynamics of Groundwater Mechanisms The hydrological dynamics within the ARB have undergone substantial changes over the past three decades (Figure 2-6). The variations in WTD across the ARB exhibit a remarkable spatial heterogeneity from east to west, serving diverse functions and roles (Figure 2-6F). The WTD varies, with shallow water tables (WTD<5m) found predominantly in the central and northwestern regions, and deeper water tables (5m5m) in the headwater catchments of the ARB emerges as the primary factor contributing to streamflow in these upstream areas (Miguez-Macho and Fan, 2012). Regions with 5m 60%) predominantly occurs in regions receiving an average of more than 1500mm of annual rainfall. Conversely, 112 savanna cover (5% < tree coverage < 60%) predominates in regions with annual rainfall less than 1500mm. In addition, the tree coverages between 20% to 60% are rare over the ARB, indicating existence of stable and unstable states in tree coverage. These results confirm the findings in the previous studies (Hirota et al., 2011; Staver et al., 2011). The probability density function of precipitation at the subbasin scale shows that the behavior is generally consistent with that of the ARB, however, over some subbasins the distribution is not bimodal (e.g., Purus) or the savanna state happens at a higher tree cover percentage (e.g., Japura). Tocantins does not receive high rates of rainfall and it dominantly falls between 1200 to 2100 mm/year; as a result, only savanna state is present in the probability density function. Negro, Solimoes, Madeira, Tapajos, and Japura subbasins show similar behavior to the ARB, however, the cutoffs for savanna and forest states happen at different precipitation levels over these subbasins. For Negro and Solimoes the cutoff between the two stable states happens around 1800 mm/year, for Madeira and Tapajos it happens around 1200 mm/year, and finally for Japura it occurs around 900 mm/year. Therefore, Japura is the most resilient subbasin in terms of sustaining the forest state. The distribution over Xingu shows that precipitation rates between 1500 to 2400 mm/year result in higher tree coverage in comparison to 2400 to 3000 mm/year, which does not follow the other subbasins. Over Purus only forest state is represented in the distribution because this subbasins generally receives more than 1500 mm/year of precipitation. 113 Figure 4-3. The probability density function constructed from 1000 samples of the arcsine transformed data described as the weighted sum for precipitation over the ARB and its subbasins. Examining the probability density function of various water table levels reveals that forest cover predominantly occurs where WTD is less than 20m, while savanna cover is prevalent in regions with WTD deeper than 20m (Figure 4-4). The probability density function 114 of WTD at subbasin scale shows that the behavior is generally consistent with that of the ARB, however, Tocantins exhibits a contrasting behavior where conditions favoring forest cover occur in regions typically dominated by savanna. Similar to the probability density function of precipitation, Negro, Solimoes, Madeira, Tapajos, and Japura subbasins show similar behavior to that of the ARB, however, the cutoffs for savanna and forest states happen at different water table levels over these subbasins. The WTD over Negro, Tapajos, Xingu, and Japura subbasins is resilient against favoring savanna state as even at WTD deeper than 40m the distribution favors the forest state. Over Solimoes, WTD<20m favors the forest state and WTD around 40m defines the cut off level between forest and savanna state over Madeira. Conversely, over Madeira, WTD>1m are more suitable for forest state than WTD<1m. The reason for this phenomenon needs further investigation. All water table levels over the Purus favor the forest state. 115 Figure 4-4. The probability density function constructed from 1000 samples of the arcsine transformed data described as the weighted sum for WTD over the ARB and its subbasins. To investigate the contrasting behavior of the Tocantins subbasin in comparison to the other subbasins, a focused analysis was conducted on grid cells within two random boxes in Tocantins and the western region of the basin (Figure 4-5). Specifically, WTD trajectories were 116 examined for different tree cover percentages, along with an evaluation of the slope of the change in WTD against the changes in tree cover percentages. The findings revealed a direct relationship between WTD and tree cover change within the western region of the basin. Conversely, over Tocantins, a reduction in tree cover did not substantially deepen the WTD. This behavior in Tocantins could potentially induce oxidative stress for the forest cover in regions with shallow WTDs, offering an explanation for the absence of forest cover in these areas. However, further comprehensive assessments of additional components are necessary to validate and confirm this observed behavior. 117 Figure 4-5. The interplay of WTD and tree coverage for two random regions in the forested regions (west Amazon box) and savanna regions (Tocantins box). 4.4 Conclusion The comprehensive analysis of the ARB highlights a significant correlation between long-term average precipitation and WTD across randomly sampled grid cells. The study suggests the presence of potential tipping points linked to WTD dynamics, akin to those 118 associated with varying levels of precipitation over tropical forests globally. The relationship between WTD and tree cover demonstrates that as the water table deepens beyond 20 meters, intermediate values of tree coverage decrease, indicating specific requirements for tree survival in drier or deeper conditions. Further investigation reveals a bimodal probability density function for precipitation, emphasizing that forest cover predominantly occurs in regions receiving over 1500mm of annual rainfall, while savanna cover prevails in drier regions. A reduction in areas conducive to forest coverage due to declining rainfall levels over the past two decades raises concerns. Similarly, the analysis of WTD levels indicates that forest cover is prominent where WTD is less than 5m, emphasizing the importance of shallow water table depths for forest ecosystems. However, a reduction in these favorable regions for forest cover is observed over the same period, suggesting potential discrepancies in existing tipping point assessments. At a sub-basin scale, the Tocantins subbasin exhibits