The rapid and accurate assessment of above ground biomass (AGB) of woody vegetation is a critical component of climate mitigation strategies, land management practices and process-based models of ecosystem function. This is especially true of semi-arid ecosystems, where the high variability in precipitation and disturbance regimes can have dramatic impacts on the global carbon budget by rapidly transitioning AGB between live and dead pools. Measuring regional AGB requires scaling ground-based measurements using remote sensing, an inherently challenging task in the sparsely-vegetated, spatially-heterogeneous landscapes characteristic of semi-arid regions. Here, we test the ability of canopy segmentation and statistic generation based on aerial LiDAR (light detection and ranging)-derived 3D point clouds to derive AGB in clumps of vegetation in a juniper savanna in central New Mexico. We show that single crown segmentation, often an error-prone and challenging task, is not required to produce accurate estimates of AGB. We leveraged the relationship between the volume of the segmented vegetation clumps and the equivalent stem diameter of the corresponding trees (R² = 0.83, p < 0.001) to drive the allometry for J. monosperma on a per segment basis. Further, we showed that making use of the full 3D point cloud from LiDAR for the generation of canopy object statistics improved that relationship by including canopy segment point density as a covariate (R² = 0.91). This work suggests the potential for LiDAR-derived estimates of AGB in spatially-heterogeneous and highly-clumped ecosystems.

Forests provide a broad set of ecosystem services, including climate regulation. Other ecosystem services can be ecosystem dependent and are in part regulated by local‐scale decision‐making. In the southwestern United States, ongoing climate change is exacerbating a legacy of fire‐exclusion that has altered forest structure and increased high‐severity wildfire risk. Management can mitigate this risk by reducing forest density and restoring frequent surface fires, but at the cost of reduced carbon stocks. We sought to quantify the role of management in building adaptive capacity to projected climate and wildfires and the carbon consequences in a forested watershed. We simulated carbon dynamics under projected climate and wildfires and two management scenarios: prioritized and optimized. The prioritized scenario involved thinning and prescribed burning in areas selected by stakeholders to mitigate high‐severity wildfire risk. The optimized scenario used the probability of high‐severity wildfires to locate thinning treatments and increased prescribed fire area burned relative to the prioritized scenario. Both scenarios reduced wildfire severity and significantly increased net photosynthesis relative to no‐management. However, the optimized scenario decreased management‐related losses by 2.4 Mg • C • ha⁻¹ and wildfire emissions by 2.9 Mg • C • ha⁻¹ relative to the prioritized scenario. By decreasing the area thinned and increasing the area burned relative to the prioritized scenario, the optimized scenario halved the time to realize a net carbon benefit relative to no‐management. Given the increasing climatic and disturbance pressures impacting southwestern forests, management will play a critical role in building adaptive capacity and ensuring the continued provision of ecosystem services.

Remote sensing is a key technology that enables us to scale up our empirical, in situ measurements of carbon uptake made at the site level. In low leaf area index ecosystems typical of semi-arid regions however, many assumptions of these remote sensing approaches fall short, given the complexities of the heterogeneous landscape and frequent disturbance. Here, we investigated the utility of remote sensing data for predicting gross primary production (GPP) in piñon-juniper woodlands in New Mexico (USA). We developed a simple model hierarchy using climate drivers and satellite vegetation indices (VIs) to predict GPP, which we validated against in situ estimates of GPP from eddy-covariance. We tested the influence of pixel size on model fit by comparing model performance when using VIs from RapidEye (5 m) and the VIs from Landsat ETM+ (30 m). We also tested the ability of the normalized difference wetness index (NDWI) and normalized difference red edge (NDRE) to improve model fits. The best predictor of GPP at the undisturbed PJ woodland was Landsat ETM+ derived NDVI (normalized difference vegetation index), whereas at the disturbed site, the red-edge VI performed best (R²_adj of 0.92 and 0.90 respectively). The RapidEye data did improve model performance, but only after we controlled for the variability in sensor view angle, which had a significant impact on the apparent cover of vegetation in our low fractional cover experimental woodland. At both sites, model performance was best either during non-stressful growth conditions, where NDVI performed best, or during severe ecosystem stress conditions (e.g., during the girdling process), where NDRE and NDWI improved model fit, suggesting the inclusion of red-edge leveraging and moisture sensitive VI in simple, data driven models can constrain GPP estimate uncertainty during periods of high ecosystem stress or disturbance.