Michael G. Sunde

Using Planetscope imagery, we trained a random forest model to detect Callery pear (Pyrus calleryana) throughout a diverse urban landscape in Columbia, Missouri. The random forest model had a classification accuracy of 89.78%, a recall score of 0.693, and an F1 score of 0.819. The key hyperparameters for model tuning were the cutoff and class–weight parameters. After the distribution of Callery pear was predicted throughout the landscape, we analyzed the distribution pattern of the predictions using Ripley’s K and then associated the distribution patterns with various socio-economic indicators. The analysis identified significant relationships between the distribution of the predicted Callery pear and population density, median household income, median year the housing infrastructure was built, and median housing value at a variety of spatial scales. The findings from this study provide a much-needed method for detecting species of interest in a heterogenous landscape that is both low cost and does not require specialized hardware or software like some alternative deep learning methods.

Spatial land cover depictions are essential for ecological and environmental management. The thematic resolution of land cover and vegetation maps is also a significant factor affecting the ability to effectively develop policy and land management decisions based on spatial data. Natural resource and conservation planners often seek to develop strategies at broad scales; however, high-quality spatial data depicting current vegetation and ecosystem types over large areas are often unavailable. Since widely available land cover and vegetation datasets are generally lacking in either thematic resolution or spatial coverage, there is a need to integrate modeling approaches and ancillary data with traditional satellite image classifications to produce more detailed ecosystem maps for large areas. In this study, we present a comprehensive approach using satellite imagery, machine learning, and ancillary modeling approaches to develop high-resolution ecological system type maps statewide for Arkansas, USA. A RandomForest land cover classification of Sentinel-2 imagery was generated and further articulated into ecological types using a comprehensive set of secondary modeling approaches. A total of 123 types were mapped in Arkansas, including common cultural and ruderal land cover and vegetation such as pine plantations and developed types. Ozark–Ouachita Dry–Mesic Forest covered the most area, 17.51% of the state. Row Crops covered 17.16%. Twenty-five pine or pine plantation types covered 19.73% of the state, with Ozark–Ouachita pine woodland or mature pine plantation covering 6.15%. Field survey points were used to assess the quality of the mapped ecological systems. The approaches presented here provide a framework for finer resolution mapping of ecological systems at broad scales in other regions.

The abstract should be an objective representation of the article and it must not contain results that are not presented and substantiated in the main text and should not exaggerate the main conclusions.

Analysing dryness/wetness trends, which is typically achieved by examining drought indices, is an essential approach for assessing extreme climate events. However, since there are differences among studies in terms of selected climate variables and timescales used to calculate drought indices, there is uncertainty regarding the magnitude and direction of the dryness/wetness, trend even for the same region. We used 40 years weather station data (1970–2010) and studied the uncertainties of dryness/wetness trends in conterminous China. The effects of selected climate variables, timescales and their interaction on the dryness/wetness trends were quantified by comparing the Standardized Precipitation Index, Standardized Precipitation Evapotranspiration Index (SPEI) with the Thornthwaite equation and SPEI with the Penman–Monteith equation at 3‐, 12‐ and 24‐month timescales. The results indicated that (a) higher uncertainties occurred in relatively arid and humid regions; (b) timescale tended to exaggerate the effect of climate variables, even though its effect was substantially lower than climate variables; (c) relative humidity and precipitation were the primary climate variables affecting observed dryness/wetness trend uncertainties. Since the key climate variables changed depending on the selected timescale, the role of timescale should be further examined the Inner Mongolia and Northeast regions. In general, the study results suggest that the trend analysis of dryness/wetness should include the interaction of both climate variable and timescale selection, particularly when a single drought index is used to estimate the dryness/wetness trend in diverse regions.

Climate change and human activities are important factors driving changes in wetland ecosystems. It is therefore crucial to quantitatively characterize the relative importance of these stressors in wetlands. Previous such analyses have generally not distinguished between wetland types, or have focused on individual wetland types. In this study, three representative wetland areas of the upper, middle and lower reaches of the Heilongjiang River Basin (HRB) were selected as the study area. An object-based classification was used with Landsat TM data to extract the spatial distribution of wetland in 1990, 2000 and 2010. We then quantified the relative importance of climate change and human activities on the wetlands by using the R package “relaimpo” package. The results indicated that: (1) the effects of human activities on wetland changes were greater (contribution rate of 63.57%) than climate change in the HRB. Specifically, there were differences in the relative importance of climate change and human activities for wetlands in different regions. Wetlands of the upper reaches were more affected by climate change, while wetlands in the middle and lower reaches were more affected by human activities; (2) climate change had a greater impact (contribution rate of 65.72%) on low intensity wetland loss, while human activities had a greater impact on moderate and severe intensity wetland loss, with respective contribution rates of 57.22% and 70.35%; (3) climate change had a larger effect on the shrub and forested wetland changes, with respective contribution rates of 58.33% and 52.58%. However, human activities had a larger effect on herbaceous wetland changes, with a contribution rate of 72.28%. Our study provides a useful framework for wetland assessment and management, and could be a useful tool for developing wetland utilization and protection approaches, particularly in sensitive environments in mid- and high-latitude areas.

Vegetation phenology plays a key role in terrestrial ecosystem nutrient and carbon cycles and is sensitive to global climate change. Compared with spring phenology, which has been well studied, autumn phenology is still poorly understood. In this study, we estimated the date of the end of the growing season (EOS) across the Greater Khingan Mountains, China, from 1982 to 2015 based on the Global Inventory Modeling and Mapping Studies (GIMMS) normalized difference vegetation index third-generation (NDVI3g) dataset. The temporal correlations between EOS and climatic factors (e.g., preseason temperature, preseason precipitation), as well as the correlation between autumn and spring phenology, were investigated using partial correlation analysis. Results showed that more than 94% of the pixels in the Greater Khingan Mountains exhibited a delayed EOS trend, with an average rate of 0.23 days/y. Increased preseason temperature resulted in earlier EOS in most of our study area, except for the semi-arid grassland region in the south, where preseason warming generally delayed EOS. Similarly, EOS in most of the mountain deciduous coniferous forest, forest grassland, and mountain grassland forest regions was earlier associated with increased preseason precipitation, but for the semi-arid grassland region, increased precipitation during the preseason mainly led to delayed EOS. However, the effect of preseason precipitation on EOS in most of the Greater Khingan Mountains was stronger than that of preseason temperature. In addition to the climatic effects on EOS, we also found an influence of spring phenology on EOS. An earlier SOS led to a delayed EOS in most of the study area, while in the southern of mountain deciduous coniferous forest region and northern of semi-arid grassland region, an earlier SOS was often followed by an earlier EOS. These findings suggest that both climatic factors and spring phenology should be incorporated into autumn phenology models in order to improve prediction accuracy under present and future climate change scenarios.