unique behavior, favoring savanna cover in regions where forest cover is typically expected. The focused analysis of this subbasin reveals a distinct relationship between WTD and tree cover change, offering insights into the potential challenges faced by the forest cover in regions with shallow water table depths. Further research and a more comprehensive assessment of various components are essential to validate and confirm this observed behavior and to better understand the dynamics and complexities of forest-savanna transitions in the ARB. 119 5. Chapter 5 Summary and Conclusion The hydrology of the ARB has been extensively studied; however, critical gaps remain in understanding key processes governing hydrological dynamics and rainforest resilience, disentangling climate and LULC change impacts, and investigating tipping points associated with dominant hydrological processes. This inhibits the understanding of hydrological considerations needed for sustainable forest management under climatic change and growing human stressors. In Chapter 2, using high resolution (~2km), long-term simulations from a process-based hydrological model (LEAF-Hydro-Flood) and an innovative groundwater area fraction analysis, the dominant hydrological processes across the ARB, their key roles in shaping basin functions, and the decadal evolution therein are investigated. Results indicated that shallow groundwater (<5m deep) strongly modulates the seasonality of the surface fluxes across the ARB. The results indicated that at least 34% of the Amazonian Forest is supported by groundwater during the dry season. A two-month lag between the seasonal peak of ET and river discharge is a key mechanism that potentially prevents the rainforest from tipping into savanna. The ARB is dominantly energy limited; however, the results suggest that in the absence of groundwater support, and with less than ~125 mm/month of precipitation, the ARB could have become water- limited, at least in some regions. The long-term basin-averaged ET—dominated by transpiration—changed with a split pattern of ±9% in the past three decades. Similarly, water table depth (±19%) and runoff (±29%) changed with a heterogenous patterns across the ARB. River discharge did not change substantially due to the crucial buffering role of groundwater. Terrestrial water storage (TWS) decreased (increased) in the 2000s (2010s) compared to that in the 1990s. Although groundwater is the dominant contributor to total TWS, the dynamics of 120 TWS over the major river channels are controlled by flood water, given relatively shallow groundwater. The chapter provides crucial insights on the dominant hydrological processes in the ARB to inform forest management practices. In Chapter 3, state-of-the-art model LEAF-Hydro-Flood (LHF), together with developing static and dynamics land use scenarios are used to disentangle the impacts of climate and LULC change. The results showed that despite extensive deforestation, climate variability remains the dominant influence on WTD dynamics; however, the impacts on ET varied across the basin. Runoff patterns were intricately tied to precipitation and water table dynamics, demonstrating regional variations influenced by both climate variability and LULC changes. This chapter provides key insights on the separate role of climate variability and LULC change in the altered water cycle of the ARB over the past four decades. In Chapter 4, using the simulated WTD from LHF, tree cover from MODIS VCF and precipitation from ERA5 dataset, potential tipping points associated with groundwater over the ARB are investigated. The area fraction analysis of WTD seasonality confirms the existence of tipping points. Further investigation reveals a bimodal probability density function for precipitation and WTD. Emphasizing that forest cover predominantly occurs in regions receiving over 1500mm of annual rainfall and/or where WTD is less than 5m. A reduction in areas conducive to forest coverage due to declining rainfall levels or deepening WTD over the past two decades raises concerns. The different amount of reduction based on rainfall and WTD suggests potential discrepancies in existing tipping point assessments. This chapter provides key insights on the resilience of the Amazonian Forest and highlights the importance of sustainable thresholds in forest management practices. 121 6. REFERENCES Abril, G., Martinez, J.-M., Artigas, L. F., Moreira-Turcq, P., Benedetti, M. F., Vidal, L., et al. (2014). Amazon River carbon dioxide outgassing fuelled by wetlands. Nature, 505(7483), 395–398. Amaral-Zettler, L. A., Zettler, E. R., & Mincer, T. J. (2020). Ecology of the plastisphere. Nature Reviews Microbiology, 18(3), 139–151. Anderson, A. B., & Ioris, E. M. (1992). Valuing the rain forest: economic strategies by small- scale forest extractivists in the Amazon estuary. Human Ecology, 20, 337–369. Andrade-Filho, J. D., Scholte, R. G. C., Amaral, A. L. G., Shimabukuro, P. H. F., Carvalho, O. dos S., & Caldeira, R. L. (2017). Occurrence and probability maps of Lutzomyia longipalpis and Lutzomyia cruzi (Diptera: Psychodidae: Phlebotominae) in Brazil. Journal of Medical Entomology, 54(5), 1430–1434. Aragão, L. C. E. O. (2012). The rainforest’s rain pump. Nature, 489(7415), 217–218. Aragão, L. E. O. C., Poulter, B., Barlow, J. B., Anderson, L. O., Malhi, Y., Saatchi, S., et al. (2014). Environmental change and the carbon balance of Amazonian forests. Biological Reviews, 89(4), 913–931. https://doi.org/10.1111/brv.12088 Aragão, L. E. O. C., Anderson, L. O., Fonseca, M. G., Rosan, T. M., Vedovato, L. B., Wagner, F. H., et al. (2018). decline of Amazon deforestation carbon emissions. Nature Communications, 9(1), 1–12. Retrieved from http://dx.doi.org/10.1038/s41467-017-02771-y Armijos, E., Crave, A., Espinoza, J. C., Filizola, N., Espinoza-Villar, R., Fonseca, P., et al. (2020). Rainfall control on Amazon sediment flux: synthesis from 20 years of monitoring. Environmental Research Communications, 2(5), 51008. Arvor, D., Funatsu, B. M., Michot, V., & Dubreui, V. (2017). Monitoring rainfall patterns in the southern amazon with PERSIANN-CDR data: Long-term characteristics and trends. Remote Sensing, 9(9), 889. https://doi.org/10.3390/rs9090889 Bagheri, O., Pokhrel, Y., Moore, N., & Phanikumar, M. S. (2024). Groundwater dominates terrestrial hydrological processes in the Amazon at the basin and subbasin scales. Journal of Hydrology, 628, 130312. Bagley, J. E., Desai, A. R., Harding, K. J., Snyder, P. K., & Foley, J. A. (2014). Drought