Given the urgency of climate change, development of fast and reliable methods is essential to understand urban building energy use in the sector that accounts for 40% of total energy use in USA. Although machine learning (ML) methods may offer promise and are less difficult to develop, discrepancy in methods, results, and recommendations have emerged that requires attention. Existing research also shows inconsistencies related to integrating climate change models into energy modeling. To address these challenges, four models: random forest (RF), extreme gradient boosting (XGBoost), single regression tree, and multiple linear regression (MLR), were developed using the Commercial Building Energy Consumption Survey dataset to predict energy use intensity (EUI) under projected heating and cooling degree days by the Intergovernmental Panel on Climate Change (IPCC) across the USA during the 21st century. The RF model provided better performance and reduced the mean absolute error by 4%, 11%, and 12% compared to XGBoost, single regression tree, and MLR, respectively. Moreover, using the RF model for climate change analysis showed that office buildings’ EUI will increase between 8.9% to 63.1% compared to 2012 baseline for different geographic regions between 2030 and 2080. One region is projected to experience an EUI reduction of almost 1.5%. Finally, good data enhance the predicting ability of ML therefore, comprehensive regional building datasets are crucial to assess counteraction of building energy use in the face of climate change at finer spatial scale.
more »
« less
Spatiotemporal trends of black walnut forest stocking under climate change
Basal area is a key measure of forest stocking and an important proxy of forest productivity in the face of climate change. Black walnut ( Juglans nigra ) is one of the most valuable timber species in North America. However, little is known about how the stocking of black walnut would change with differed bioclimatic conditions under climate change. In this study, we projected the current and future basal area of black walnut. We trained different machine learning models using more than 1.4 million tree records from 10,162 Forest Inventory and Analysis (FIA) sample plots and 42 spatially explicit bioclimate and other environmental attributes. We selected random forests (RF) as the final model to estimate the basal area of black walnut under climate change because RF had a higher coefficient of determination ( R 2 ), lower root mean square error (RMSE), and lower mean absolute error (MAE) than the other two models (XGBoost and linear regression). The most important variables to predict basal area were the mean annual temperature and precipitation, potential evapotranspiration, topology, and human footprint. Under two emission scenarios (Representative Concentration Pathway 4.5 and 8.5), the RF model projected that black walnut stocking would increase in the northern part of the current range in the USA by 2080, with a potential shift of species distribution range although uncertainty still exists due to unpredictable events, including extreme abiotic (heat, drought) and biotic (pests, disease) occurrences. Our models can be adapted to other hardwood tree species to predict tree changes in basal area based on future climate scenarios.
more »
« less
- Award ID(s):
- 1916587
- NSF-PAR ID:
- 10382642
- Date Published:
- Journal Name:
- Frontiers in Forests and Global Change
- Volume:
- 10
- ISSN:
- 2624-893X
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
more » « less
Abstract Changes in growing season climate are often the foci of research exploring forest response to climate change. By contrast, little is known about tree growth response to projected declines in winter snowpack and increases in soil freezing in seasonally snow‐covered forest ecosystems, despite extensive documentation of the importance of winter climate in mediating ecological processes. We conducted a 5‐year snow‐removal experiment whereby snow was removed for the first 4–5 weeks of winter in a northern hardwood forest at the Hubbard Brook Experimental Forest in New Hampshire,
USA . Our results indicate that adverse impacts of reduced snowpack and increased soil freezing on the physiology ofAcer saccharum (sugar maple), a dominant species across northern temperate forests, are accompanied by a 40 ± 3% reduction in aboveground woody biomass increment, averaged across the 6 years following the start of the experiment. Further, we find no indication of growth recovery 1 year after cessation of the experiment. Based on these findings, we integrate spatial modeling of snowpack depth with forest inventory data to develop a spatially explicit, regional‐scale assessment of the vulnerability of forest aboveground growth to projected declines in snowpack depth and increased soil frost. These analyses indicate that nearly 65% of sugar maple basal area in the northeastern United States resides in areas that typically experience insulating snowpack. However, under theRCP 4.5 and 8.5 emissions scenarios, we project a 49%–95% reduction in forest area experiencing insulating snowpack by the year 2099 in the northeastern United States, leaving large areas of northern forest vulnerable to these changes in winter climate, particularly along the northern edge of the region. Our study demonstrates that research focusing on growing season climate alone overestimates the stimulatory effect of warming temperatures on tree and forest growth in seasonally snow‐covered forests. -
