Machine-learning emulation of a forest biogeochemistry model for efficient biosphere optimization

<div> <div> <div> <p>Management controls the spatial configuration of a number of landscapes globally, from forests to rangelands. The majority of landcover change and all land-use change is the result of human decision-making. As human populations and global temperatures continue to increase, an engineering approach is needed to ensure the persistence of biological diversity and natural capital critical to human well-being. Such an approach may be based on manipulating ecosystems to achieve desired future states, informed by the latest simulation models. Models of the land surface are now being used to inform policy in the form of planning and management practices. This often involves the application of models that include spatial dynamics and operate at a landscape scale. The strong correspondence between the resolution and extent of modeling and management activities at this scale, and ability to efficiently simulate the decadal-to-centennial time-scales of interest, provide managers with a credible scientific tool for anticipating future land states under different scenarios. The importance of such tools to managers has grown dramatically with the challenges posed by anthropogenic climate change. As ecosystem simulation models continually improve in precision, accuracy, and robustness, we posit that models may be mathematically optimized as a basis for optimizing the management of real-world systems. Since current ecosystem simulation models are coarse approximations of highly complex and dynamic real-world systems, such optimizations should ideally account for uncertainty and physical or biochemical constraints, thereby improving the tractability of the optimization problem. In this work, we demonstrate the emulation and optimization of a forest biogeochemistry model from the SORTIE-PPA family of models. In doing so, we provide the first demonstration of the concept of biosphere optimization (Erickson 2015), which may one day be extended to include computational genetic manipulation experiments. To perform this work, we utilize the open-source Earth-systems Research and Development Environment (ERDE) library, which contains built-in functions for performing these and other analyses with land models, with a particular focus on forests.</p> </div> </div> </div>

Modeling the Effects of Global Change on Ecosystem Processes in a Tropical Rainforest

Research Highlights: Ongoing land-use change and climate change in wet tropical forests can potentially drive shifts in tree species composition, representing a change in individual species within a functional group, tropical evergreen trees. The impacts on the global carbon cycle are potentially large, but unclear. We explored the differential effects of species within this functional group, in comparison with the effects of climate change, using the Century model as a research tool. Simulating effects of individual tree species on biome-level biogeochemical cycles constituted a novel application for Century. Background and Objectives: A unique, long-term, replicated field experiment containing five evergreen tree species in monodominant stands under similar environmental conditions in a Costa Rican wet forest provided data for model evaluation. Our objectives were to gain insights about this forest’s biogeochemical cycles and effects of tree species within this functional group, in comparison with climate change. Materials and Methods: We calibrated Century, using long-term meteorological, soil, and plant data from the field-based experiment. In modeling experiments, we evaluated effects on forest biogeochemistry of eight plant traits that were both observed and modeled. Climate-change simulation experiments represented two climate-change aspects observed in this region. Results: Model calibration revealed that unmodeled soil processes would be required to sustain observed P budgets. In species-traits experiments, three separate plant traits (leaf death rate, leaf C:N, and allocation to fine roots) resulted in modeled biomass C stock changes of >50%, compared with a maximum 21% change in the climate-change experiments. Conclusions: Modeled ecosystem properties and processes in Century were sensitive to changes in plant traits and nutrient limitations to productivity. Realistic model output was attainable for some species, but unusual plant traits thwarted predictions for one species. Including more plant traits and soil processes could increase realism, but less-complex models provide an accessible means for exploring plant-soil-atmosphere interactions.

Arboreal Epiphytes in the Soil-Atmosphere Interface: How Often Are the Biggest “Buckets” in the Canopy Empty?