and deforestation: Has land cover change influenced recent precipitation extremes in the Amazon? Journal of Climate, 27(1), 345–361. https://doi.org/10.1175/JCLI-D-12-00369.1 Bala, G., & Nag, B. (2012). Albedo enhancement over land to counteract global warming: impacts on hydrological cycle. Climate Dynamics, 39, 1527–1542. Barichivich, J., Gloor, E., Peylin, P., Brienen, R. J. W., Schöngart, J., Espinoza, J. C., & Pattnayak, K. C. (2018). Recent intensification of Amazon flooding extremes driven by 122 strengthened Walker circulation. Science Advances, 4(9), eaat8785. Barkhordarian, A., Saatchi, S. S., Behrangi, A., Loikith, P. C., & Mechoso, C. R. (2019). A Recent Systematic Increase in Vapor Pressure Deficit over Tropical South America. Scientific Reports, 9(1), 1–12. https://doi.org/10.1038/s41598-019-51857-8 Barlow, J., & Peres, C. A. (2008). Fire-mediated dieback and compositional cascade in an Amazonian forest. Philosophical Transactions of the Royal Society B: Biological Sciences, 363(1498), 1787–1794. Barlow, J., Lennox, G. D., Ferreira, J., Berenguer, E., Lees, A. C., Nally, R. Mac, et al. (2016). Anthropogenic disturbance in tropical forests can double biodiversity loss from deforestation. Nature, 535(7610), 144–147. Barreto, J. R., Berenguer, E., Ferreira, J., Joly, C. A., Malhi, Y., de Seixas, M. M. M., & Barlow, J. (2021). Assessing invertebrate herbivory in human‐modified tropical forest canopies. Ecology and Evolution, 11(9), 4012–4022. Bäse, F., Elsenbeer, H., Neill, C., & Krusche, A. V. (2012). Differences in throughfall and net precipitation between soybean and transitional tropical forest in the southern Amazon, Brazil. Agriculture, Ecosystems & Environment, 159, 19–28. Bates, P. D., Horritt, M. S., & Fewtrell, T. J. (2010). A simple inertial formulation of the shallow water equations for efficient two-dimensional flood inundation modelling. Journal of Hydrology, 387(1–2), 33–45. https://doi.org/10.1016/j.jhydrol.2010.03.027 Berenguer, E., Ferreira, J., Gardner, T. A., Aragão, L. E. O. C., De Camargo, P. B., Cerri, C. E., et al. (2014). A large‐scale field assessment of carbon stocks in human‐modified tropical forests. Global Change Biology, 20(12), 3713–3726. Beven, K. J., & Kirkby, M. J. (1979). A physically based, variable contributing area model of basin hydrology. Hydrological Sciences Bulletin, 24(1), 43–69. https://doi.org/10.1080/02626667909491834 Bonell, M., Purandara, B. K., Venkatesh, B., Krishnaswamy, J., Acharya, H. A. K., Singh, U. V, et al. (2010). The impact of forest use and reforestation on soil hydraulic conductivity in the Western Ghats of India: Implications for surface and sub-surface hydrology. Journal of Hydrology, 391(1–2), 47–62. Bosmans, J. H. C., Van Beek, L. P. H., Sutanudjaja, E. H., & Bierkens, M. F. P. (2017). Hydrological impacts of global land cover change and human water use. Hydrology and Earth System Sciences, 21(11), 5603–5626. https://doi.org/10.5194/hess-21-5603-2017 Brando, P. M., Nepstad, D. C., Balch, J. K., Bolker, B., Christman, M. C., Coe, M., & Putz, F. E. (2012). Fire‐induced tree mortality in a neotropical forest: the roles of bark traits, tree size, wood density and fire behavior. Global Change Biology, 18(2), 630–641. Brown, E., Johansen, I. C., Bortoleto, A. P., Pokhrel, Y., Chaudhari, S., Cak, A., et al. (2022). 123 Feasibility of hybrid in-stream generator–photovoltaic systems for Amazonian off-grid communities. PNAS Nexus, 1(3), pgac077. https://doi.org/10.1093/pnasnexus/pgac077 Bruno, R. D., Da Rocha, H. R., De Freitas, H. C., Goulden, M. L., & Miller, S. D. (2006). Soil moisture dynamics in an eastern Amazonian tropical forest. Hydrological Processes: An International Journal, 20(12), 2477–2489. Builes-Jaramillo, A., Marwan, N., Poveda, G., & Kurths, J. (2018). Nonlinear interactions between the Amazon River basin and the Tropical North Atlantic at interannual timescales. Climate Dynamics, 50, 2951–2969. Bullock, E. L., Woodcock, C. E., Souza, C., & Olofsson, P. (2020). Satellite-based estimates reveal widespread forest degradation in the Amazon. Global Change Biology, 26(5), 2956– 2969. https://doi.org/10.1111/gcb.15029 Bustamante, M. G., Cruz, F. W., Vuille, M., Apaéstegui, J., Strikis, N., Panizo, G., et al. (2016). Holocene changes in monsoon precipitation in the Andes of NE Peru based on δ18O speleothem records. Quaternary Science Reviews, 146, 274–287. Butt, N., De Oliveira, P. A., & Costa, M. H. (2011). Evidence that deforestation affects the onset of the rainy season in Rondonia, Brazil. Journal of Geophysical Research Atmospheres, 116(11). https://doi.org/10.1029/2010JD015174 Callede, J., Cochonneau, G., VIEIRA, A. F., Guyot, J.-L., Guimaraes, V. S., & De Oliveira, E. (2010). The river amazon water contribution to the atlantic Ocean. J Water Sci, 23(3), 247– 273. Campos-Silva, J. V, Peres, C. A., Hawes, J. E., Haugaasen, T., Freitas, C. T., Ladle, R. J., & Lopes, P. F. M. (2021). Sustainable-use protected areas catalyze enhanced livelihoods in rural Amazonia. Proceedings of the National Academy of Sciences, 118(40), e2105480118. Cerri, C. E. P., Sparovek, G., Bernoux, M., Easterling, W. E., Melillo, J. M., & Cerri, C. C. (2007). Tropical agriculture and global warming: impacts and mitigation options. Scientia Agricola, 64, 83–99. Chagas, V. B. P., Chaffe, P. L. B., & Blöschl, G. (2022). Climate and land management accelerate the Brazilian water cycle. Nature Communications, 13(1), 5136. Chaudhari, S., & Pokhrel, Y. (2022). Alteration of River Flow and Flood Dynamics by Existing and Planned Hydropower Dams in the Amazon River Basin. Water Resources Research, 58(5), e2021WR030555. https://doi.org/10.1029/2021WR030555 Chaudhari, S., Felfelani, F., Shin, S., & Pokhrel, Y. (2018). Climate and anthropogenic contributions to the desiccation of the second largest saline lake in the twentieth century. Journal of Hydrology, 560, 342–353. https://doi.org/10.1016/j.jhydrol.2018.03.034 Chaudhari, S., Pokhrel, Y., Moran, E., & Miguez-Macho, G. (2019). Multi-decadal hydrologic change and variability in the Amazon River basin: Understanding terrestrial water storage 124 variations and drought characteristics. Hydrology and Earth System Sciences, 23(7), 2841– 2862. https://doi.org/10.5194/hess-23-2841-2019 Chaudhari, S., Brown, E., Quispe-Abad, R., Moran, E., Müller, N., & Pokhrel, Y. (2021). In- stream turbines for rethinking hydropower development in the Amazon basin. Nature Sustainability, 4(8), 680–687. https://doi.org/10.1038/s41893-021-00712-8 