null (Ed.)When forest conditions are mapped from empirical models, uncertainty in remotely sensed predictor variables can cause the systematic overestimation of low values, underestimation of high values, and suppression of variability. This regression dilution or attenuation bias is a well-recognized problem in remote sensing applications, with few practical solutions. Attenuation is of particular concern for applications that are responsive to prediction patterns at the high end of observed data ranges, where systematic error is typically greatest. We addressed attenuation bias in models of tree species relative abundance (percent of total aboveground live biomass) based on multitemporal Landsat and topoclimatic predictor data. We developed a multi-objective support vector regression (MOSVR) algorithm that simultaneously minimizes total prediction error and systematic error caused by attenuation bias. Applied to 13 tree species in the Acadian Forest Region of the northeastern U.S., MOSVR performed well compared to other prediction methods including single-objective SVR (SOSVR) minimizing total error, Random Forest (RF), gradient nearest neighbor (GNN), and Random Forest nearest neighbor (RFNN) algorithms. SOSVR and RF yielded the lowest total prediction error but produced the greatest systematic error, consistent with strong attenuation bias. Underestimation at high relative abundance caused strong deviations between predicted patterns of species dominance/codominance and those observed at field plots. In contrast, GNN and RFNN produced dominance/codominance patterns that deviated little from observed patterns, but predicted species relative abundance with lower accuracy and substantial systematic error. MOSVR produced the least systematic error for all species with total error often comparable to SOSVR or RF. Predicted patterns of dominance/codominance matched observations well, though not quite as well as GNN or RFNN. Overall, MOSVR provides an effective machine learning approach to the reduction of systematic prediction error and should be fully generalizable to other remote sensing applications and prediction problems.more » « less
-
Increasing tropical cyclone (TC) pressure on temperate forests is inevitable under the recent global increase of the in- tensity and poleward migration of TCs. However, the long-term effects of TCs on large-scale structure and diversity of temperate forests remain unclear. Here, we aim to ascertain the legacy of TCs on forest structure and tree species rich- ness by using structural equation models that consider several environmental gradients and use an extensive dataset containing >140,000 plots with >3 million trees from natural temperate forests across eastern United States impacted by TCs. We found that high TC activity (a combination of TC frequency and intensity) leads to a decrease in maximum tree sizes (height and diameter), an increase in tree density and basal area, and a decline in the number of tree species and recruits. We identified TC activity as the strongest predictor of forest structure and species richness in xeric (dry) forests, while it had a weaker impact on hydric (wet) forests. We highlight the sensitivity of forest structure and tree species richness to impacts of likely further increase of TC activity in interaction with climate extremes, especially drought. Our results show that increased TC activity leads to the homogenization of forest structure and reduced tree species richness in U.S. temperate forests. These findings suggest that further declines in tree species richness may be expected because of the projected increase of future levels of TC activity.more » « less
-
more » « less
Abstract Phenological escape, a strategy that deciduous understory plants use to access direct light in spring by leafing out before the canopy closes, plays an important role in shaping the recruitment of temperate tree seedlings. Previous studies have investigated how climate change will alter these dynamics for herbaceous species, but there is a knowledge gap related to how woody species such as tree seedlings will be affected. Here, we modeled temperate tree seedling leaf‐out phenology and canopy close phenology in response to environmental drivers and used climate change projections to forecast changes to the duration of spring phenological escape. We then used these predictions to estimate changes in annual carbon assimilation while accounting for reduced carbon assimilation rates associated with hotter and drier summers. Lastly, we applied these estimates to previously published models of seedling growth and survival to investigate the net effect on seedling demographic performance. Our models predict that temperate tree seedlings will experience improved phenological escape and, therefore, increased spring carbon assimilation under climate change conditions. However, increased summer respiration costs will offset the gains in spring under extreme climate change leading to a net loss in annual carbon assimilation and demographic performance. Furthermore, we found that annual carbon assimilation predictions depend strongly on the species of nearby canopy tree that seedlings were planted near, with all seedlings projected to assimilate less carbon (and therefore experience worse demographic performance) when planted near
Quercus rubra canopy trees as opposed toAcer saccharum canopy trees. We conclude that changes to spring phenological escape will have important effects on how tree seedling recruitment is affected by climate change, with the magnitude of these effects dependent upon climate change severity and biological interactions with neighboring adults. Thus, future studies of temperate forest recruitment should account for phenological escape dynamics in their models.
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.3389/ffgc.2022.970379