Arboreal epiphytes (plants residing in forest canopies) are present across all major climate zones and play important roles in forest biogeochemistry. The substantial water storage capacity per unit area of the epiphyte “bucket” is a key attribute underlying their capability to influence forest hydrological processes and their related mass and energy flows. It is commonly assumed that the epiphyte bucket remains saturated, or near-saturated, most of the time; thus, epiphytes (particularly vascular epiphytes) can store little precipitation, limiting their impact on the forest canopy water budget. We present evidence that contradicts this common assumption from (i) an examination of past research; (ii) new datasets on vascular epiphyte and epi-soil water relations at a tropical montane cloud forest (Monteverde, Costa Rica); and (iii) a global evaluation of non-vascular epiphyte saturation state using a process-based vegetation model, LiBry. All analyses found that the external and internal water storage capacity of epiphyte communities is highly dynamic and frequently available to intercept precipitation. Globally, non-vascular epiphytes spend <20% of their time near saturation and regionally, including the humid tropics, model results found that non-vascular epiphytes spend ~1/3 of their time in the dry state (0–10% of water storage capacity). Even data from Costa Rican cloud forest sites found the epiphyte community was saturated only 1/3 of the time and that internal leaf water storage was temporally dynamic enough to aid in precipitation interception. Analysis of the epi-soils associated with epiphytes further revealed the extent to which the epiphyte bucket emptied—as even the canopy soils were often <50% saturated (29–53% of all days observed). Results clearly show that the epiphyte bucket is more dynamic than currently assumed, meriting further research on epiphyte roles in precipitation interception, redistribution to the surface and chemical composition of “net” precipitation waters reaching the surface.

The role of feeding strategy in the tolerance of a terrestrial salamander (Plethodon cinereus) to biogeochemical changes in northern hardwood forests

We investigated whether the trophic ecology of an apex predator is influenced by ecosystem-level nutrient depletion. The feeding behavior and nutrient assimilation of a terrestrial salamander, Eastern Red-backed Salamander (Plethodon cinereus (Green, 1818)), was surveyed along a gradient of forest biogeochemistry. Recent studies have documented populations of these salamanders in forests with low-pH soils that were long thought to be fatal. One mechanism that may enable P. cinereus to tolerate acid-impaired habitats is its generalist life history. We sampled diet, invertebrate prey abundance, and tissue composition of P. cinereus from sites that range in calcium availability and soil pH in northern forests of North America. We found that P. cinereus consistently exhibited a generalist feeding strategy, having diverse diets closely representing resource availability. Prey abundances were unrelated to the biogeochemical gradient (excluding gastropods), indicating relatively intact food webs. Although P. cinereus at the two most acid-impaired sites consumed more prey, overall trophic strategies were consistent across the gradient. Salamander tissue composition was unrelated to variation in forest biogeochemistry, although manganese levels were elevated in the most acid-impaired forests. We suggest that a generalist feeding strategy, combined with diverse and compositionally stable food webs, facilitates tolerance by this abundant predator of the challenges imposed by acid-impaired habitats.

Nitrogen and Phosphorus Concentration in Leaf Litter and Soil in Xishuangbanna Tropical Forests: Does Precipitation Limitation Matter?

Tropical forests are generally expected to be limited by the availability of nitrogen (N) and phosphorus (P), and these nutrient limitations could be increased by changes in forest biogeochemistry due to limited precipitation. This effect is presumed to be enhanced in the forests predominated by monsoon climate. The present study examined the impacts of monthly precipitation on total N and P in leaf litter and soil of Xishuangbanna tropical forests. Litterfall and top soil were sampled from each of the five 20 × 20 m plots established in the primary (PTF) and secondary tropical forests (STF), at monthly interval for one year. Soils were strongly acidic and showed significant differences between the sites. The monthly amounts of soil and leaf litter nutrients showed great variations between the PTF and STF. Leaf litter N and P were associated with precipitation in both dry and rainy seasons. Soil N and P were not significantly related to precipitation, indicating that changes in vegetation composition and litterfall production together accounted for variation in soil N and P. Our results suggest that the precipitation limitation may affect the leaf litter N and P changes, but did not support the prediction that precipitation limitation can immediately lead to effects on soil N and P in the Xishuangbanna tropical forests.