Clark, M. P., Fan, Y., Lawrence, D. M., Adam, J. C., Bolster, D., Gochis, D. J., et al. (2015). Improving the representation of hydrologic processes in Earth System Models. Water Resources Research, 51(8), 5929–5956. https://doi.org/10.1002/2015WR017096 Costa, Marcos H., Yanagi, S. N. M., Souza, P. J. O. P., Ribeiro, A., & Rocha, E. J. P. (2007). Climate change in Amazonia caused by soybean cropland expansion, as compared to caused by pastureland expansion. Geophysical Research Letters, 34(7). https://doi.org/10.1029/2007GL029271 Costa, Marcos Heil, & Foley, J. A. (1999). Trends in the hydrologic cycle of the Amazon basin. Journal of Geophysical Research Atmospheres, 104(D12), 14189–14198. https://doi.org/10.1029/1998JD200126 Costa, Marcos Heil, & Pires, G. F. (2010a). Effects of Amazon and Central Brazil deforestation scenarios on the duration of the dry season in the arc of deforestation. International Journal of Climatology, 30(13), 1970–1979. Costa, Marcos Heil, & Pires, G. F. (2010b). Effects of Amazon and Central Brazil deforestation scenarios on the duration of the dry season in the arc of deforestation. International Journal of Climatology, 30(13), 1970–1979. https://doi.org/10.1002/joc.2048 Cox, P. M., Betts, R. A., Collins, M., Harris, P. P., Huntingford, C., & Jones, C. D. (2004). Amazonian forest dieback under climate-carbon cycle projections for the 21st century. Theoretical and Applied Climatology, 78(1–3), 137–156. https://doi.org/10.1007/s00704- 004-0049-4 Davidson, E. A., De Araüjo, A. C., Artaxo, P., Balch, J. K., Brown, I. F., Mercedes, M. M., et al. (2012). The Amazon basin in transition. Nature, 481(7381), 321–328. https://doi.org/10.1038/nature10717 DeAngelis, D. L., Post, W. M., & Travis, C. C. (2012). Positive feedback in natural systems (Vol. 15). Springer Science & Business Media. Dexter, K. G., Pennington, R. T., Oliveira-Filho, A. T., Bueno, M. L., Silva de Miranda, P. L., & Neves, D. M. (2018). Inserting tropical dry forests into the discussion on biome transitions in the tropics. Frontiers in Ecology and Evolution, 6, 104. Duponchelle, F., Isaac, V. J., Rodrigues Da Costa Doria, C., Van Damme, P. A., Herrera-R, G. A., Anderson, E. P., et al. (2021). Conservation of migratory fishes in the Amazon basin. Aquatic Conservation: Marine and Freshwater Ecosystems, 31(5), 1087–1105. https://doi.org/10.1002/aqc.3550 125 Eltahir, E. A. B., & Bras, R. L. (1994). Precipitation recycling in the Amazon basin. Quarterly Journal of the Royal Meteorological Society, 120(518), 861–880. https://doi.org/10.1002/qj.49712051806 Endo, W., Peres, C. A., & Haugaasen, T. (2016). Flood pulse dynamics affects exploitation of both aquatic and terrestrial prey by Amazonian floodplain settlements. Biological Conservation, 201, 129–136. Van Der Ent, R. J., Savenije, H. H. G., Schaefli, B., & Steele-Dunne, S. C. (2010). Origin and fate of atmospheric moisture over continents. Water Resources Research, 46(9). https://doi.org/10.1029/2010WR009127 Espinoza Villar, J. C., Ronchail, J., Guyot, J. L., Cochonneau, G., Naziano, F., Lavado, W., et al. (2009). Spatio‐temporal rainfall variability in the Amazon basin countries (Brazil, Peru, Bolivia, Colombia, and Ecuador). International Journal of Climatology: A Journal of the Royal Meteorological Society, 29(11), 1574–1594. Esquivel‐Muelbert, A., Baker, T. R., Dexter, K. G., Lewis, S. L., Brienen, R. J. W., Feldpausch, T. R., et al. (2019). Compositional response of Amazon forests to climate change. Global Change Biology, 25(1), 39–56. Fan, Y., & Miguez-Macho, G. (2010). Potential groundwater contribution to Amazon evapotranspiration. Hydrology and Earth System Sciences, 14(10), 2039–2056. https://doi.org/10.5194/hess-14-2039-2010 Fan, Y., Li, H., & Miguez-Macho, G. (2013). Global patterns of groundwater table depth. Science, 339(6122), 940–943. https://doi.org/10.1126/science.1229881 Fan, Ying, Miguez-Macho, G., Jobbágy, E. G., Jackson, R. B., & Otero-Casal, C. (2017). Hydrologic regulation of plant rooting depth. Proceedings of the National Academy of Sciences of the United States of America, 114(40), 10572–10577. https://doi.org/10.1073/pnas.1712381114 Fassoni-Andrade, A. C., Fleischmann, A. S., Papa, F., Paiva, R. C. D. de, Wongchuig, S., Melack, J. M., et al. (2021). Amazon Hydrology From Space: Scientific Advances and Future Challenges. Reviews of Geophysics, 59(4), e2020RG000728. https://doi.org/10.1029/2020RG000728 Felfelani, F., Wada, Y., Longuevergne, L., & Pokhrel, Y. N. (2017). Natural and human-induced terrestrial water storage change: A global analysis using hydrological models and GRACE. Journal of Hydrology, 553, 105–118. https://doi.org/10.1016/j.jhydrol.2017.07.048 Field, C. B., Behrenfeld, M. J., Randerson, J. T., & Falkowski, P. (1998). Primary production of the biosphere: Integrating terrestrial and oceanic components. Science, 281(5374), 237–240. https://doi.org/10.1126/science.281.5374.237 Flores, B. M., Fagoaga, R., Nelson, B. W., & Holmgren, M. (2016). Repeated fires trap Amazonian blackwater floodplains in an open vegetation state. Journal of Applied Ecology, 126 53(5), 1597–1603. Foley, J. A., DeFries, R., Asner, G. P., Barford, C., Bonan, G., Carpenter, S. R., et al. (2005). Global consequences of land use. Science, 309(5734), 570–574. https://doi.org/10.1126/science.1111772 Frappart, F., Papa, F., Güntner, A., Tomasella, J., Pfeffer, J., Ramillien, G., et al. (2019). The spatio-temporal variability of groundwater storage in the Amazon River Basin. Advances in Water Resources, 124, 41–52. Fu, R., Yin, L., Li, W., Arias, P. A., Dickinson, R. E., Huang, L., et al. (2013). Increased dry- season length over southern Amazonia in recent decades and its implication for future climate projection. Proceedings of the National Academy of Sciences of the United States of America, 110(45), 18110–18115. https://doi.org/10.1073/pnas.1302584110 Gash, J. H. C. (1979). An analytical model of rainfall interception by forests. Quarterly Journal of the Royal Meteorological Society, 105(443), 43–55. Gatti, L. V, Basso, L. S., Miller, J. B., Gloor, M., Gatti Domingues, L., Cassol, H. L. G., et al. (2021). Amazonia as a carbon source linked to deforestation and climate change. Nature, 595(7867), 388–393. Getirana, A. C. V, Boone, A., Yamazaki, D., Decharme, B., Papa, F., & Mognard, N. (2012). The hydrological modeling and analysis platform (HyMAP): Evaluation in the Amazon basin. Journal of Hydrometeorology, 13(6), 1641–1665. Ghimire, C. P., Bruijnzeel, L. A., Bonell, M., Coles, N., Lubczynski, M. W., & Gilmour, D. A. (2014). The effects of sustained forest use on hillslope soil hydraulic conductivity in the Middle Mountains of Central Nepal. Ecohydrology, 7(2), 478–495. Guan, K., Pan, M., Li, H., Wolf, A., Wu, J., Medvigy, D., et al. (2015). Photosynthetic seasonality of global tropical forests constrained by hydroclimate. Nature Geoscience, 8(4), 284–289. https://doi.org/10.1038/ngeo2382 Guilhen, J., Al Bitar, A., Sauvage, S., Parrens, M., Martinez, J.