New indicator toolkit brings ecosystem services down to the crop field

Agriculture fed the twentieth century by stripping it. Intensive fertilisation, pesticides, modern plant breeding and mechanical tillage delivered unprecedented harvests at low prices, but they also drained biodiversity, degraded soils, eutrophied waters and amplified greenhouse gas emissions from farmland. Now, as scientists and policymakers push for a transition toward ecologically intensive cropping systems that lean on natural processes instead of off-farm inputs, a deceptively practical question has become central: how do you actually measure the ecosystem services a crop field provides? A new study published in the journal One Ecosystem offers the most comprehensive answer yet — a collection of 128 field-based indicators designed specifically for empirical, in situ assessments of provisioning and regulating ecosystem services in crop fields.

The research team, working from an exhaustive global review of ecosystem service indicators, set out to bridge a stubborn gap between broad ecological frameworks and the gritty, site-specific knowledge farmers need. Ecosystem services — defined as the contributions ecosystems make to human well-being — include provisioning services like food, fibre and biofuel outputs, regulating services like climate regulation, soil quality regulation, pollination and biological pest control, and cultural services such as recreation and education. Agroecosystems can also generate dis-services: pest damage to crops and soil erosion are negative contributions that arise when services are not adequately provided. While the concept of leveraging these services through agricultural management has gained political momentum, comprehensive lists of indicators applicable directly within crop fields remained scarce.

To build their collection, the researchers started from a previously published list of 507 indicators applicable across various ecosystem types, compiled from 405 peer-reviewed papers. They then applied a series of strict filtering criteria: indicators had to be applicable to crop fields, sit at measurable levels of the ecosystem service cascade model — including ecosystem property, ecosystem function and human benefit, but excluding monetary value — and rely on direct data collection methods such as field observation or remote sensing, excluding all indirect approaches like modelling, expert judgement or literature extraction. Cultural services were excluded because their provision depends on landscape-level factors that can overshadow the effects of local cultivation practices. The filtering process first yielded 59 indicators, from which 20 assemblies of multiple indicators were disaggregated into single components, ultimately producing the final collection of 128 indicators drawn from 66 peer-reviewed articles published between 2006 and 2014.

The cascade model, central to the study’s analytical framework, represents ecosystem service supply as a flow through five levels. Upstream, within the ecological system, biophysical structures called ecosystem properties influence ecosystem functions, which in turn influence final ecosystem services. Downstream, within the socio-economic system, benefits and values derive from service supply. Properties are biophysical structures of the ecosystem, while functions are biological, chemical or physical changes or reactions. Notably, none of the reviewed indicators aligned uniquely with the conceptual centre of the cascade — the final ecosystem service level — so the researchers treated it as a global concept encompassing both the ecosystem function and human benefit levels. Each indicator in the new collection was explicitly tagged with its cascade level, allowing future users to design assessments that integrate both ecological and socio-economic dimensions.

The distribution of the 128 indicators revealed striking disparities. Only 8 per cent related to provisioning services, while 92 per cent related to regulating services — a gap that cannot be explained by the number of services in each section alone, since provisioning services accounted for 44 per cent of the total. In terms of the cascade, the imbalance was even more pronounced: for provisioning services, 80 per cent of indicators measured human benefits, while only 20 per cent measured ecosystem properties. For regulating services, the pattern reversed, with 63 per cent of indicators related to properties and 31 per cent to functions, and only 6 per cent to human benefits or dis-services. Soil quality regulation topped the list with 56 indicators, followed by pollination with 18. Remarkably, no indicator met the filtering criteria for three services — water provision, air quality regulation and life cycle maintenance — all of which are output services without direct income for the farmer.

The analysis of how the original 66 reference papers applied these indicators in practice was equally revealing. Most articles — 64 per cent — assessed a single ecosystem service, 22 per cent assessed two, and only 14 per cent assessed three to six. Just 14 per cent of studies measured both provisioning and regulating services simultaneously. Only four papers used both ecological and socio-economic indicators: one evaluated pollination through function and benefit measures, one assessed soil retention through function and dis-service, and two evaluated water purification. Most studies focused on a single cascade level per service, suggesting that the theoretical ideal of comprehensive, multidimensional assessments integrating both ecological and socio-economic perspectives remains rarely realised in empirical field research.

The researchers explored several explanations for these gaps. Some services, like water provision, function at spatial scales — river basins, watersheds — too large for meaningful field-level measurement, since processes like groundwater recharge depend on water flows and land uses at the basin scale. Others, like air quality regulation, are dominated by ecosystems with dense perennial vegetation such as forests and urban parks, which are considered more relevant hotspots for pollutant removal than crop fields. For services like life cycle maintenance, in situ assessment is so resource-intensive — requiring significant time, expertise and materials to characterise species survival and reproduction — that researchers often favour modelling or mapping instead. The abundance of indicators for soil quality regulation reflects both the complexity of soil as a multidimensional system and the lack of scientific consensus on how to best measure soil quality or health, a concept that remains contested across disciplines.

The study also raises a provocative conceptual question about the relationship between input and output services. Input services — provided by the cultivated field and its surroundings for the benefit of the farmer — support crop production through processes like pollination, soil retention, microclimate regulation and nutrient cycling, and can substitute off-farm inputs. Output services, delivered from agroecosystems for the benefit of society, include food and fibre production, water purification and climate change mitigation. The researchers observed that indicators for each service tended to cluster either on the ecological or on the socio-economic side of the cascade, and they suggest that input services could represent the ecological side of the ecosystem, driving the delivery of output services on the socio-economic side. This raises the opportunity of reclassifying input services as property and function indicators of output services, rather than as independent services — though the precise benefits of input substitution, such as human irrigation being replaced by water regulation, could also be biophysically quantified.

Encouragingly, 91 per cent of the indicators in the collection can be obtained through field observation, and all services except the three absent ones can be quantified in crop fields this way. Remote sensing, currently limited to property indicators for five services, is identified as a major area of growth. The researchers advise future users to explore new remote sensing applications — especially for plant production, water regulation and climate-related services — and to engage with researchers in adjacent disciplines when indicators are lacking, since well-established biophysical measures from fields like air quality research may already exist without being labelled as ecosystem service research. Recent innovations in soil health research, such as soil enzyme production measures and refined reference ranges for soil organic carbon, are also highlighted as examples of indicators emerging from growing research interest.

The collection is available as an interactive dataset, including an Excel file on the Zenodo repository and a Shiny application for exploring the indicators online. The researchers acknowledge that their methodology carries biases — including the specific ecosystem service terminology adopted in the original review and the limitation of references to literature published until 2014 — but they argue that the pronounced disparities in indicator distribution are unlikely to change significantly with more recent studies, since many underlying causes are not time-dependent. By providing a flexible yet robust set of indicators, explicitly tagged with cascade levels and linked to detailed protocols, the collection aims to support researchers and practitioners in selecting context-specific measures that enhance site-specific knowledge for developing sustainable cropping systems — a foundational step toward making ecologically intensive agriculture not just a theoretical ideal, but an empirically grounded practice.

Subject of Research: A collection of field-based indicators for assessing ecosystem services in crop fields

Article Title: A collection of field-based indicators for assessing ecosystem services in crop fields

Article References: Leveau, L., Biot, N., Lobet, G., & Bertin, P. (2025). A collection of field-based indicators for assessing ecosystem services in crop fields. One Ecosystem, 10, Article e151491. https://doi.org/10.3897/oneeco.10.e151491

Image Credits: AI Generated

DOI: 10.3897/oneeco.10.e151491

Keywords: collection, field-based, indicators, assessing, ecosystem, services, crop, fields, scientific research