A Forest Model Intercomparison Framework and Application at Two Temperate Forests Along the East Coast of the United States

State-of-the-art forest models are often complex, analytically intractable, and computationally expensive, due to the explicit representation of detailed biogeochemical and ecological processes. Different models often produce distinct results while predictions from the same model vary with parameter values. In this project, we developed a rigorous quantitative approach for conducting model intercomparisons and assessing model performance. We have applied our original methodology to compare two forest biogeochemistry models, the Perfect Plasticity Approximation with Simple Biogeochemistry (PPA-SiBGC) and Landscape Disturbance and Succession with Net Ecosystem Carbon and Nitrogen (LANDIS-II NECN). We simulated past-decade conditions at flux tower sites located within Harvard Forest, MA, USA (HF-EMS) and Jones Ecological Research Center, GA, USA (JERC-RD). We mined field data available from both sites to perform model parameterization, validation, and intercomparison. We assessed model performance using the following time-series metrics: Net ecosystem exchange, aboveground net primary production, aboveground biomass, C, and N, belowground biomass, C, and N, soil respiration, and species total biomass and relative abundance. We also assessed static observations of soil organic C and N, and concluded with an assessment of general model usability, performance, and transferability. Despite substantial differences in design, both models achieved good accuracy across the range of pool metrics. While LANDIS-II NECN showed better fidelity to interannual NEE fluxes, PPA-SiBGC indicated better overall performance for both sites across the 11 temporal and two static metrics tested (HF-EMS R 2 ¯ = 0.73 , + 0.07 , R M S E ¯ = 4.68 , − 9.96 ; JERC-RD R 2 ¯ = 0.73 , + 0.01 , R M S E ¯ = 2.18 , − 1.64 ). To facilitate further testing of forest models at the two sites, we provide pre-processed datasets and original software written in the R language of statistical computing. In addition to model intercomparisons, our approach may be employed to test modifications to forest models and their sensitivity to different parameterizations.

DINÂMICA DA ACUMULAÇÃO DE FITOMASSA E NUTRIENTES NA SERAPILHEIRA SOB PLANTIOS CLONAIS DE EUCALIPTO

O objetivo geral deste trabalho foi estimar a quantidade de fitomassa e nutrientes estocada na serapilheira acumulada e taxas de decomposição da serapilheira acumulada ao longo do ano em plantios com diferentes materiais genéticos de eucalipto, em áreas submetidas ou não ao desbaste florestal. O trabalho foi conduzido em Itatinga-SP. O experimento foi um delineamento de parcelas subdivididas. As parcelas foram compostas por 8 clones e um plantio semental, sendo subparcelas áreas em que foi realizado o desbaste e áreas sem desbaste. Para estimativa da serapilheira acumulada foram realizadas coletas em cada estação do ano, entre junho de 2015 e março de 2016. Após as avaliações concluiu-se que existiu um padrão de acúmulo de serapilheira associado às estações, com maiores valores na primavera e verão. A fração folhas foi a mais representativa quanto à quantidade de serapilheira acumulada no solo. Os valores de acúmulo total de serapilheira variaram pouco entre os materiais genéticos avaliados. A decomposição foi ligeiramente maior nas áreas desbastadas comparativamente às áreas sem desbaste. A sequência decrescente dos teores nas frações foi: Folhas > Galhos > Casca > Órgãos reprodutivos > Outros materiais. Quanto aos elementos a sequência foi: N > Ca > Mg > K > P.Palavras-chave: biogeoquímica florestal, serapilheira acumulada, desbaste, decomposição. DYNAMICS OF LITTER STOCK AND NUTRIENTS UNDER CLONAL EUCALYPTUS PLANTATIONS ABSTRACT: The objectives of this work were: to estimate the amount of phytomass and nutrients stored in the accumulated litter; to evaluate the seasonal variability of the amount of phytomass and nutrients in the accumulated litter and to estimate the decomposition rates of accumulated litter throughout the year. The work was conducted in Itatinga-SP. The experiment was an outline of subdivided plots. The plots were composed of 8 clones and one stallion planting, being subplots areas where thinning and non - roughing areas were performed. In order to estimate the accumulated litter, collections were made in each season of the year, between June 2015 and June 2016. After the evaluations it was concluded that there was a pattern of litter accumulation associated with the seasons, with higher values in spring and summer. The most representative leaf fraction was the amount of litter accumulated on the soil. Values of total litter accumulation were close between genetic materials. The decomposition was slightly higher in the thinned areas compared to the thinned areas. The decreasing sequence of the contents in the fractions was: Leaves> Branches> Bark> Other materials> Vegetable remains. As for the elements the sequence was: N> Ca> Mg> K> P.Keywords: forest biogeochemistry, litter stock, thinning, decomposition.