-M., Abril, G., et al. (2020). Denitrification, carbon and nitrogen emissions over the Amazonianwetlands. Biogeochemistry: Wetlands. Https://Doi. Org/10.5194/Bg-2020-3. Hassler, S. K., Zimmermann, B., van Breugel, M., Hall, J. S., & Elsenbeer, H. (2011). Recovery of saturated hydraulic conductivity under secondary succession on former pasture in the humid tropics. Forest Ecology and Management, 261(10), 1634–1642. Heerspink, B. P., Kendall, A. D., Coe, M. T., & Hyndman, D. W. (2020). Trends in streamflow, evapotranspiration, and groundwater storage across the Amazon Basin linked to changing precipitation and land cover. Journal of Hydrology: Regional Studies, 32, 100755. https://doi.org/10.1016/j.ejrh.2020.100755 Hersbach, H., B, B., P, B., S, H., A, H., J, M., et al. (2020). The ERA5 global reanalysis. 127 Quarterly Journal of the Royal Meteorological Society, 146(730), 1999–2049. Hirota, M., Holmgren, M., Van Nes, E. H., & Scheffer, M. (2011). Global resilience of tropical forest and savanna to critical transitions. Science, 334(6053), 232–235. Hoek van Dijke, A. J., Herold, M., Mallick, K., Benedict, I., Machwitz, M., Schlerf, M., et al. (2022). Shifts in regional water availability due to global tree restoration. Nature Geoscience, 15(5), 363–368. https://doi.org/10.1038/s41561-022-00935-0 Hubbell, S. P., He, F., Condit, R., Borda-de-Água, L., Kellner, J., & Ter Steege, H. (2008). How many tree species are there in the Amazon and how many of them will go extinct? Proceedings of the National Academy of Sciences, 105(supplement_1), 11498–11504. Jones, C., Lowe, J., Liddicoat, S., & Betts, R. (2009). Committed terrestrial ecosystem changes due to climate change. Nature Geoscience, 2(7), 484–487. https://doi.org/10.1038/ngeo555 Junk, W. J., Piedade, M. T. F., Schöngart, J., Cohn-Haft, M., Adeney, J. M., & Wittmann, F. (2011). A classification of major naturally-occurring Amazonian lowland wetlands. Wetlands, 31, 623–640. Kendall, M. G. (1948). Rank correlation methods. Khanna, J., Medvigy, D., Fueglistaler, S., & Walko, R. (2017). Regional dry-season climate changes due to three decades of Amazonian deforestation. Nature Climate Change, 7(3), 200–204. https://doi.org/10.1038/nclimate3226 Kirschke, S., Bousquet, P., Ciais, P., Saunois, M., Canadell, J. G., Dlugokencky, E. J., et al. (2013). Three decades of global methane sources and sinks. Nature Geoscience, 6(10), 813– 823. Kumagai, T., Kanamori, H., & Yasunari, T. (2013). Deforestation‐induced reduction in rainfall. Hydrological Processes, 27(25), 3811–3814. Lanckriet, S., Araya, T., Cornelis, W., Verfaillie, E., Poesen, J., Govaerts, B., et al. (2012). Impact of conservation agriculture on catchment runoff and soil loss under changing climate conditions in May Zeg-zeg (Ethiopia). Journal of Hydrology, 475, 336–349. Laurance, W. F., Cochrane, M. A., Bergen, S., Fearnside, P. M., Delamônica, P., Barber, C., et al. (2001). The future of the Brazilian Amazon. Science, 291(5503), 438–439. https://doi.org/10.1126/science.291.5503.438 Laurance, William F., Lovejoy, T. E., Vasconcelos, H. L., Bruna, E. M., Didham, R. K., Stouffer, P. C., et al. (2002). Ecosystem decay of Amazonian forest fragments: A 22-year investigation. Conservation Biology, 16(3), 605–618. https://doi.org/10.1046/j.1523- 1739.2002.01025.x Lavagnini, I., Badocco, D., Pastore, P., & Magno, F. (2011). Theil–Sen nonparametric regression technique on univariate calibration, inverse regression and detection limits. Talanta, 87, 128 180–188. Lawrence, D. M., Fisher, R. A., Koven, C. D., Oleson, K. W., Swenson, S. C., Bonan, G., et al. (2019). The Community Land Model Version 5: Description of New Features, Benchmarking, and Impact of Forcing Uncertainty. Journal of Advances in Modeling Earth Systems, 11(12), 4245–4287. https://doi.org/10.1029/2018MS001583 Legates, D. R., & McCabe, G. J. (1999). Evaluating the use of “goodness-of-fit” measures in hydrologic and hydroclimatic model validation. Water Resources Research, 35(1), 233– 241. https://doi.org/10.1029/1998WR900018 Leite-Filho, A. T., de Sousa Pontes, V. Y., & Costa, M. H. (2019). Effects of Deforestation on the Onset of the Rainy Season and the Duration of Dry Spells in Southern Amazonia. Journal of Geophysical Research: Atmospheres, 124(10), 5268–5281. https://doi.org/10.1029/2018JD029537 Lenton, T. M., Held, H., Kriegler, E., Hall, J. W., Lucht, W., Rahmstorf, S., & Schellnhuber, H. J. (2008). Tipping elements in the Earth’s climate system. Proceedings of the National Academy of Sciences, 105(6), 1786–1793. Levy, M. C., Lopes, A. V, Cohn, A., Larsen, L. G., & Thompson, S. E. (2018). Land use change increases streamflow across the arc of deforestation in Brazil. Geophysical Research Letters, 45(8), 3520–3530. Lewinsohn, T. M., & Prado, P. I. (2005). How many species are there in Brazil? Conservation Biology, 19(3), 619–624. Li, F., Zhang, G., & Xu, Y. J. (2014). Spatiotemporal variability of climate and streamflow in the Songhua River Basin, northeast China. Journal of Hydrology, 514, 53–64. Longo, M., Saatchi, S., Keller, M., Bowman, K., Ferraz, A., Moorcroft, P. R., et al. (2020). Impacts of degradation on water, energy, and carbon cycling of the Amazon tropical forests. Journal of Geophysical Research: Biogeosciences, 125(8), e2020JG005677. Longuevergne, L., Scanlon, B. R., & Wilson, C. R. (2010). GRACE hydrological estimates for small basins: Evaluating processing approaches on the High Plains aquifer, USA. Water Resources Research, 46(11). https://doi.org/10.1029/2009WR008564 Lovejoy, T. E., & Nobre, C. (2019). Amazon tipping point: Last chance for action. Science Advances. American Association for the Advancement of Science. https://doi.org/10.1126/sciadv.aba2949 Maeda, E. E., Ma, X., Wagner, F. H., Kim, H., Oki, T., Eamus, D., & Huete, A. (2017). Evapotranspiration seasonality across the Amazon Basin. Earth System Dynamics, 8(2), 439–454. https://doi.org/10.5194/esd-8-439-2017 Malhi, Y., Wood, D., Baker, T. R., Wright, J., Phillips, O. L., Cochrane, T., et