The optical properties of secondary brown carbon (BrC) aerosols are poorly understood, hampering quantitative assessments of their impact. We propose a new method for estimating secondary source of BrC using excitation-emission matrix (EEM) fluorescence spectroscopy, combined with parallel factor analysis (PARAFAC) and partial least squares regression (PLSR). Experiments were conducted on a collection of PM2.5 samples from urban areas in five Chinese cities during winter and summer. The humic-like component with long-emission wavelengths (L-HULIS) was identified as a secondary source tracer of BrC. This was confirmed by correlating PARAFAC components with secondary organic aerosol tracers and molecular oxidation indices obtained from Fourier transform ion cyclotron resonance mass spectrometry analysis. Using L-HULIS as a secondary tracer of BrC, it was determined that the contribution of secondary sources to water-soluble BrC (WS-BrC) in source emission samples is significantly smaller than in PM2.5 from five Chinese cities, supporting our method. In the five cities, secondary source derived via L-HULIS contributes a dominant potion (80% +/- 3.5%) of WS-BrC at 365 nm during the summer, which is approximately twice as high as during the winter (45% +/- 4.9%). Radiocarbon isotope (14C) analysis provides additional constraints to the sources of L-HULIS-derived secondary WS-BrC in urban PM2.5, suggesting that aged biomass burning is the dominant contributor to secondary WS-BrC in winter, and biogenic emission is dominant during summer. This study is the first report on identification of secondary sources of BrC using the fluorescence technique. It demonstrates the potential of this method in characterizing non-fossil source secondary BrC in the atmosphere. Brown carbon (BrC) originates from primary combustion emissions and secondary formation, with large source-dependent uncertainties of radiative forcing. Direct measurements to separate the primary and secondary BrC are challenging due to the chemical complexity. Recent online studies have shown that excitation-emission matrix fluorescence spectroscopy coupled with parallel factor (PARAFAC) analysis identified some fluorescent components that may be linked to secondary sources. However, there is a knowledge gap on whether PARAFAC components correlate closely with atmospheric secondary chemical components, particularly biogenic and anthropogenic secondary organic aerosol, as their precursors can also form secondary BrC chromophores. We established the correlations between PARAFAC components and secondary organic aerosol tracers and compound oxidations to identify the long-emission-wavelength humic-like component as a secondary source tracer of BrC. Then, we estimated non-fossil source secondary BrC in urban aerosols during the winter and summer. Our studies provide references for quantifying secondary sources of BrC in the atmosphere. A fluorescence-based method was developed to investigate secondary sources of water-soluble brown carbon in five cities in China The contribution of secondary sources to water-soluble brown carbon in the summer is approximately twice as high as during the winter This secondary water-soluble brown carbon was more associated with aging biomass burning in winter and biogenic emissions in summer

Recent warming in the Andes is affecting the region's water resources including glaciers and lakes, which supply water to tens of millions of people downstream. High-elevation wetlands, known locally as bofedales, are an understudied Andean ecosystem despite their key role in carbon sequestration, maintenance of biodiversity, and regulation of water flow. Here, we analyze subfossil diatom assemblages and other siliceous bioindicators preserved in a peat core collected from a bofedal in Peru's Cordillera Vilcanota. Basal radiocarbon ages show the bofedal likely formed during a wet period of the Little Ice Age (1520-1680 CE), as inferred from nearby ice core data. The subfossil diatom record is marked by several dynamic assemblage shifts documenting a hydrosere succession from an open-water system to mature peatland. The diatoms appear to be responding largely to changes in hydrology that occur within the natural development of the bofedal, but also to pH and possibly nutrient enrichment from grazing animals. The rapid peat accretion recorded post-1950 at this site is consistent with recent peat growth rates elsewhere in the Andes. Given the many threats to Peruvian bofedales including climate change, overgrazing, peat extraction, and mining, these baseline data will be critical to assessing future change in these important ecosystems.