Implementation of the Perfect Plasticity Approximation with biogeochemical compartments in R

Modeling forest ecosystems is a landmark challenge in science, due to the complexity of the processes involved and their importance in predicting future planetary conditions. While there are a number of open-source forest biogeochemistry models, few papers exist detailing the software development approach used to develop these models. This has left many forest biogeochemistry models large, opaque, and/or difficult to use, typically implemented in compiled languages for speed. Here, we present a forest biogeochemistry model from the SORTIE-PPA class of models, PPA-SiBGC. Our model is based on the Perfect Plasticity Approximation with simple biogeochemistry compartments and uses empirical vegetation dynamics rather than detailed prognostic processes to drive the estimation of carbon and nitrogen fluxes. This allows our model to be used with traditional forest inventory data, making it widely applicable and simple to parameterize. We detail the conceptual design of the model as well as the software implementation in the R language for statistical computing. Our aim is to provide a useful tool for the biogeochemistry modeling community that demonstrates the importance of vegetation dynamics in biogeochemical models.

A forest model intercomparison framework and application at two temperate forests along the East Coast of the United States

Forest models often reflect the dominant management paradigm of their time. Until the late 1970s, this meant sustaining yields. Following landmark work in forest ecology, physiology, and biogeochemistry, the current generation of models is further intended to inform ecological and climatic forest management in alignment with national biodiversity and climate mitigation targets. This has greatly increased the complexity of models used to inform management, making them difficult to diagnose and understand. State-of-the-art forest models are often complex, analytically intractable, and computationally-expensive, due to the explicit representation of detailed biogeochemical and ecological processes. Different models often produce distinct results while predictions from the same model vary with parameter values. In this project, we developed a rigorous quantitative approach for conducting model intercomparisons and assessing model performance. We have applied our original methodology to compare two forest biogeochemistry models, the Perfect Plasticity Approximation with Simple Biogeochemistry (PPA-SiBGC) and Landscape Disturbance and Succession with Net Ecosystem Carbon and Nitrogen (LANDIS-II NECN). We simulated past-decade conditions at flux tower sites located within Harvard Forest, MA, USA (HF-EMS) and Jones Ecological Research Center, GA, USA (JERC-RD). We mined field data available for both sites to perform model parameterization, validation, and intercomparison. We assessed model performance using the following time-series metrics: net ecosystem exchange, aboveground net primary production, aboveground biomass, C, and N, belowground biomass, C, and N, soil respiration, and, species total biomass and relative abundance. We also assessed static observations of soil organic C and N, and concluded with an assessment of general model usability, performance, and transferability. Despite substantial differences in design, both models achieved good accuracy across the range of pool metrics. While LANDIS-II NECN showed better fidelity to interannual NEE fluxes, PPA-SiBGC indicated better overall performance for both sites across the 11 temporal and 2 static metrics tested (HF-EMS = 0.73, +0.07, = 4.84, −10.02; JERC-RD = 0.76, +0.04, = 2.69, −1.86). To facilitate further testing of forest models at the two sites, we provide pre-processed datasets and original software written in the R language of statistical computing. In addition to model intercomparisons, our approach may be employed to test modifications to forest models and their sensitivity to different parameterizations.