al. (2006). The regional variation of aboveground live biomass in old‐growth Amazonian forests. Global 129 Change Biology, 12(7), 1107–1138. Malhi, Y., Roberts, J. T., Betts, R. A., Killeen, T. J., Li, W., & Nobre, C. A. (2008). Climate change, deforestation, and the fate of the Amazon. Science, 319(5860), 169–172. https://doi.org/10.1126/science.1146961 Malhi, Y., Aragão, L. E. O. C., Galbraith, D., Huntingford, C., Fisher, R., Zelazowski, P., et al. (2009). Exploring the likelihood and mechanism of a climate-change-induced dieback of the Amazon rainforest. Proceedings of the National Academy of Sciences, 106(49), 20610– 20615. Mann, H. B. (1945). Nonparametric tests against trend. Econometrica: Journal of the Econometric Society, 245–259. Mapbiomas_Amazonia. (2022). Collection 2.0 of annual maps of land cover, land use and land use changes between 1985 to 2018 in the Pan-Amazon. The Amazonian Network of Georeferenced Socio-Environmental Information (RAISG). Retrieved from https://amazonia.mapbiomas.org/ Marengo, Jose A., Souza, C. M., Thonicke, K., Burton, C., Halladay, K., Betts, R. A., et al. (2018). Changes in Climate and Land Use Over the Amazon Region: Current and Future Variability and Trends. Frontiers in Earth Science, 6, 228. https://doi.org/10.3389/feart.2018.00228 Marengo, Jose Antonio, Tomasella, J., Soares, W. R., Alves, L. M., & Nobre, C. A. (2012). Extreme climatic events in the Amazon basin Climatological and hydrological context of recent floods. Theoretical and Applied Climatology, 107(1–2), 73–85. Marengo, José Antonio, & Espinoza, J. C. (2016). Extreme seasonal droughts and floods in Amazonia: causes, trends and impacts. International Journal of Climatology, 36(3), 1033– 1050. Martens, B., Miralles, D. G., Lievens, H., Van Der Schalie, R., De Jeu, R. A. M., Fernández- Prieto, D., et al. (2017). GLEAM v3: Satellite-based land evaporation and root-zone soil moisture. Geoscientific Model Development, 10(5), 1903–1925. Matricardi, E. A. T., Skole, D. L., Costa, O. B., Pedlowski, M. A., Samek, J. H., & Miguel, E. P. (2020). Long-term forest degradation surpasses deforestation in the Brazilian Amazon. Science, 369(6509), 1378–1382. https://doi.org/10.1126/SCIENCE.ABB3021 Mattos, C. R. C., Hirota, M., Oliveira, R. S., Flores, B. M., Miguez-Macho, G., Pokhrel, Y., & Fan, Y. (2023). Double stress of waterlogging and drought drives forest–savanna coexistence. Proceedings of the National Academy of Sciences, 120(33), e2301255120. Melack, J. M., Hess, L. L., Gastil, M., Forsberg, B. R., Hamilton, S. K., Lima, I. B. T., & Novo, E. M. L. M. (2004). Regionalization of methane emissions in the Amazon Basin with microwave remote sensing. Global Change Biology, 10(5), 530–544. 130 Mercedes, G., & Montenegro, L. (2005). Globalwarming and Tropical Land-Use Change: Greenhouse Gas Emissions Frombiomass Burning, Decomposition and Soils in Forest Conversion, Shifting Cultivation and Secondaryvegetation. Climatic Change, 46(1), 115– 158. Miguez-Macho, G., & Fan, Y. (2012). The role of groundwater in the Amazon water cycle: 2. Influence on seasonal soil moisture and evapotranspiration. Journal of Geophysical Research Atmospheres, 117(15). https://doi.org/10.1029/2012JD017540 Miguez-Macho, G., Fan, Y., Weaver, C. P., Walko, R., & Robock, A. (2007). Incorporating water table dynamics in climate modeling: 2. Formulation, validation, and soil moisture simulation. Journal of Geophysical Research Atmospheres, 112(13). https://doi.org/10.1029/2006JD008112 Miralles, D. G., Holmes, T. R. H., De Jeu, R. A. M., Gash, J. H., Meesters, A., & Dolman, A. J. (2011). Global land-surface evaporation estimated from satellite-based observations. Hydrology and Earth System Sciences, 15(2), 453–469. Mishra, V., Cherkauer, K. A., Niyogi, D., Lei, M., Pijanowski, B. C., Ray, D. K., et al. (2010). A regional scale assessment of land use/land cover and climatic changes on water and energy cycle in the upper Midwest United States. International Journal of Climatology, 30(13), 2025–2044. Monteith, J. L. (1965). Evaporation and environment. In Symposia of the society for experimental biology (Vol. 19, pp. 205–234). Cambridge University Press (CUP) Cambridge. Moore, N., Arima, E., Walker, R., & Ramos da Silva, R. (2007). Uncertainty and the changing hydroclimatology of the Amazon. Geophysical Research Letters, 34(14). https://doi.org/10.1029/2007GL030157 Moran, E. F. (1993). Deforestation and land use in the Brazilian Amazon. Human Ecology, 21, 1–21. Morrison, E. B., & Lindell, C. A. (2011). Active or Passive Forest Restoration? Assessing Restoration Alternatives with Avian Foraging Behavior. Restoration Ecology, 19(201), 170–177. https://doi.org/10.1111/j.1526-100X.2010.00725.x Morton, D. C., DeFries, R. S., Shimabukuro, Y. E., Anderson, L. O., Arai, E., Del Bon Espirito- Santo, F., et al. (2006). Cropland expansion changes deforestation dynamics in the southern Brazilian Amazon. Proceedings of the National Academy of Sciences of the United States of America, 103(39), 14637–14641. https://doi.org/10.1073/pnas.0606377103 Muma, M., Assani, A. A., Landry, R., Quessy, J.-F., & Mesfioui, M. (2011). Effects of the change from forest to agriculture land use on the spatial variability of summer extreme daily flow characteristics in southern Quebec (Canada). Journal of Hydrology, 407(1–4), 153– 163. 131 Nagy, L., Artaxo, P., & Forsberg, B. R. (2016). Interactions between biosphere, atmosphere, and human land use in the Amazon basin: an introduction. Interactions Between Biosphere, Atmosphere and Human Land Use in the Amazon Basin, 3–15. Neill, C., Coe, M. T., Riskin, S. H., Krusche, A. V, Elsenbeer, H., Macedo, M. N., et al. (2013). Watershed responses to Amazon soya bean cropland expansion and intensification. Philosophical Transactions of the Royal Society B: Biological Sciences, 368(1619), 20120425. Nepstad, D. C., Stickler, C. M., Filho, B. S.