Introduction
Ecosystem services are the benefits that people obtain from various ecosystems that can be described and measured (Tamayo et al. 2018; Costanza et al. 1997). Mendelsohn and Olmstead (2009) described the value of ecosystem services (ESV) as “the sum of what all members of society would be willing to pay” for “the economic benefit provided by environmental products or service” (Mendelsohn & Olmstead 2009). Hence, the estimation of ESV can make a vital contribution to biodiversity protection and sustainable development (Billé et al. 2012). Assessments of ESV at the national, regional, basin and even single ecosystem levels can show how these services support our lives and how people develop natural resources rationally (Wei et al. 2018). The valuation methods now available are highly developed and can be mainly divided into behavioral (revealed preference) methods and attitudinal (stated preference) methods (Mendelsohn and Olmstead 2009). Behavioral methods attempt to calculate the environmental value of goods indirectly through market analysis (Braden et al. 2010; Phaneuf et al. 2008; Harrington & Portney 1987). Attitudinal methods use subjectively designed surveys to create a table of ecological value equivalents. Two common valuation systems include the system created by Costanza in 1997 (Costanza et al. 1997) and the millennium ecosystem assessment framework (Alcamo 2003).
However, the understanding of ESV is not comprehensive because multiple types of service interrelate in complex and dynamic ways (Spake et al. 2017). The current research perspectives on ESV consider it to be the result of a process: “human-driven factors of ecosystem change ecosystem process and functions ecosystem services”. “Human-driven factors of ecosystem change” can be interpreted as basic socioeconomic conditions, including population, GDP, industry structure, and energy consumption. “Ecosystem process and functions” can be represented by land and land cover change at the geospatial level, which is traditionally the most important part of the information used to estimate ESV (Barbier et al. 2011). However, how ESV interacts with socioeconomic factors remains ambiguous (Meacham et al. 2016), which leads to difficulties in the application of ESV in ecological management. In other words, even if a low ESV area is identified, we still do not know how to promote it efficiently through regional planning or industry regulation. Studies started to include the socioeconomic drivers of ESV into consideration for the implementation of responsive policies. Yang et al. (2019) found that ESV is tightly correlated with socioeconomic status. Wu et al. (2019) found nonlinear relations between GDP and ESV and between population density and ESV, but no more complete causality was explained.
As one kind of machine learning algorithm, deep learning is a multilayer perceptron neural network (Reichstein et al. 2019). It offers significant breakthroughs in solving classification and nonlinear regression problems (Sze et al. 2017). Deep learning can extract the valid features of data input through complex computational models and represent them at a higher level of abstraction, eventually achieving complex self-learning functions through multiple transformations and combinations (LeCun et al. 2015). Traditional evaluation and analysis methods are often not sufficiently effective in describing the continuous and quantitative rules in a complicated ecosystem (Moore et al. 2017). Deep learning may be an effective tool for dealing with this problem.
In this work, deep learning was used to explore the relationships between “human drivers of ecosystem change” and “ESV” on a dataset from Nanjing, China. The city of Nanjing is one of the megacities in the Yangtze River basin; it has experienced rapid economic development since the 1970s that is still occurring today (Li et al. 2016) (Figure S1). At the end of the 20th century, the urbanization of Nanjing entered an accelerated phase, which led to a rapid increase in population, unreasonable industrial structure, unbalanced land use, high energy consumption, and environmental degradation (Yuan et al. 2018; Shi et al. 2019). Over the last two decades, the population has increased from 3 M to 8.5 M, and its GDP has increased from 338.12 billion CNY in 2008 to 1171.51 billion CNY in 2018. As an ecologically sensitive area, the changes in its ecological system and services have been continuously monitored and studied. Taking ESV as a parameter of the ecosystem, a better understanding of their internal driving mechanism will be conducive to optimizing local policies and regional planning (Shiferaw et al. 2019).
Methods