-, & Merry, F. (2008). Interactions among Amazon land use, forests and climate: prospects for a near-term forest tipping point. Philosophical Transactions of the Royal Society B: Biological Sciences, 363(1498), 1737–1746. van Nes, E. H., Arani, B. M. S., Staal, A., van der Bolt, B., Flores, B. M., Bathiany, S., & Scheffer, M. (2016). What do you mean,‘tipping point’? Trends in Ecology & Evolution, 31(12), 902–904. Nobre, Carlos A., Sellers, P. J., & Shukla, J. (1991). Amazonian Deforestation and Regional Climate Change. Journal of Climate, 4(10), 957–988. https://doi.org/10.1175/1520- 0442(1991)004<0957:adarcc>2.0.co;2 Nobre, Carlos A., Sampaio, G., Borma, L. S., Castilla-Rubio, J. C., Silva, J. S., & Cardoso, M. (2016). Land-use and climate change risks in the amazon and the need of a novel sustainable development paradigm. Proceedings of the National Academy of Sciences of the United States of America, 113(39), 10759–10768. https://doi.org/10.1073/pnas.1605516113 Nobre, Carlos Afonso, & Borma, L. D. S. (2009). “Tipping points” for the Amazon forest. Current Opinion in Environmental Sustainability, 1(1), 28–36. https://doi.org/10.1016/j.cosust.2009.07.003 De Paiva, R. C. D., Buarque, D. C., Collischonn, W., Bonnet, M., Frappart, F., Calmant, S., & Bulhões Mendes, C. A. (2013). Large‐scale hydrologic and hydrodynamic modeling of the Amazon River basin. Water Resources Research, 49(3), 1226–1243. Pangala, S. R., Enrich-Prast, A., Basso, L. S., Peixoto, R. B., Bastviken, D., Hornibrook, E. R. C., et al. (2017). Large emissions from floodplain trees close the Amazon methane budget. Nature, 552(7684), 230–234. Parrotta, J. A., Wildburger, C., & Mansourian, S. (2012). Understanding relationships between biodiversity, carbon, forests and people: the key to achieving REDD+ objectives. A global assessment report prepared by the Global Forest Expert Panel on Biodiversity, Forest Management and REDD+. (Vol. 31). IUFRO (International Union of Forestry Research Organizations) Secretariat. Pfeffer, J., Seyler, F., Bonnet, M., Calmant, S., Frappart, F., Papa, F., et al. (2014). Low‐water maps of the groundwater table in the central Amazon by satellite altimetry. Geophysical Research Letters, 41(6), 1981–1987. 132 Phillips, J. N., & Derryberry, E. P. (2017). Equivalent effects of bandwidth and trill rate: support for a performance constraint as a competitive signal. Animal Behaviour, 132, 209–215. Pielke, R. A., Cotton, W. R., Walko, R. L., Tremback, C. J., Lyons, W. A., Grasso, L. D., et al. (1992). A comprehensive meteorological modeling system-RAMS. Meteorology and Atmospheric Physics, 49(1–4), 69–91. https://doi.org/10.1007/BF01025401 Pison, I., Ringeval, B., Bousquet, P., Prigent, C., & Papa, F. (2013). Stable atmospheric methane in the 2000s: key-role of emissions from natural wetlands. Atmospheric Chemistry and Physics, 13(23), 11609–11623. Pokhrel, Y. N., Fan, Y., Miguez-Macho, G., Yeh, P. J. F., & Han, S. C. (2013). The role of groundwater in the Amazon water cycle: 3. Influence on terrestrial water storage computations and comparison with GRACE. Journal of Geophysical Research Atmospheres, 118(8), 3233–3244. https://doi.org/10.1002/jgrd.50335 Pokhrel, Y. N., Fan, Y., & Miguez-Macho, G. (2014). Potential hydrologic changes in the Amazon by the end of the 21st century and the groundwater buffer. Environmental Research Letters, 9(8), 84004. https://doi.org/10.1088/1748-9326/9/8/084004 Poorter, L., Bongers, F., Aide, T. M., Almeyda Zambrano, A. M., Balvanera, P., Becknell, J. M., et al. (2016). Biomass resilience of Neotropical secondary forests. Nature, 530(7589), 211– 214. Poorter, L., Craven, D., Jakovac, C. C., van der Sande, M. T., Amissah, L., Bongers, F., et al. (2021). Multidimensional tropical forest recovery. Science, 374(6573), 1370–1376. Poveda, G., Jaramillo, L., & Vallejo, L. F. (2014). Seasonal precipitation patterns along pathways of South American low‐level jets and aerial rivers. Water Resources Research, 50(1), 98–118. Priestley, C. H. B., & Taylor, R. J. (1972). On the assessment of surface heat flux and evaporation using large-scale parameters. Monthly Weather Review, 100(2), 81–92. Putz, F. E., & Redford, K. H. (2010). The importance of defining ‘forest’: Tropical forest degradation, deforestation, long‐term phase shifts, and further transitions. Biotropica, 42(1), 10–20. Raymond, P. A., Hartmann, J., Lauerwald, R., Sobek, S., McDonald, C., Hoover, M., et al. (2013). Global carbon dioxide emissions from inland waters. Nature, 503(7476), 355–359. Reis, A., Magne, K., Massot, S., Tallini, L. R., Scopel, M., Bastida, J., et al. (2019). Amaryllidaceae alkaloids: identification and partial characterization of montanine production in Rhodophiala bifida plant. Scientific Reports, 9(1), 8471. Richey, J. E., Melack, J. M., Aufdenkampe, A. K., Ballester, V. M., & Hess, L. L. (2002). Outgassing from Amazonian rivers and wetlands as a large tropical source of atmospheric CO2. Nature, 416(6881), 617–620. 133 Rozendaal, D. M. A., Bongers, F., Aide, T. M., Alvarez-Dávila, E., Ascarrunz, N., Balvanera, P., et al. (2019). Biodiversity recovery of Neotropical secondary forests. Science Advances, 5(3), eaau3114. Running, S., Mu, Q., Zhao, M., & Moreno, A. (2021). MODIS/Terra Net Evapotranspiration 8- Day L4 Global 500m SIN Grid V061. NASA EOSDIS Land Processes DAAC. Retrieved from https://doi.org/10.5067/MODIS/MOD16A3GF.061 Rutishauser, E., Hérault, B., Baraloto, C., Blanc, L., Descroix, L., Sotta, E. D., et al. (2015). Rapid tree carbon stock recovery in managed Amazonian forests. Current Biology, 25(18), R787–R788. Saatchi, S. S., Harris, N. L., Brown, S., Lefsky, M., Mitchard, E. T. A., Salas, W., et al. (2011). Benchmark map of forest carbon stocks in tropical regions across three continents. Proceedings of the National Academy of Sciences, 108(24), 9899–9904. Salati, E., Dall’Olio, A., Matsui, E., & Gat, J. R. (1979). Recycling of water in the Amazon basin: an isotopic study. Water Resources Research, 15(5), 1250–1258. Sampaio, G., Nobre, C., Costa, M. H., Satyamurty, P., Soares-Filho, B. S., & Cardoso, M. (2007). Regional climate change over eastern Amazonia caused by pasture and soybean cropland expansion. Geophysical Research Letters, 34(17). https://doi.org/10.1029/2007GL030612 Santoro, M., Kirches, G., Wevers, J., Boettcher, M., Brockmann, C., Lamarche, C., & Defourny, P. (2017). Land cover CCI: Product user guide version 2.0. Climate Change Initiative Belgium. Satyamurty, P., da Costa, C. P. W., & Manzi, A. O. (2013). Moisture source for the Amazon Basin: a study of contrasting years. Theoretical and Applied Climatology, 111, 195–209. Scheffer, M., Carpenter, S., Foley, J. A., Folke, C., & Walker, B. (2001). Catastrophic shifts in ecosystems. Nature, 413(6856), 591–596. Schellnhuber, H. J. (2009). Tipping elements in the Earth System. Proceedings of the National Academy of Sciences of the United States of America, 106(49), 20561–20563. https://doi.org/10.1073/pnas.0911106106 Sen, P. K. (1968). Estimates of the regression coefficient based on Kendall’s tau. Journal of the American Statistical Association, 63(324), 1379–1389. Shukla, J., Nobre, C., & Sellers, P. (1990). Amazon deforestation and climate changeShukla, J., Nobre, C., Sellers, P., 1990. Amazon deforestation and climate change. Science (80-. ). 247, 1322–1325. https://doi.org/10.1126/science.247.4948.1322. Science, 247(4948), 1322– 1325. https://doi.org/10.1126/science.247.4948.1322 Silva, B. N., Khan, M., & Han, K. (2018). Towards sustainable smart cities: A review of trends, architectures, components, and open challenges in smart cities. Sustainable Cities and 134 Society, 38, 697–713. Siqueira, V. A., Paiva, R. C. D., Fleischmann, A. S., Fan, F. M., Ruhoff, A. L., Pontes, P. R. M., et al. (2018). Toward continental hydrologic-hydrodynamic modeling in South America. Hydrology and Earth System Sciences, 22(9), 4815–4842. https://doi.org/10.5194/hess-22- 4815-2018 Smith, C. C., Healey, J. R., Berenguer, E., Young, P. J., Taylor, B., Elias, F., et al. (2021). Old- growth forest loss and secondary forest recovery across Amazonian countries. Environmental Research Letters, 16(8), 85009. Soares-Filho, B., Moutinho, P., Nepstad, D., Anderson, A., Rodrigues, H., Garcia, R., et al. (2010). Role of Brazilian Amazon protected areas in climate change mitigation. Proceedings of the National Academy of Sciences of the United States of America, 107(24), 10821–10826. https://doi.org/10.1073/pnas.0913048107 Soares-Filho, B. S., Nepstad, D. C., Curran, L. M., Cerqueira, G. C., Garcia, R. A., Ramos, C. A., et al. (2006). Modelling conservation in the Amazon basin. Nature, 440(7083), 520– 523. https://doi.org/10.1038/nature04389 Souza Jr, C. M., Z. Shimbo, J., Rosa, M. R., Parente, L. L., A. Alencar, A., Rudorff, B. F. T., et al. (2020). Reconstructing three decades of land use and land cover changes in brazilian biomes with landsat archive and earth engine. Remote Sensing, 12(17), 2735. Spracklen, D. V., Arnold, S. R., & Taylor, C. M. (2012). Observations of increased tropical rainfall preceded by air passage over forests. Nature, 489(7415), 282–285. https://doi.org/10.1038/nature11390 Spracklen, Dominick V, Arnold, S. R., & Taylor, C. M. (2012). Observations of increased tropical rainfall preceded by air passage over forests. Nature, 489(7415), 282–285. Staal, A., Tuinenburg, O. A., Bosmans, J. H. C., Holmgren, M., Van Nes, E. H., Scheffer, M., et al. (2018). Forest-rainfall cascades buffer against drought across the Amazon. Nature Climate Change, 8(6), 539–543. https://doi.org/10.1038/s41558-018-0177-y Staal, A., Fetzer, I., Wang-Erlandsson, L., Bosmans, J. H. C., Dekker, S. C., van Nes, E. H., et al. (2020). Hysteresis of tropical forests in the 21st century. Nature Communications, 11(1), 1– 8. https://doi.org/10.1038/s41467-020-18728-7 Staver, A. C., Archibald, S., & Levin, S. A. (2011). The global extent and determinants of savanna and forest as alternative biome states. Science, 334(6053), 230–232. https://doi.org/10.1126/science.1210465 Sterling, S. M., Ducharne, A., & Polcher, J. (2013). The impact of global land-cover change on the terrestrial water cycle. Nature Climate Change, 3(4), 385–390. https://doi.org/10.1038/nclimate1690 Theil, H. (1950). A rank-invariant method of linear and polynomial regression analysis. 135 Indagationes Mathematicae, 12(85), 173. Tropek, R., Sedláček, O., Beck, J., Keil, P., Musilová, Z., Šímová, I., & Storch, D. (2014). Comment on “High-resolution global maps of 21st-century forest cover change.” Science, 344(6187), 850–853. https://doi.org/10.1126/science.1248753 Walker, R. T. (2020). Collision Course: Development Pushes Amazonia Toward Its Tipping Point. Environment, 63(1), 15–25. https://doi.org/10.1080/00139157.2021.1842711 Walko, R. L., Band, L. E., Baron, J., Kittel, T. G. F., Lammers, R., Lee, T. J., et al. (2000). Coupled atmosphere-biophysics-hydrology models for environmental modeling. Journal of Applied Meteorology, 39(6), 931–944. https://doi.org/10.1175/1520- 0450(2000)039<0931:CABHMF>2.0.CO;2 Weng, W., Luedeke, M., Zemp, D., Lakes, T., & Kropp, J. (2018). Aerial and surface rivers: Downwind impacts on water availability from land use changes in Amazonia. Hydrology and Earth System Sciences, 22(1), 911–927. https://doi.org/10.5194/hess-22-911-2018 Werth, D., & Avissar, R. (2005). The local and global effects of Southeast Asian deforestation. Geophysical Research Letters, 32(20), 1–4. https://doi.org/10.1029/2005GL022970 Wohl, E., Barros, A., Brunsell, N., Chappell, N. A., Coe, M., Giambelluca, T., et al. (2012). The hydrology of the humid tropics. Nature Climate Change, 2(9), 655–662. https://doi.org/10.1038/nclimate1556 Wright, J. S., Fu, R., Worden, J. R., Chakraborty, S., Clinton, N. E., Risi, C., et al. (2017). Rainforest-initiated wet season onset over the southern Amazon. Proceedings of the National Academy of Sciences, 114(32), 8481–8486. Xu, D., Agee, E., Wang, J., & Ivanov, V. Y. (2019). Estimation of evapotranspiration of Amazon rainforest using the maximum entropy production method. Geophysical Research Letters, 46(3), 1402–1412. Yamazaki, D., Kanae, S., Kim, H., & Oki, T. (2011). A physically based description of floodplain inundation dynamics in a global river routing model. Water Resources Research, 47(4). https://doi.org/10.1029/2010WR009726 Zak, R., & Nippert, J. B. (2012). Comment on “Global resilience of tropical forest and Savanna to critical transitions.” Science, 336(6081), 232–235. https://doi.org/10.1126/science.1219346 Zelazowski, P., Malhi, Y., Huntingford, C., Sitch, S., & Fisher, J. B. (2011). Changes in the potential distribution of humid tropical forests on a warmer planet. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 369(1934), 137–160. Zemp, D C, Schleussner, C.-F., Barbosa, H. M. J., Van der Ent, R. J., Donges, J. F., Heinke, J., et al. (2014). On the importance of cascading moisture recycling in South America. 136 Atmospheric Chemistry and Physics, 14(23), 13337–13359. Zemp, Delphine Clara, Schleussner, C. F., Barbosa, H. M. J., Hirota, M., Montade, V., Sampaio, G., et al. (2017). Self-amplified Amazon forest loss due to vegetation-atmosphere feedbacks. Nature Communications, 8(1), 1–10. https://doi.org/10.1038/ncomms14681 Zeng, N. (1999). Seasonal cycle and interannual variability in the Amazon hydrologic cycle. Journal of Geophysical Research Atmospheres, 104(D8), 9097–9106. https://doi.org/10.1029/1998JD200088 Zeng, N., Dickinson, R. E., & Zeng, X. (1996). Climatic impact of Amazon deforestation - A mechanistic model study. Journal of Climate, 9(4), 859–883. https://doi.org/10.1175/1520- 0442(1996)009<0859:CIOADM>2.0.CO;2 137